no force, deployd from gitea

debug: with force
feat: move map from nginx sigil to nginx global config
2025-04-07 11:14:58 -03:00 · 2025-04-07 11:11:01 -03:00 · 2025-01-30 12:40:24 -03:00 · 2024-12-21 23:31:19 +03:00 · 2024-12-21 23:00:30 +03:00 · 2024-08-06 21:01:03 +03:00
154 changed files with 11746 additions and 11704 deletions
--- a/.cursorignore
+++ b/.cursorignore
@@ -1 +0,0 @@
-# Add directories or file patterns to ignore during indexing (e.g. foo/ or *.csv)
--- a/.gitea/workflows/main.yml
+++ b/.gitea/workflows/main.yml
@@ -1,5 +1,9 @@
-name: 'Deploy on push'
-on: [push]
+name:  'Deploy to discoursio-api'
+on: 
+  push:
+    branches:
+      - main
+
 jobs:
  deploy:
    runs-on: ubuntu-latest
@@ -17,19 +21,11 @@ jobs:
        id: branch_name
        run: echo "::set-output name=branch::$(echo ${GITHUB_REF##*/})"

-      - name: Push to dokku for main branch
-        if: github.ref == 'refs/heads/main'
+      - name: Push to dokku
        uses: dokku/github-action@master
        with:
          branch: 'main'
          git_remote_url: 'ssh://dokku@v2.discours.io:22/discoursio-api'
          ssh_private_key: ${{ secrets.SSH_PRIVATE_KEY }}

-      - name: Push to dokku for dev branch
-        if: github.ref == 'refs/heads/dev'
-        uses: dokku/github-action@master
-        with:
-          branch: 'dev'
-          force: true
-          git_remote_url: 'ssh://dokku@v2.discours.io:22/core'
-          ssh_private_key: ${{ secrets.SSH_PRIVATE_KEY }}
+
--- a/.github/workflows/checks.yml
+++ b/.github/workflows/checks.yml
@@ -0,0 +1,16 @@
+name: Checks
+on: [pull_request]
+
+jobs:
+  build:
+    runs-on: ubuntu-latest
+    name: Checks
+    steps:
+    - uses: actions/checkout@v2
+    - uses: actions/setup-python@v2
+      with:
+        python-version: 3.10.6
+    - run: pip install --upgrade pip
+    - run: pip install -r requirements.txt
+    - run: pip install -r requirements-dev.txt
+    - run: ./checks.sh
--- a/.github/workflows/deploy.yml
+++ b/.github/workflows/deploy.yml
@@ -17,11 +17,11 @@ jobs:

      - uses: webfactory/ssh-agent@v0.8.0
        with:
-            ssh-private-key: ${{ github.action.secrets.SSH_PRIVATE_KEY }}
+            ssh-private-key: ${{ secrets.SSH_PRIVATE_KEY }}

      - name: Push to dokku
        env:
-            HOST_KEY: ${{ github.action.secrets.HOST_KEY }}
+            HOST_KEY: ${{ secrets.HOST_KEY }}
        run: |
            echo $HOST_KEY > ~/.ssh/known_hosts
            git remote add dokku dokku@v2.discours.io:discoursio-api
--- a/.gitignore
+++ b/.gitignore
@@ -147,19 +147,11 @@ migration/content/**/*.md
 *.csv
 dev-server.pid
 backups/
-poetry.lock
 .ruff_cache
-.jj
-.zed
-
-dokku_config
-
-*.db
-*.sqlite3
-views.json
-*.pem
-*.key
-*.crt
-*cache.json
-.cursor
+.venv
+poetry.lock
 .devcontainer/devcontainer.json
+localhost-key.pem
+.gitignore
+discoursio.db
+localhost.pem
--- a/.pre-commit-config.yaml
+++ b/.pre-commit-config.yaml
@@ -1,18 +1,44 @@
+exclude: |
+  (?x)(
+    ^tests/unit_tests/resource|
+    _grpc.py|
+    _pb2.py
+  )
+
+default_language_version:
+  python: python3.10
+
 repos:
  - repo: https://github.com/pre-commit/pre-commit-hooks
    rev: v4.5.0
    hooks:
-      - id: check-yaml
+      - id: check-added-large-files
+      - id: check-case-conflict
+      - id: check-docstring-first
+      - id: check-json
+      - id: check-merge-conflict
      - id: check-toml
+      - id: check-yaml
      - id: end-of-file-fixer
      - id: trailing-whitespace
-      - id: check-added-large-files
-      - id: detect-private-key
-      - id: check-ast
-      - id: check-merge-conflict
+      - id: requirements-txt-fixer

-  - repo: https://github.com/astral-sh/ruff-pre-commit
-    rev: v0.4.7
+  - repo: https://github.com/timothycrosley/isort
+    rev: 5.12.0
    hooks:
-      - id: ruff
-        args: [--fix]
+      - id: isort
+
+  - repo: https://github.com/ambv/black
+    rev: 23.10.1
+    hooks:
+      - id: black
+
+  - repo: https://github.com/PyCQA/flake8
+    rev: 6.1.0
+    hooks:
+      - id: flake8
+
+#  - repo: https://github.com/python/mypy
+#    rev: v1.6.1
+#    hooks:
+#      - id: mypy
--- a/CHANGELOG.md
+++ b/CHANGELOG.md
@@ -1,340 +0,0 @@
-#### [0.4.20] - 2025-05-03
- Исправлена ошибка в классе `CacheRevalidationManager`: добавлена инициализация атрибута `_redis`
- Улучшена обработка соединения с Redis в менеджере ревалидации кэша:
-  - Автоматическое восстановление соединения в случае его потери
-  - Проверка соединения перед выполнением операций с кэшем
-  - Дополнительное логирование для упрощения диагностики проблем
- Исправлен резолвер `unpublish_shout`:
-  - Корректное формирование синтетического поля `publication` с `published_at: null`
-  - Возвращение полноценного словаря с данными вместо объекта модели
-  - Улучшена загрузка связанных данных (авторы, темы) для правильного формирования ответа
-
-#### [0.4.19] - 2025-04-14
- dropped `Shout.description` and `Draft.description` to be UX-generated
- use redis to init views counters after migrator
-
-#### [0.4.18] - 2025-04-10
- Fixed `Topic.stat.authors` and `Topic.stat.comments`
- Fixed unique constraint violation for empty slug values:
-  - Modified `update_draft` resolver to handle empty slug values
-  - Modified `create_draft` resolver to prevent empty slug values
-  - Added validation to prevent inserting or updating drafts with empty slug
-  - Fixed database error "duplicate key value violates unique constraint draft_slug_key"
-
-#### [0.4.17] - 2025-03-26
- Fixed `'Reaction' object is not subscriptable` error in hierarchical comments:
-  - Modified `get_reactions_with_stat()` to convert Reaction objects to dictionaries
-  - Added default values for limit/offset parameters
-  - Fixed `load_first_replies()` implementation with proper parameter passing
-  - Added doctest with example usage
-  - Limited child comments to 100 per parent for performance
-
-#### [0.4.16] - 2025-03-22
- Added hierarchical comments pagination:
-  - Created new GraphQL query `load_comments_branch` for efficient loading of hierarchical comments
-  - Ability to load root comments with their first N replies
-  - Added pagination for both root and child comments
-  - Using existing `comments_count` field in `Stat` type to display number of replies
-  - Added special `first_replies` field to store first replies to a comment
-  - Optimized SQL queries for efficient loading of comment hierarchies
-  - Implemented flexible comment sorting system (by time, rating)
-
-#### [0.4.15] - 2025-03-22
- Upgraded caching system described `docs/caching.md`
- Module `cache/memorycache.py` removed
- Enhanced caching system with backward compatibility:
-  - Unified cache key generation with support for existing naming patterns
-  - Improved Redis operation function with better error handling
-  - Updated precache module to use consistent Redis interface
-  - Integrated revalidator with the invalidation system for better performance
-  - Added comprehensive documentation for the caching system
-  - Enhanced cached_query to support template-based cache keys
-  - Standardized error handling across all cache operations
- Optimized cache invalidation system:
-  - Added targeted invalidation for individual entities (authors, topics)
-  - Improved revalidation manager with individual object processing
-  - Implemented batched processing for high-volume invalidations
-  - Reduced Redis operations by using precise key invalidation instead of prefix-based wipes
-  - Added special handling for slug changes in topics
- Unified caching system for all models:
-  - Implemented abstract functions `cache_data`, `get_cached_data` and `invalidate_cache_by_prefix`
-  - Added `cached_query` function for unified approach to query caching
-  - Updated resolvers `author.py` and `topic.py` to use the new caching API
-  - Improved logging for cache operations to simplify debugging
-  - Optimized Redis memory usage through key format unification
- Improved caching and sorting in Topic and Author modules:
-  - Added support for dictionary sorting parameters in `by` for both modules
-  - Optimized cache key generation for stable behavior with various parameters
-  - Enhanced sorting logic with direction support and arbitrary fields
-  - Added `by` parameter support in the API for getting topics by community
- Performance optimizations for author-related queries:
-  - Added SQLAlchemy-managed indexes to `Author`, `AuthorFollower`, `AuthorRating` and `AuthorBookmark` models
-  - Implemented persistent Redis caching for author queries without TTL (invalidated only on changes)
-  - Optimized author retrieval with separate endpoints:
-    - `get_authors_all` - returns all non-deleted authors without statistics
-    - `load_authors_by` - optimized to use caching and efficient sorting and pagination
-  - Improved SQL queries with optimized JOIN conditions and efficient filtering
-  - Added pre-aggregation of statistics (shouts count, followers count) in single efficient queries
-  - Implemented robust cache invalidation on author updates
-  - Created necessary indexes for author lookups by user ID, slug, and timestamps
-
-#### [0.4.14] - 2025-03-21
- Significant performance improvements for topic queries:
-  - Added database indexes to optimize JOIN operations
-  - Implemented persistent Redis caching for topic queries (no TTL, invalidated only on changes)
-  - Optimized topic retrieval with separate endpoints for different use cases:
-    - `get_topics_all` - returns all topics without statistics for lightweight listing
-    - `get_topics_by_community` - adds pagination and optimized filtering by community
-  - Added SQLAlchemy-managed indexes directly in ORM models for automatic schema maintenance
-  - Created `sync_indexes()` function for automatic index synchronization during app startup
-  - Reduced database load by pre-aggregating statistics in optimized SQL queries
-  - Added robust cache invalidation on topic create/update/delete operations
-  - Improved query optimization with proper JOIN conditions and specific partial indexes
-
-#### [0.4.13] - 2025-03-20
- Fixed Topic objects serialization error in cache/memorycache.py
- Improved CustomJSONEncoder to support SQLAlchemy models with dict() method
- Enhanced error handling in cache_on_arguments decorator
- Modified `load_reactions_by` to include deleted reactions when `include_deleted=true` for proper comment tree building
- Fixed featured/unfeatured logic in reaction processing:
-  - Dislike reactions now properly take precedence over likes
-  - Featured status now requires more than 4 likes from users with featured articles
-  - Removed unnecessary filters for deleted reactions since rating reactions are physically deleted
-  - Author's featured status now based on having non-deleted articles with featured_at
-
-#### [0.4.12] - 2025-03-19
- `delete_reaction` detects comments and uses `deleted_at` update
- `check_to_unfeature` etc. update
- dogpile dep in `services/memorycache.py` optimized
-
-#### [0.4.11] - 2025-02-12
- `create_draft` resolver requires draft_id fixed
- `create_draft` resolver defaults body and title fields to empty string
-
-
-#### [0.4.9] - 2025-02-09
- `Shout.draft` field added
- `Draft` entity added
- `create_draft`, `update_draft`, `delete_draft` mutations and resolvers added
- `create_shout`, `update_shout`, `delete_shout` mutations removed from GraphQL API
- `load_drafts` resolver implemented
- `publish_` and `unpublish_` mutations and resolvers added
- `create_`, `update_`, `delete_` mutations and resolvers added for `Draft` entity
- tests with pytest for original auth, shouts, drafts
- `Dockerfile` and `pyproject.toml` removed for the simplicity: `Procfile` and `requirements.txt`
-
-#### [0.4.8] - 2025-02-03
- `Reaction.deleted_at` filter on `update_reaction` resolver added
- `triggers` module updated with `after_shout_handler`, `after_reaction_handler` for cache revalidation
- `after_shout_handler`, `after_reaction_handler` now also handle `deleted_at` field
- `get_cached_topic_followers` fixed
- `get_my_rates_comments` fixed
-
-#### [0.4.7]
- `get_my_rates_shouts` resolver added with:
-  - `shout_id` and `my_rate` fields in response
-  - filters by `Reaction.deleted_at.is_(None)`
-  - filters by `Reaction.kind.in_([ReactionKind.LIKE.value, ReactionKind.DISLIKE.value])`
-  - filters by `Reaction.reply_to.is_(None)`
-  - uses `local_session()` context manager
-  - returns empty list on errors
- SQLAlchemy syntax updated:
-  - `select()` statement fixed for newer versions
-  - `Reaction` model direct selection instead of labeled columns
-  - proper row access with `row[0].shout` and `row[0].kind`
- GraphQL resolver fixes:
-  - added root parameter `_` to match schema
-  - proper async/await handling with `@login_required`
-  - error logging added via `logger.error()`
-
-#### [0.4.6]
- login_accepted decorator added
- `docs` added
- optimized and unified `load_shouts_*` resolvers with `LoadShoutsOptions`
- `load_shouts_bookmarked` resolver fixed
- resolvers updates:
-    - new resolvers group `feed`
-    - `load_shouts_authored_by` resolver added
-    - `load_shouts_with_topic` resolver added
-    - `load_shouts_followed` removed
-    - `load_shouts_random_topic` removed
-    - `get_topics_random` removed
- model updates:
-    - `ShoutsOrderBy` enum added
-    - `Shout.main_topic` from `ShoutTopic.main` as `Topic` type output
-    - `Shout.created_by` as `Author` type output
-
-#### [0.4.5]
- `bookmark_shout` mutation resolver added
- `load_shouts_bookmarked` resolver added
- `get_communities_by_author` resolver added
- `get_communities_all` resolver fixed
- `Community` stats in orm
- `Community` CUDL resolvers added
- `Reaction` filter by `Reaction.kind`s
- `ReactionSort` enum added
- `CommunityFollowerRole` enum added
- `InviteStatus` enum added
- `Topic.parents` ids added
- `get_shout` resolver accepts slug or shout_id
-
-#### [0.4.4]
- `followers_stat` removed for shout
- sqlite3 support added
- `rating_stat` and `comments_count` fixes
-
-#### [0.4.3]
- cache reimplemented
- load shouts queries unified
- `followers_stat` removed from shout
-
-#### [0.4.2]
- reactions load resolvers separated for ratings (no stats) and comments
- reactions stats improved
- `load_comment_ratings` separate resolver
-
-#### [0.4.1]
- follow/unfollow logic updated and unified with cache
-
-#### [0.4.0]
- chore: version migrator synced
- feat: precache_data on start
- fix: store id list for following cache data
- fix: shouts stat filter out deleted
-
-#### [0.3.5]
- cache isolated to services
- topics followers and authors cached
- redis stores lists of ids
-
-#### [0.3.4]
- `load_authors_by` from cache
-
-#### [0.3.3]
- feat: sentry integration enabled with glitchtip
- fix: reindex on update shout
- packages upgrade, isort
- separated stats queries for author and topic
- fix: feed featured filter
- fts search removed
-
-#### [0.3.2]
- redis cache for what author follows
- redis cache for followers
- graphql add query: get topic followers
-
-#### [0.3.1]
- enabling sentry
- long query log report added
- editor fixes
- authors links cannot be updated by `update_shout` anymore
-
-#### [0.3.0]
- `Shout.featured_at` timestamp of the frontpage featuring event
- added proposal accepting logics
- schema modulized
- Shout.visibility removed
-
-#### [0.2.22]
- added precommit hook
- fmt
- granian asgi
-
-#### [0.2.21]
- fix: rating logix
- fix: `load_top_random_shouts`
- resolvers: `add_stat_*` refactored
- services: use google analytics
- services: minor fixes search
-
-#### [0.2.20]
- services: ackee removed
- services: following manager fixed
- services: import views.json
-
-#### [0.2.19]
- fix: adding `author` role
- fix: stripping `user_id` in auth connector
-
-#### [0.2.18]
- schema: added `Shout.seo` string field
- resolvers: added `/new-author` webhook resolver
- resolvers: added reader.load_shouts_top_random
- resolvers: added reader.load_shouts_unrated
- resolvers: community follower id property name is `.author`
- resolvers: `get_authors_all` and `load_authors_by`
- services: auth connector upgraded
-
-#### [0.2.17]
- schema: enum types workaround, `ReactionKind`, `InviteStatus`, `ShoutVisibility`
- schema: `Shout.created_by`, `Shout.updated_by`
- schema: `Shout.authors` can be empty
- resolvers: optimized `reacted_shouts_updates` query
-
-#### [0.2.16]
- resolvers: collab inviting logics
- resolvers: queries and mutations revision and renaming
- resolvers: `delete_topic(slug)` implemented
- resolvers: added `get_shout_followers`
- resolvers: `load_shouts_by` filters implemented
- orm: invite entity
- schema: `Reaction.range` -> `Reaction.quote`
- filters: `time_ago` -> `after`
- httpx -> aiohttp
-
-#### [0.2.15]
- schema: `Shout.created_by` removed
- schema: `Shout.mainTopic` removed
- services: cached elasticsearch connector
- services: auth is using `user_id` from authorizer
- resolvers: `notify_*` usage fixes
- resolvers: `getAuthor` now accepts slug, `user_id` or `author_id`
- resolvers: login_required usage fixes
-
-#### [0.2.14]
- schema: some fixes from migrator
- schema: `.days` -> `.time_ago`
- schema: `excludeLayout` + `layout` in filters -> `layouts`
- services: db access simpler, no contextmanager
- services: removed Base.create() method
- services: rediscache updated
- resolvers: get_reacted_shouts_updates as followedReactions query
-
-#### [0.2.13]
- services: db context manager
- services: `ViewedStorage` fixes
- services: views are not stored in core db anymore
- schema: snake case in model fields names
- schema: no DateTime scalar
- resolvers: `get_my_feed` comments filter reactions body.is_not('')
- resolvers: `get_my_feed` query fix
- resolvers: `LoadReactionsBy.days` -> `LoadReactionsBy.time_ago`
- resolvers: `LoadShoutsBy.days` -> `LoadShoutsBy.time_ago`
-
-#### [0.2.12]
- `Author.userpic` -> `Author.pic`
- `CommunityFollower.role` is string now
- `Author.user` is string now
-
-#### [0.2.11]
- redis interface updated
- `viewed` interface updated
- `presence` interface updated
- notify on create, update, delete for reaction and shout
- notify on follow / unfollow author
- use pyproject
- devmode fixed
-
-#### [0.2.10]
- community resolvers connected
-
-#### [0.2.9]
- starlette is back, aiohttp removed
- aioredis replaced with aredis
-
-#### [0.2.8]
- refactored
-
-
-#### [0.2.7]
- `loadFollowedReactions` now with `
--- a/5
+++ b/5
@@ -0,0 +1,5 @@
+WAIT=10
+TIMEOUT=10
+ATTEMPTS=3
+
+/
--- a/17
+++ b/17
@@ -1,18 +1,11 @@
-FROM python:slim
-
-RUN apt-get update && apt-get install -y \
-    postgresql-client \
-    curl \
-    && rm -rf /var/lib/apt/lists/*
-
+FROM python:3.11-slim
 WORKDIR /app

+EXPOSE 8080
+ADD nginx.conf.sigil ./
 COPY requirements.txt .
-
+RUN apt update && apt install -y git gcc curl postgresql
 RUN pip install -r requirements.txt
-
 COPY . .

-EXPOSE 8000
-
-CMD ["python", "-m", "granian", "main:app", "--interface", "asgi", "--host", "0.0.0.0", "--port", "8000"]
+CMD python server.py
--- a/README.md
+++ b/README.md
@@ -1,102 +1,47 @@
-# GraphQL API Backend
+# discoursio-api

-Backend service providing GraphQL API for content management system with reactions, ratings and comments.

-## Core Features
+- sqlalchemy
+- redis
+- ariadne
+- starlette
+- uvicorn

-### Shouts (Posts)
- CRUD operations via GraphQL mutations
- Rich filtering and sorting options
- Support for multiple authors and topics
- Rating system with likes/dislikes
- Comments and nested replies
- Bookmarks and following
-
-### Reactions System
- `ReactionKind` types: LIKE, DISLIKE, COMMENT
- Rating calculation for shouts and comments
- User-specific reaction tracking
- Reaction stats and aggregations
- Nested comments support
-
-### Authors & Topics
- Author profiles with stats
- Topic categorization and hierarchy
- Following system for authors/topics
- Activity tracking and stats
- Community features
-
-## Tech Stack
-
- **(Python)[https://www.python.org/]** 3.12+
- **GraphQL** with [Ariadne](https://ariadnegraphql.org/)
- **(SQLAlchemy)[https://docs.sqlalchemy.org/en/20/orm/]**
- **(PostgreSQL)[https://www.postgresql.org/]/(SQLite)[https://www.sqlite.org/]** support
- **(Starlette)[https://www.starlette.io/]** for ASGI server
- **(Redis)[https://redis.io/]** for caching
-
-## Development
-
-### Prepare environment:
-
-```shell
-mkdir .venv
-python3.12 -m venv venv
-source venv/bin/activate
+on osx
+```
+brew install redis nginx postgres
+brew services start redis
 ```

-### Run server
-
-First, certifcates are required to run the server.
-
-```shell
-mkcert -install
-mkcert localhost
+on debian/ubuntu
+```
+apt install redis nginx
 ```

-Then, run the server:
+# Local development

-```shell
-python -m granian main:app --interface asgi --host 0.0.0.0 --port 8000
+Install deps first
+
+```
+pip install -r requirements.txt
+pip install -r requirements-dev.txt
+pre-commit install
 ```

-### Useful Commands
-
-```shell
-# Linting and import sorting
-ruff check . --fix --select I 
-
-# Code formatting
-ruff format . --line-length=120 
-
-# Run tests
-pytest
-
-# Type checking
-mypy .
+Create database from backup
+```
+./restdb.sh
 ```

-### Code Style
-
-We use:
- Ruff for linting and import sorting
- Line length: 120 characters
- Python type hints
- Docstrings for public methods
-
-### GraphQL Development
-
-Test queries in GraphQL Playground at `http://localhost:8000`:
-
-```graphql
-# Example query
-query GetShout($slug: String) {
-  get_shout(slug: $slug) {
-    id
-    title
-    main_author {
-      name
-    }
-  }
-}
+Start local server
 ```
+python3 server.py dev
+```
+
+# How to do an authorized request
+
+Put the header 'Authorization' with token from signIn query or registerUser mutation.
+
+# How to debug Ackee
+
+Set ACKEE_TOKEN var
--- a/init.py
+++ b/init.py
@@ -1,6 +0,0 @@
-import os
-import sys
-
-# Получаем путь к корневой директории проекта
-root_path = os.path.abspath(os.path.dirname(__file__))
-sys.path.append(root_path)
--- a/alembic.ini
+++ b/alembic.ini
@@ -0,0 +1,110 @@
+# A generic, single database configuration.
+
+[alembic]
+# path to migration scripts
+script_location = alembic
+
+# template used to generate migration file names; The default value is %%(rev)s_%%(slug)s
+# Uncomment the line below if you want the files to be prepended with date and time
+# see https://alembic.sqlalchemy.org/en/latest/tutorial.html#editing-the-ini-file
+# for all available tokens
+# file_template = %%(year)d_%%(month).2d_%%(day).2d_%%(hour).2d%%(minute).2d-%%(rev)s_%%(slug)s
+
+# sys.path path, will be prepended to sys.path if present.
+# defaults to the current working directory.
+prepend_sys_path = .
+
+# timezone to use when rendering the date within the migration file
+# as well as the filename.
+# If specified, requires the python-dateutil library that can be
+# installed by adding `alembic[tz]` to the pip requirements
+# string value is passed to dateutil.tz.gettz()
+# leave blank for localtime
+# timezone =
+
+# max length of characters to apply to the
+# "slug" field
+# truncate_slug_length = 40
+
+# set to 'true' to run the environment during
+# the 'revision' command, regardless of autogenerate
+# revision_environment = false
+
+# set to 'true' to allow .pyc and .pyo files without
+# a source .py file to be detected as revisions in the
+# versions/ directory
+# sourceless = false
+
+# version location specification; This defaults
+# to alembic/versions.  When using multiple version
+# directories, initial revisions must be specified with --version-path.
+# The path separator used here should be the separator specified by "version_path_separator" below.
+# version_locations = %(here)s/bar:%(here)s/bat:alembic/versions
+
+# version path separator; As mentioned above, this is the character used to split
+# version_locations. The default within new alembic.ini files is "os", which uses os.pathsep.
+# If this key is omitted entirely, it falls back to the legacy behavior of splitting on spaces and/or commas.
+# Valid values for version_path_separator are:
+#
+# version_path_separator = :
+# version_path_separator = ;
+# version_path_separator = space
+version_path_separator = os  # Use os.pathsep. Default configuration used for new projects.
+
+# set to 'true' to search source files recursively
+# in each "version_locations" directory
+# new in Alembic version 1.10
+# recursive_version_locations = false
+
+# the output encoding used when revision files
+# are written from script.py.mako
+# output_encoding = utf-8
+
+sqlalchemy.url = %(DB_URL)
+
+
+[post_write_hooks]
+# post_write_hooks defines scripts or Python functions that are run
+# on newly generated revision scripts.  See the documentation for further
+# detail and examples
+
+# format using "black" - use the console_scripts runner, against the "black" entrypoint
+# hooks = black
+# black.type = console_scripts
+# black.entrypoint = black
+# black.options = -l 79 REVISION_SCRIPT_FILENAME
+
+# Logging configuration
+[loggers]
+keys = root,sqlalchemy,alembic
+
+[handlers]
+keys = console
+
+[formatters]
+keys = generic
+
+[logger_root]
+level = WARN
+handlers = console
+qualname =
+
+[logger_sqlalchemy]
+level = WARN
+handlers =
+qualname = sqlalchemy.engine
+
+[logger_alembic]
+level = INFO
+handlers =
+qualname = alembic
+
+[handler_console]
+class = StreamHandler
+args = (sys.stderr,)
+level = NOTSET
+formatter = generic
+
+[formatter_generic]
+format = %(levelname)-5.5s [%(name)s] %(message)s
+datefmt = %H:%M:%S
--- a/alembic/README
+++ b/alembic/README
@@ -0,0 +1,3 @@
+Generic single-database configuration.
+
+https://alembic.sqlalchemy.org/en/latest/tutorial.html
--- a/alembic/env.py
+++ b/alembic/env.py
@@ -3,7 +3,7 @@ from logging.config import fileConfig
 from sqlalchemy import engine_from_config, pool

 from alembic import context
-from services.db import Base
+from base.orm import Base
 from settings import DB_URL

 # this is the Alembic Config object, which provides
--- a/alembic/script.py.mako
+++ b/alembic/script.py.mako
@@ -0,0 +1,26 @@
+"""${message}
+
+Revision ID: ${up_revision}
+Revises: ${down_revision | comma,n}
+Create Date: ${create_date}
+
+"""
+from typing import Sequence, Union
+
+from alembic import op
+import sqlalchemy as sa
+${imports if imports else ""}
+
+# revision identifiers, used by Alembic.
+revision: str = ${repr(up_revision)}
+down_revision: Union[str, None] = ${repr(down_revision)}
+branch_labels: Union[str, Sequence[str], None] = ${repr(branch_labels)}
+depends_on: Union[str, Sequence[str], None] = ${repr(depends_on)}
+
+
+def upgrade() -> None:
+    ${upgrades if upgrades else "pass"}
+
+
+def downgrade() -> None:
+    ${downgrades if downgrades else "pass"}
--- a/alembic/versions/fe943b098418_init_alembic.py
+++ b/alembic/versions/fe943b098418_init_alembic.py
@@ -0,0 +1,26 @@
+"""init alembic
+
+Revision ID: fe943b098418
+Revises:
+Create Date: 2023-08-19 01:37:57.031933
+
+"""
+from typing import Sequence, Union
+
+# import sqlalchemy as sa
+
+# from alembic import op
+
+# revision identifiers, used by Alembic.
+revision: str = "fe943b098418"
+down_revision: Union[str, None] = None
+branch_labels: Union[str, Sequence[str], None] = None
+depends_on: Union[str, Sequence[str], None] = None
+
+
+def upgrade() -> None:
+    pass
+
+
+def downgrade() -> None:
+    pass
--- a/app.json
+++ b/app.json
@@ -1,15 +0,0 @@
-{
-  "healthchecks": {
-    "web": [
-      {
-        "type": "startup",
-        "name": "web check",
-        "description": "Checking if the app responds to the GET /",
-        "path": "/",
-        "attempts": 3,
-        "warn": true,
-        "initialDelay": 1
-      }
-    ]
-  }
-}
--- a/app/resolvers/draft.py
+++ b/app/resolvers/draft.py
@@ -1 +0,0 @@
- 
--- a/auth/authenticate.py
+++ b/auth/authenticate.py
@@ -7,22 +7,26 @@ from starlette.authentication import AuthenticationBackend
 from starlette.requests import HTTPConnection

 from auth.credentials import AuthCredentials, AuthUser
-from auth.exceptions import OperationNotAllowed
 from auth.tokenstorage import SessionToken
-from auth.usermodel import Role, User
-from services.db import local_session
+from base.exceptions import OperationNotAllowed
+from base.orm import local_session
+from orm.user import Role, User
 from settings import SESSION_TOKEN_HEADER


 class JWTAuthenticate(AuthenticationBackend):
-    async def authenticate(self, request: HTTPConnection) -> Optional[Tuple[AuthCredentials, AuthUser]]:
+    async def authenticate(
+        self, request: HTTPConnection
+    ) -> Optional[Tuple[AuthCredentials, AuthUser]]:
        if SESSION_TOKEN_HEADER not in request.headers:
            return AuthCredentials(scopes={}), AuthUser(user_id=None, username="")

        token = request.headers.get(SESSION_TOKEN_HEADER)
        if not token:
            print("[auth.authenticate] no token in header %s" % SESSION_TOKEN_HEADER)
-            return AuthCredentials(scopes={}, error_message=str("no token")), AuthUser(user_id=None, username="")
+            return AuthCredentials(scopes={}, error_message=str("no token")), AuthUser(
+                user_id=None, username=""
+            )

        if len(token.split(".")) > 1:
            payload = await SessionToken.verify(token)
@@ -48,14 +52,20 @@ class JWTAuthenticate(AuthenticationBackend):
                except exc.NoResultFound:
                    pass

-        return AuthCredentials(scopes={}, error_message=str("Invalid token")), AuthUser(user_id=None, username="")
+        return AuthCredentials(scopes={}, error_message=str("Invalid token")), AuthUser(
+            user_id=None, username=""
+        )


 def login_required(func):
    @wraps(func)
    async def wrap(parent, info: GraphQLResolveInfo, *args, **kwargs):
+        # debug only
+        # print('[auth.authenticate] login required for %r with info %r' % (func, info))
        auth: AuthCredentials = info.context["request"].auth
+        # print(auth)
        if not auth or not auth.logged_in:
+            # raise Unauthorized(auth.error_message or "Please login")
            return {"error": "Please login first"}
        return await func(parent, info, *args, **kwargs)

@@ -65,7 +75,9 @@ def login_required(func):
 def permission_required(resource, operation, func):
    @wraps(func)
    async def wrap(parent, info: GraphQLResolveInfo, *args, **kwargs):
-        print("[auth.authenticate] permission_required for %r with info %r" % (func, info))  # debug only
+        print(
+            "[auth.authenticate] permission_required for %r with info %r" % (func, info)
+        )  # debug only
        auth: AuthCredentials = info.context["request"].auth
        if not auth.logged_in:
            raise OperationNotAllowed(auth.error_message or "Please login")
@@ -75,22 +87,3 @@ def permission_required(resource, operation, func):
        return await func(parent, info, *args, **kwargs)

    return wrap
-
-
-def login_accepted(func):
-    @wraps(func)
-    async def wrap(parent, info: GraphQLResolveInfo, *args, **kwargs):
-        auth: AuthCredentials = info.context["request"].auth
-
-        # Если есть авторизация, добавляем данные автора в контекст
-        if auth and auth.logged_in:
-            info.context["author"] = auth.author
-            info.context["user_id"] = auth.author.get("id")
-        else:
-            # Очищаем данные автора из контекста если авторизация отсутствует
-            info.context["author"] = None
-            info.context["user_id"] = None
-
-        return await func(parent, info, *args, **kwargs)
-
-    return wrap
--- a/auth/identity.py
+++ b/auth/identity.py
@@ -1,15 +1,15 @@
 from binascii import hexlify
 from hashlib import sha256

+from jwt import DecodeError, ExpiredSignatureError
 from passlib.hash import bcrypt

-from auth.exceptions import ExpiredToken, InvalidToken
 from auth.jwtcodec import JWTCodec
 from auth.tokenstorage import TokenStorage
-from orm.user import User

 # from base.exceptions import InvalidPassword, InvalidToken
-from services.db import local_session
+from base.orm import local_session
+from orm import User


 class Password:
@@ -33,8 +33,8 @@ class Password:
        Verify that password hash is equal to specified hash. Hash format:

        $2a$10$Ro0CUfOqk6cXEKf3dyaM7OhSCvnwM9s4wIX9JeLapehKK5YdLxKcm
-        \__/\/ \____________________/\_____________________________/  # noqa: W605
-        |   |        Salt                     Hash
+         __ __ ____________________________________________________  # noqa: W605
+        |   |  |     Salt  (22)    |          Hash
        |  Cost
        Version

@@ -80,10 +80,10 @@ class Identity:
            if not await TokenStorage.exist(f"{payload.user_id}-{payload.username}-{token}"):
                # raise InvalidToken("Login token has expired, please login again")
                return {"error": "Token has expired"}
-        except ExpiredToken:
+        except ExpiredSignatureError:
            # raise InvalidToken("Login token has expired, please try again")
            return {"error": "Token has expired"}
-        except InvalidToken:
+        except DecodeError:
            # raise InvalidToken("token format error") from e
            return {"error": "Token format error"}
        with local_session() as session:
--- a/auth/jwtcodec.py
+++ b/auth/jwtcodec.py
@@ -1,23 +1,15 @@
 from datetime import datetime, timezone

 import jwt
-from pydantic import BaseModel

-from auth.exceptions import ExpiredToken, InvalidToken
+from base.exceptions import ExpiredToken, InvalidToken
 from settings import JWT_ALGORITHM, JWT_SECRET_KEY
-
-
-class TokenPayload(BaseModel):
-    user_id: str
-    username: str
-    exp: datetime
-    iat: datetime
-    iss: str
+from validations.auth import AuthInput, TokenPayload


 class JWTCodec:
    @staticmethod
-    def encode(user, exp: datetime) -> str:
+    def encode(user: AuthInput, exp: datetime) -> str:
        payload = {
            "user_id": user.id,
            "username": user.email or user.phone,
@@ -31,7 +23,7 @@ class JWTCodec:
            print("[auth.jwtcodec] JWT encode error %r" % e)

    @staticmethod
-    def decode(token: str, verify_exp: bool = True):
+    def decode(token: str, verify_exp: bool = True) -> TokenPayload:
        r = None
        payload = None
        try:
--- a/auth/tokenstorage.py
+++ b/auth/tokenstorage.py
@@ -1,9 +1,9 @@
 from datetime import datetime, timedelta, timezone

 from auth.jwtcodec import JWTCodec
-from auth.validations import AuthInput
-from services.redis import redis
+from base.redis import redis
 from settings import ONETIME_TOKEN_LIFE_SPAN, SESSION_TOKEN_LIFE_SPAN
+from validations.auth import AuthInput


 async def save(token_key, life_span, auto_delete=True):
--- a/auth/usermodel.py
+++ b/auth/usermodel.py
@@ -1,119 +0,0 @@
-import time
-
-from sqlalchemy import (
-    JSON,
-    Boolean,
-    Column,
-    DateTime,
-    ForeignKey,
-    Integer,
-    String,
-    func,
-)
-from sqlalchemy.orm import relationship
-
-from services.db import Base
-
-
-class Permission(Base):
-    __tablename__ = "permission"
-
-    id = Column(String, primary_key=True, unique=True, nullable=False, default=None)
-    resource = Column(String, nullable=False)
-    operation = Column(String, nullable=False)
-
-
-class Role(Base):
-    __tablename__ = "role"
-
-    id = Column(String, primary_key=True, unique=True, nullable=False, default=None)
-    name = Column(String, nullable=False)
-    permissions = relationship(Permission)
-
-
-class AuthorizerUser(Base):
-    __tablename__ = "authorizer_users"
-
-    id = Column(String, primary_key=True, unique=True, nullable=False, default=None)
-    key = Column(String)
-    email = Column(String, unique=True)
-    email_verified_at = Column(Integer)
-    family_name = Column(String)
-    gender = Column(String)
-    given_name = Column(String)
-    is_multi_factor_auth_enabled = Column(Boolean)
-    middle_name = Column(String)
-    nickname = Column(String)
-    password = Column(String)
-    phone_number = Column(String, unique=True)
-    phone_number_verified_at = Column(Integer)
-    # preferred_username = Column(String, nullable=False)
-    picture = Column(String)
-    revoked_timestamp = Column(Integer)
-    roles = Column(String, default="author,reader")
-    signup_methods = Column(String, default="magic_link_login")
-    created_at = Column(Integer, default=lambda: int(time.time()))
-    updated_at = Column(Integer, default=lambda: int(time.time()))
-
-
-class UserRating(Base):
-    __tablename__ = "user_rating"
-
-    id = None
-    rater: Column = Column(ForeignKey("user.id"), primary_key=True, index=True)
-    user: Column = Column(ForeignKey("user.id"), primary_key=True, index=True)
-    value: Column = Column(Integer)
-
-    @staticmethod
-    def init_table():
-        pass
-
-
-class UserRole(Base):
-    __tablename__ = "user_role"
-
-    id = None
-    user = Column(ForeignKey("user.id"), primary_key=True, index=True)
-    role = Column(ForeignKey("role.id"), primary_key=True, index=True)
-
-
-class User(Base):
-    __tablename__ = "user"
-    default_user = None
-
-    email = Column(String, unique=True, nullable=False, comment="Email")
-    username = Column(String, nullable=False, comment="Login")
-    password = Column(String, nullable=True, comment="Password")
-    bio = Column(String, nullable=True, comment="Bio")  # status description
-    about = Column(String, nullable=True, comment="About")  # long and formatted
-    userpic = Column(String, nullable=True, comment="Userpic")
-    name = Column(String, nullable=True, comment="Display name")
-    slug = Column(String, unique=True, comment="User's slug")
-    links = Column(JSON, nullable=True, comment="Links")
-    oauth = Column(String, nullable=True)
-    oid = Column(String, nullable=True)
-
-    muted = Column(Boolean, default=False)
-    confirmed = Column(Boolean, default=False)
-
-    created_at = Column(DateTime(timezone=True), nullable=False, server_default=func.now(), comment="Created at")
-    updated_at = Column(DateTime(timezone=True), nullable=False, server_default=func.now(), comment="Updated at")
-    last_seen = Column(DateTime(timezone=True), nullable=False, server_default=func.now(), comment="Was online at")
-    deleted_at = Column(DateTime(timezone=True), nullable=True, comment="Deleted at")
-
-    ratings = relationship(UserRating, foreign_keys=UserRating.user)
-    roles = relationship(lambda: Role, secondary=UserRole.__tablename__)
-
-    def get_permission(self):
-        scope = {}
-        for role in self.roles:
-            for p in role.permissions:
-                if p.resource not in scope:
-                    scope[p.resource] = set()
-                scope[p.resource].add(p.operation)
-        print(scope)
-        return scope
-
-
-# if __name__ == "__main__":
-#   print(User.get_permission(user_id=1))
--- a/auth/validations.py
+++ b/auth/validations.py
@@ -1,116 +0,0 @@
-import re
-from datetime import datetime
-from typing import Dict, List, Optional, Union
-
-from pydantic import BaseModel, Field, field_validator
-
-# RFC 5322 compliant email regex pattern
-EMAIL_PATTERN = r"^[a-zA-Z0-9._%+-]+@[a-zA-Z0-9.-]+\.[a-zA-Z]{2,}$"
-
-
-class AuthInput(BaseModel):
-    """Base model for authentication input validation"""
-
-    user_id: str = Field(description="Unique user identifier")
-    username: str = Field(min_length=2, max_length=50)
-    token: str = Field(min_length=32)
-
-    @field_validator("user_id")
-    @classmethod
-    def validate_user_id(cls, v: str) -> str:
-        if not v.strip():
-            raise ValueError("user_id cannot be empty")
-        return v
-
-
-class UserRegistrationInput(BaseModel):
-    """Validation model for user registration"""
-
-    email: str = Field(max_length=254)  # Max email length per RFC 5321
-    password: str = Field(min_length=8, max_length=100)
-    name: str = Field(min_length=2, max_length=50)
-
-    @field_validator("email")
-    @classmethod
-    def validate_email(cls, v: str) -> str:
-        """Validate email format"""
-        if not re.match(EMAIL_PATTERN, v):
-            raise ValueError("Invalid email format")
-        return v.lower()
-
-    @field_validator("password")
-    @classmethod
-    def validate_password_strength(cls, v: str) -> str:
-        """Validate password meets security requirements"""
-        if not any(c.isupper() for c in v):
-            raise ValueError("Password must contain at least one uppercase letter")
-        if not any(c.islower() for c in v):
-            raise ValueError("Password must contain at least one lowercase letter")
-        if not any(c.isdigit() for c in v):
-            raise ValueError("Password must contain at least one number")
-        if not any(c in "!@#$%^&*()_+-=[]{}|;:,.<>?" for c in v):
-            raise ValueError("Password must contain at least one special character")
-        return v
-
-
-class UserLoginInput(BaseModel):
-    """Validation model for user login"""
-
-    email: str = Field(max_length=254)
-    password: str = Field(min_length=8, max_length=100)
-
-    @field_validator("email")
-    @classmethod
-    def validate_email(cls, v: str) -> str:
-        if not re.match(EMAIL_PATTERN, v):
-            raise ValueError("Invalid email format")
-        return v.lower()
-
-
-class TokenPayload(BaseModel):
-    """Validation model for JWT token payload"""
-
-    user_id: str
-    username: str
-    exp: datetime
-    iat: datetime
-    scopes: Optional[List[str]] = []
-
-
-class OAuthInput(BaseModel):
-    """Validation model for OAuth input"""
-
-    provider: str = Field(pattern="^(google|github|facebook)$")
-    code: str
-    redirect_uri: Optional[str] = None
-
-    @field_validator("provider")
-    @classmethod
-    def validate_provider(cls, v: str) -> str:
-        valid_providers = ["google", "github", "facebook"]
-        if v.lower() not in valid_providers:
-            raise ValueError(f"Provider must be one of: {', '.join(valid_providers)}")
-        return v.lower()
-
-
-class AuthResponse(BaseModel):
-    """Validation model for authentication responses"""
-
-    success: bool
-    token: Optional[str] = None
-    error: Optional[str] = None
-    user: Optional[Dict[str, Union[str, int, bool]]] = None
-
-    @field_validator("error")
-    @classmethod
-    def validate_error_if_not_success(cls, v: Optional[str], info) -> Optional[str]:
-        if not info.data.get("success") and not v:
-            raise ValueError("Error message required when success is False")
-        return v
-
-    @field_validator("token")
-    @classmethod
-    def validate_token_if_success(cls, v: Optional[str], info) -> Optional[str]:
-        if info.data.get("success") and not v:
-            raise ValueError("Token required when success is True")
-        return v
--- a/base/exceptions.py
+++ b/base/exceptions.py
--- a/base/orm.py
+++ b/base/orm.py
@@ -0,0 +1,57 @@
+from typing import Any, Callable, Dict, Generic, TypeVar
+
+from sqlalchemy import Column, Integer, create_engine
+from sqlalchemy.ext.declarative import declarative_base
+from sqlalchemy.orm import Session
+from sqlalchemy.sql.schema import Table
+
+from settings import DB_URL
+
+engine = create_engine(DB_URL, echo=False, pool_size=10, max_overflow=20)
+
+T = TypeVar("T")
+
+REGISTRY: Dict[str, type] = {}
+
+
+def local_session():
+    return Session(bind=engine, expire_on_commit=False)
+
+
+DeclarativeBase = declarative_base()  # type: Any
+
+
+class Base(DeclarativeBase):
+    __table__: Table
+    __tablename__: str
+    __new__: Callable
+    __init__: Callable
+    __allow_unmapped__ = True
+    __abstract__ = True
+    __table_args__ = {"extend_existing": True}
+
+    id = Column(Integer, primary_key=True)
+
+    def __init_subclass__(cls, **kwargs):
+        REGISTRY[cls.__name__] = cls
+
+    @classmethod
+    def create(cls: Generic[T], **kwargs) -> Generic[T]:
+        instance = cls(**kwargs)
+        return instance.save()
+
+    def save(self) -> Generic[T]:
+        with local_session() as session:
+            session.add(self)
+            session.commit()
+        return self
+
+    def update(self, input):
+        column_names = self.__table__.columns.keys()
+        for name, value in input.items():
+            if name in column_names:
+                setattr(self, name, value)
+
+    def dict(self) -> Dict[str, Any]:
+        column_names = self.__table__.columns.keys()
+        return {c: getattr(self, c) for c in column_names}
--- a/services/redis.py
+++ b/services/redis.py
@@ -1,41 +1,32 @@
-import logging
-
-from redis.asyncio import Redis
+import redis.asyncio as aredis

 from settings import REDIS_URL
+import logging

-# Set redis logging level to suppress DEBUG messages
-logger = logging.getLogger("redis")
-logger.setLevel(logging.WARNING)
+logger = logging.getLogger("[services.redis] ")
+logger.setLevel(logging.DEBUG)


-class RedisService:
+class RedisCache:
    def __init__(self, uri=REDIS_URL):
        self._uri: str = uri
        self.pubsub_channels = []
        self._client = None

    async def connect(self):
-        if self._uri:
-            self._client = await Redis.from_url(self._uri, decode_responses=True)
-            logger.info("Redis connection was established.")
+        self._client = aredis.Redis.from_url(self._uri, decode_responses=True)

    async def disconnect(self):
-        if isinstance(self._client, Redis):
+        if self._client:
            await self._client.close()
-            logger.info("Redis connection was closed.")

    async def execute(self, command, *args, **kwargs):
        if self._client:
            try:
-                logger.debug(f"{command}")  # {args[0]}") # {args} {kwargs}")
-                for arg in args:
-                    if isinstance(arg, dict):
-                        if arg.get("_sa_instance_state"):
-                            del arg["_sa_instance_state"]
+                logger.debug(f"{command} {args} {kwargs}")
                r = await self._client.execute_command(command, *args, **kwargs)
-                # logger.debug(type(r))
-                # logger.debug(r)
+                logger.debug(type(r))
+                logger.debug(r)
                return r
            except Exception as e:
                logger.error(e)
@@ -60,22 +51,12 @@ class RedisService:
            return
        await self._client.publish(channel, data)

-    async def set(self, key, data, ex=None):
-        # Prepare the command arguments
-        args = [key, data]
+    async def mget(self, *keys):
+        return await self.execute('MGET', *keys)

-        # If an expiration time is provided, add it to the arguments
-        if ex is not None:
-            args.append("EX")
-            args.append(ex)
+    async def lrange(self, key, start, stop):
+        return await self.execute('LRANGE', key, start, stop)

-        # Execute the command with the provided arguments
-        await self.execute("set", *args)
-
-    async def get(self, key):
-        return await self.execute("get", key)
-
-
-redis = RedisService()
+redis = RedisCache()

 __all__ = ["redis"]
--- a/base/resolvers.py
+++ b/base/resolvers.py
@@ -0,0 +1,13 @@
+from ariadne import MutationType, QueryType, ScalarType
+
+datetime_scalar = ScalarType("DateTime")
+
+
+@datetime_scalar.serializer
+def serialize_datetime(value):
+    return value.isoformat()
+
+
+query = QueryType()
+mutation = MutationType()
+resolvers = [query, mutation, datetime_scalar]
--- a/cache/cache.py
+++ b/cache/cache.py
@@ -1,628 +0,0 @@
-"""
-Caching system for the Discours platform
----------------------------------------
-
-This module provides a comprehensive caching solution with these key components:
-
-1. KEY NAMING CONVENTIONS:
-   - Entity-based keys: "entity:property:value" (e.g., "author:id:123")
-   - Collection keys: "entity:collection:params" (e.g., "authors:stats:limit=10:offset=0")
-   - Special case keys: Maintained for backwards compatibility (e.g., "topic_shouts_123")
-
-2. CORE FUNCTIONS:
-   - cached_query(): High-level function for retrieving cached data or executing queries
-
-3. ENTITY-SPECIFIC FUNCTIONS:
-   - cache_author(), cache_topic(): Cache entity data
-   - get_cached_author(), get_cached_topic(): Retrieve entity data from cache
-   - invalidate_cache_by_prefix(): Invalidate all keys with a specific prefix
-
-4. CACHE INVALIDATION STRATEGY:
-   - Direct invalidation via invalidate_* functions for immediate changes
-   - Delayed invalidation via revalidation_manager for background processing
-   - Event-based triggers for automatic cache updates (see triggers.py)
-
-To maintain consistency with the existing codebase, this module preserves
-the original key naming patterns while providing a more structured approach
-for new cache operations.
-"""
-
-import asyncio
-import json
-from typing import Any, Dict, List, Optional, Union
-
-import orjson
-from sqlalchemy import and_, join, select
-
-from orm.author import Author, AuthorFollower
-from orm.shout import Shout, ShoutAuthor, ShoutTopic
-from orm.topic import Topic, TopicFollower
-from services.db import local_session
-from services.redis import redis
-from utils.encoders import CustomJSONEncoder
-from utils.logger import root_logger as logger
-
-DEFAULT_FOLLOWS = {
-    "topics": [],
-    "authors": [],
-    "shouts": [],
-    "communities": [{"id": 1, "name": "Дискурс", "slug": "discours", "pic": ""}],
-}
-
-CACHE_TTL = 300  # 5 minutes
-
-# Key templates for common entity types
-# These are used throughout the codebase and should be maintained for compatibility
-CACHE_KEYS = {
-    "TOPIC_ID": "topic:id:{}",
-    "TOPIC_SLUG": "topic:slug:{}",
-    "TOPIC_AUTHORS": "topic:authors:{}",
-    "TOPIC_FOLLOWERS": "topic:followers:{}",
-    "TOPIC_SHOUTS": "topic_shouts_{}",
-    "AUTHOR_ID": "author:id:{}",
-    "AUTHOR_USER": "author:user:{}",
-    "SHOUTS": "shouts:{}",
-}
-
-
-# Cache topic data
-async def cache_topic(topic: dict):
-    payload = json.dumps(topic, cls=CustomJSONEncoder)
-    await asyncio.gather(
-        redis.execute("SET", f"topic:id:{topic['id']}", payload),
-        redis.execute("SET", f"topic:slug:{topic['slug']}", payload),
-    )
-
-
-# Cache author data
-async def cache_author(author: dict):
-    payload = json.dumps(author, cls=CustomJSONEncoder)
-    await asyncio.gather(
-        redis.execute("SET", f"author:user:{author['user'].strip()}", str(author["id"])),
-        redis.execute("SET", f"author:id:{author['id']}", payload),
-    )
-
-
-# Cache follows data
-async def cache_follows(follower_id: int, entity_type: str, entity_id: int, is_insert=True):
-    key = f"author:follows-{entity_type}s:{follower_id}"
-    follows_str = await redis.execute("GET", key)
-    follows = orjson.loads(follows_str) if follows_str else DEFAULT_FOLLOWS[entity_type]
-    if is_insert:
-        if entity_id not in follows:
-            follows.append(entity_id)
-    else:
-        follows = [eid for eid in follows if eid != entity_id]
-    await redis.execute("SET", key, json.dumps(follows, cls=CustomJSONEncoder))
-    await update_follower_stat(follower_id, entity_type, len(follows))
-
-
-# Update follower statistics
-async def update_follower_stat(follower_id, entity_type, count):
-    follower_key = f"author:id:{follower_id}"
-    follower_str = await redis.execute("GET", follower_key)
-    follower = orjson.loads(follower_str) if follower_str else None
-    if follower:
-        follower["stat"] = {f"{entity_type}s": count}
-        await cache_author(follower)
-
-
-# Get author from cache
-async def get_cached_author(author_id: int, get_with_stat):
-    author_key = f"author:id:{author_id}"
-    result = await redis.execute("GET", author_key)
-    if result:
-        return orjson.loads(result)
-    # Load from database if not found in cache
-    q = select(Author).where(Author.id == author_id)
-    authors = get_with_stat(q)
-    if authors:
-        author = authors[0]
-        await cache_author(author.dict())
-        return author.dict()
-    return None
-
-
-# Function to get cached topic
-async def get_cached_topic(topic_id: int):
-    """
-    Fetch topic data from cache or database by id.
-
-    Args:
-        topic_id (int): The identifier for the topic.
-
-    Returns:
-        dict: Topic data or None if not found.
-    """
-    topic_key = f"topic:id:{topic_id}"
-    cached_topic = await redis.execute("GET", topic_key)
-    if cached_topic:
-        return orjson.loads(cached_topic)
-
-    # If not in cache, fetch from the database
-    with local_session() as session:
-        topic = session.execute(select(Topic).where(Topic.id == topic_id)).scalar_one_or_none()
-        if topic:
-            topic_dict = topic.dict()
-            await redis.execute("SET", topic_key, json.dumps(topic_dict, cls=CustomJSONEncoder))
-            return topic_dict
-
-    return None
-
-
-# Get topic by slug from cache
-async def get_cached_topic_by_slug(slug: str, get_with_stat):
-    topic_key = f"topic:slug:{slug}"
-    result = await redis.execute("GET", topic_key)
-    if result:
-        return orjson.loads(result)
-    # Load from database if not found in cache
-    topic_query = select(Topic).where(Topic.slug == slug)
-    topics = get_with_stat(topic_query)
-    if topics:
-        topic_dict = topics[0].dict()
-        await cache_topic(topic_dict)
-        return topic_dict
-    return None
-
-
-# Get list of authors by ID from cache
-async def get_cached_authors_by_ids(author_ids: List[int]) -> List[dict]:
-    # Fetch all author data concurrently
-    keys = [f"author:id:{author_id}" for author_id in author_ids]
-    results = await asyncio.gather(*(redis.execute("GET", key) for key in keys))
-    authors = [orjson.loads(result) if result else None for result in results]
-    # Load missing authors from database and cache
-    missing_indices = [index for index, author in enumerate(authors) if author is None]
-    if missing_indices:
-        missing_ids = [author_ids[index] for index in missing_indices]
-        with local_session() as session:
-            query = select(Author).where(Author.id.in_(missing_ids))
-            missing_authors = session.execute(query).scalars().all()
-            await asyncio.gather(*(cache_author(author.dict()) for author in missing_authors))
-            for index, author in zip(missing_indices, missing_authors):
-                authors[index] = author.dict()
-    return authors
-
-
-async def get_cached_topic_followers(topic_id: int):
-    """
-    Получает подписчиков темы по ID, используя кеш Redis.
-
-    Args:
-        topic_id: ID темы
-
-    Returns:
-        List[dict]: Список подписчиков с их данными
-    """
-    try:
-        cache_key = CACHE_KEYS["TOPIC_FOLLOWERS"].format(topic_id)
-        cached = await redis.execute("GET", cache_key)
-
-        if cached:
-            followers_ids = orjson.loads(cached)
-            logger.debug(f"Found {len(followers_ids)} cached followers for topic #{topic_id}")
-            return await get_cached_authors_by_ids(followers_ids)
-
-        with local_session() as session:
-            followers_ids = [
-                f[0]
-                for f in session.query(Author.id)
-                .join(TopicFollower, TopicFollower.follower == Author.id)
-                .filter(TopicFollower.topic == topic_id)
-                .all()
-            ]
-
-            await redis.execute("SETEX", cache_key, CACHE_TTL, orjson.dumps(followers_ids))
-            followers = await get_cached_authors_by_ids(followers_ids)
-            logger.debug(f"Cached {len(followers)} followers for topic #{topic_id}")
-            return followers
-
-    except Exception as e:
-        logger.error(f"Error getting followers for topic #{topic_id}: {str(e)}")
-        return []
-
-
-# Get cached author followers
-async def get_cached_author_followers(author_id: int):
-    # Check cache for data
-    cached = await redis.execute("GET", f"author:followers:{author_id}")
-    if cached:
-        followers_ids = orjson.loads(cached)
-        followers = await get_cached_authors_by_ids(followers_ids)
-        logger.debug(f"Cached followers for author #{author_id}: {len(followers)}")
-        return followers
-
-    # Query database if cache is empty
-    with local_session() as session:
-        followers_ids = [
-            f[0]
-            for f in session.query(Author.id)
-            .join(AuthorFollower, AuthorFollower.follower == Author.id)
-            .filter(AuthorFollower.author == author_id, Author.id != author_id)
-            .all()
-        ]
-        await redis.execute("SET", f"author:followers:{author_id}", orjson.dumps(followers_ids))
-        followers = await get_cached_authors_by_ids(followers_ids)
-        return followers
-
-
-# Get cached follower authors
-async def get_cached_follower_authors(author_id: int):
-    # Attempt to retrieve authors from cache
-    cached = await redis.execute("GET", f"author:follows-authors:{author_id}")
-    if cached:
-        authors_ids = orjson.loads(cached)
-    else:
-        # Query authors from database
-        with local_session() as session:
-            authors_ids = [
-                a[0]
-                for a in session.execute(
-                    select(Author.id)
-                    .select_from(join(Author, AuthorFollower, Author.id == AuthorFollower.author))
-                    .where(AuthorFollower.follower == author_id)
-                ).all()
-            ]
-            await redis.execute("SET", f"author:follows-authors:{author_id}", orjson.dumps(authors_ids))
-
-    authors = await get_cached_authors_by_ids(authors_ids)
-    return authors
-
-
-# Get cached follower topics
-async def get_cached_follower_topics(author_id: int):
-    # Attempt to retrieve topics from cache
-    cached = await redis.execute("GET", f"author:follows-topics:{author_id}")
-    if cached:
-        topics_ids = orjson.loads(cached)
-    else:
-        # Load topics from database and cache them
-        with local_session() as session:
-            topics_ids = [
-                t[0]
-                for t in session.query(Topic.id)
-                .join(TopicFollower, TopicFollower.topic == Topic.id)
-                .where(TopicFollower.follower == author_id)
-                .all()
-            ]
-            await redis.execute("SET", f"author:follows-topics:{author_id}", orjson.dumps(topics_ids))
-
-    topics = []
-    for topic_id in topics_ids:
-        topic_str = await redis.execute("GET", f"topic:id:{topic_id}")
-        if topic_str:
-            topic = orjson.loads(topic_str)
-            if topic and topic not in topics:
-                topics.append(topic)
-
-    logger.debug(f"Cached topics for author#{author_id}: {len(topics)}")
-    return topics
-
-
-# Get author by user ID from cache
-async def get_cached_author_by_user_id(user_id: str, get_with_stat):
-    """
-    Retrieve author information by user_id, checking the cache first, then the database.
-
-    Args:
-        user_id (str): The user identifier for which to retrieve the author.
-
-    Returns:
-        dict: Dictionary with author data or None if not found.
-    """
-    # Attempt to find author ID by user_id in Redis cache
-    author_id = await redis.execute("GET", f"author:user:{user_id.strip()}")
-    if author_id:
-        # If ID is found, get full author data by ID
-        author_data = await redis.execute("GET", f"author:id:{author_id}")
-        if author_data:
-            return orjson.loads(author_data)
-
-    # If data is not found in cache, query the database
-    author_query = select(Author).where(Author.user == user_id)
-    authors = get_with_stat(author_query)
-    if authors:
-        # Cache the retrieved author data
-        author = authors[0]
-        author_dict = author.dict()
-        await asyncio.gather(
-            redis.execute("SET", f"author:user:{user_id.strip()}", str(author.id)),
-            redis.execute("SET", f"author:id:{author.id}", orjson.dumps(author_dict)),
-        )
-        return author_dict
-
-    # Return None if author is not found
-    return None
-
-
-# Get cached topic authors
-async def get_cached_topic_authors(topic_id: int):
-    """
-    Retrieve a list of authors for a given topic, using cache or database.
-
-    Args:
-        topic_id (int): The identifier of the topic for which to retrieve authors.
-
-    Returns:
-        List[dict]: A list of dictionaries containing author data.
-    """
-    # Attempt to get a list of author IDs from cache
-    rkey = f"topic:authors:{topic_id}"
-    cached_authors_ids = await redis.execute("GET", rkey)
-    if cached_authors_ids:
-        authors_ids = orjson.loads(cached_authors_ids)
-    else:
-        # If cache is empty, get data from the database
-        with local_session() as session:
-            query = (
-                select(ShoutAuthor.author)
-                .select_from(join(ShoutTopic, Shout, ShoutTopic.shout == Shout.id))
-                .join(ShoutAuthor, ShoutAuthor.shout == Shout.id)
-                .where(and_(ShoutTopic.topic == topic_id, Shout.published_at.is_not(None), Shout.deleted_at.is_(None)))
-            )
-            authors_ids = [author_id for (author_id,) in session.execute(query).all()]
-            # Cache the retrieved author IDs
-            await redis.execute("SET", rkey, orjson.dumps(authors_ids))
-
-    # Retrieve full author details from cached IDs
-    if authors_ids:
-        authors = await get_cached_authors_by_ids(authors_ids)
-        logger.debug(f"Topic#{topic_id} authors fetched and cached: {len(authors)} authors found.")
-        return authors
-
-    return []
-
-
-async def invalidate_shouts_cache(cache_keys: List[str]):
-    """
-    Инвалидирует кэш выборок публикаций по переданным ключам.
-    """
-    for key in cache_keys:
-        try:
-            # Формируем полный ключ кэша
-            cache_key = f"shouts:{key}"
-
-            # Удаляем основной кэш
-            await redis.execute("DEL", cache_key)
-            logger.debug(f"Invalidated cache key: {cache_key}")
-
-            # Добавляем ключ в список инвалидированных с TTL
-            await redis.execute("SETEX", f"{cache_key}:invalidated", CACHE_TTL, "1")
-
-            # Если это кэш темы, инвалидируем также связанные ключи
-            if key.startswith("topic_"):
-                topic_id = key.split("_")[1]
-                related_keys = [
-                    f"topic:id:{topic_id}",
-                    f"topic:authors:{topic_id}",
-                    f"topic:followers:{topic_id}",
-                    f"topic:stats:{topic_id}",
-                ]
-                for related_key in related_keys:
-                    await redis.execute("DEL", related_key)
-                    logger.debug(f"Invalidated related key: {related_key}")
-
-        except Exception as e:
-            logger.error(f"Error invalidating cache key {key}: {e}")
-
-
-async def cache_topic_shouts(topic_id: int, shouts: List[dict]):
-    """Кэширует список публикаций для темы"""
-    key = f"topic_shouts_{topic_id}"
-    payload = json.dumps(shouts, cls=CustomJSONEncoder)
-    await redis.execute("SETEX", key, CACHE_TTL, payload)
-
-
-async def get_cached_topic_shouts(topic_id: int) -> List[dict]:
-    """Получает кэшированный список публикаций для темы"""
-    key = f"topic_shouts_{topic_id}"
-    cached = await redis.execute("GET", key)
-    if cached:
-        return orjson.loads(cached)
-    return None
-
-
-async def cache_related_entities(shout: Shout):
-    """
-    Кэширует все связанные с публикацией сущности (авторов и темы)
-    """
-    tasks = []
-    for author in shout.authors:
-        tasks.append(cache_by_id(Author, author.id, cache_author))
-    for topic in shout.topics:
-        tasks.append(cache_by_id(Topic, topic.id, cache_topic))
-    await asyncio.gather(*tasks)
-
-
-async def invalidate_shout_related_cache(shout: Shout, author_id: int):
-    """
-    Инвалидирует весь кэш, связанный с публикацией и её связями
-
-    Args:
-        shout: Объект публикации
-        author_id: ID автора
-    """
-    cache_keys = {
-        "feed",  # основная лента
-        f"author_{author_id}",  # публикации автора
-        "random_top",  # случайные топовые
-        "unrated",  # неоцененные
-        "recent",  # последние
-        "coauthored",  # совместные
-    }
-
-    # Добавляем ключи авторов
-    cache_keys.update(f"author_{a.id}" for a in shout.authors)
-    cache_keys.update(f"authored_{a.id}" for a in shout.authors)
-
-    # Добавляем ключи тем
-    cache_keys.update(f"topic_{t.id}" for t in shout.topics)
-    cache_keys.update(f"topic_shouts_{t.id}" for t in shout.topics)
-
-    await invalidate_shouts_cache(list(cache_keys))
-
-
-# Function removed - direct Redis calls used throughout the module instead
-
-
-async def get_cached_entity(entity_type: str, entity_id: int, get_method, cache_method):
-    """
-    Универсальная функция получения кэшированной сущности
-
-    Args:
-        entity_type: 'author' или 'topic'
-        entity_id: ID сущности
-        get_method: метод получения из БД
-        cache_method: метод кэширования
-    """
-    key = f"{entity_type}:id:{entity_id}"
-    cached = await redis.execute("GET", key)
-    if cached:
-        return orjson.loads(cached)
-
-    entity = await get_method(entity_id)
-    if entity:
-        await cache_method(entity)
-        return entity
-    return None
-
-
-async def cache_by_id(entity, entity_id: int, cache_method):
-    """
-    Кэширует сущность по ID, используя указанный метод кэширования
-
-    Args:
-        entity: класс сущности (Author/Topic)
-        entity_id: ID сущности
-        cache_method: функция кэширования
-    """
-    from resolvers.stat import get_with_stat
-
-    caching_query = select(entity).filter(entity.id == entity_id)
-    result = get_with_stat(caching_query)
-    if not result or not result[0]:
-        logger.warning(f"{entity.__name__} with id {entity_id} not found")
-        return
-    x = result[0]
-    d = x.dict()
-    await cache_method(d)
-    return d
-
-
-# Универсальная функция для сохранения данных в кеш
-async def cache_data(key: str, data: Any, ttl: Optional[int] = None) -> None:
-    """
-    Сохраняет данные в кеш по указанному ключу.
-
-    Args:
-        key: Ключ кеша
-        data: Данные для сохранения
-        ttl: Время жизни кеша в секундах (None - бессрочно)
-    """
-    try:
-        payload = json.dumps(data, cls=CustomJSONEncoder)
-        if ttl:
-            await redis.execute("SETEX", key, ttl, payload)
-        else:
-            await redis.execute("SET", key, payload)
-        logger.debug(f"Данные сохранены в кеш по ключу {key}")
-    except Exception as e:
-        logger.error(f"Ошибка при сохранении данных в кеш: {e}")
-
-
-# Универсальная функция для получения данных из кеша
-async def get_cached_data(key: str) -> Optional[Any]:
-    """
-    Получает данные из кеша по указанному ключу.
-
-    Args:
-        key: Ключ кеша
-
-    Returns:
-        Any: Данные из кеша или None, если данных нет
-    """
-    try:
-        cached_data = await redis.execute("GET", key)
-        if cached_data:
-            loaded = orjson.loads(cached_data)
-            logger.debug(f"Данные получены из кеша по ключу {key}: {len(loaded)}")
-            return loaded
-        return None
-    except Exception as e:
-        logger.error(f"Ошибка при получении данных из кеша: {e}")
-        return None
-
-
-# Универсальная функция для инвалидации кеша по префиксу
-async def invalidate_cache_by_prefix(prefix: str) -> None:
-    """
-    Инвалидирует все ключи кеша с указанным префиксом.
-
-    Args:
-        prefix: Префикс ключей кеша для инвалидации
-    """
-    try:
-        keys = await redis.execute("KEYS", f"{prefix}:*")
-        if keys:
-            await redis.execute("DEL", *keys)
-            logger.debug(f"Удалено {len(keys)} ключей кеша с префиксом {prefix}")
-    except Exception as e:
-        logger.error(f"Ошибка при инвалидации кеша: {e}")
-
-
-# Универсальная функция для получения и кеширования данных
-async def cached_query(
-    cache_key: str,
-    query_func: callable,
-    ttl: Optional[int] = None,
-    force_refresh: bool = False,
-    use_key_format: bool = True,
-    **query_params,
-) -> Any:
-    """
-    Gets data from cache or executes query and saves result to cache.
-    Supports existing key formats for compatibility.
-
-    Args:
-        cache_key: Cache key or key template from CACHE_KEYS
-        query_func: Function to execute the query
-        ttl: Cache TTL in seconds (None - indefinite)
-        force_refresh: Force cache refresh
-        use_key_format: Whether to check if cache_key matches a key template in CACHE_KEYS
-        **query_params: Parameters to pass to the query function
-
-    Returns:
-        Any: Data from cache or query result
-    """
-    # Check if cache_key matches a pattern in CACHE_KEYS
-    actual_key = cache_key
-    if use_key_format and "{}" in cache_key:
-        # Look for a template match in CACHE_KEYS
-        for key_name, key_format in CACHE_KEYS.items():
-            if cache_key == key_format:
-                # We have a match, now look for the id or value to format with
-                for param_name, param_value in query_params.items():
-                    if param_name in ["id", "slug", "user", "topic_id", "author_id"]:
-                        actual_key = cache_key.format(param_value)
-                        break
-
-    # If not forcing refresh, try to get data from cache
-    if not force_refresh:
-        cached_result = await get_cached_data(actual_key)
-        if cached_result is not None:
-            return cached_result
-
-    # If data not in cache or refresh required, execute query
-    try:
-        result = await query_func(**query_params)
-        if result is not None:
-            # Save result to cache
-            await cache_data(actual_key, result, ttl)
-        return result
-    except Exception as e:
-        logger.error(f"Error executing query for caching: {e}")
-        # In case of error, return data from cache if not forcing refresh
-        if not force_refresh:
-            return await get_cached_data(actual_key)
-        raise
--- a/cache/precache.py
+++ b/cache/precache.py
@@ -1,133 +0,0 @@
-import asyncio
-import json
-
-from sqlalchemy import and_, join, select
-
-from cache.cache import cache_author, cache_topic
-from orm.author import Author, AuthorFollower
-from orm.shout import Shout, ShoutAuthor, ShoutReactionsFollower, ShoutTopic
-from orm.topic import Topic, TopicFollower
-from resolvers.stat import get_with_stat
-from services.db import local_session
-from services.redis import redis
-from utils.encoders import CustomJSONEncoder
-from utils.logger import root_logger as logger
-
-
-# Предварительное кеширование подписчиков автора
-async def precache_authors_followers(author_id, session):
-    authors_followers = set()
-    followers_query = select(AuthorFollower.follower).where(AuthorFollower.author == author_id)
-    result = session.execute(followers_query)
-    authors_followers.update(row[0] for row in result if row[0])
-
-    followers_payload = json.dumps(list(authors_followers), cls=CustomJSONEncoder)
-    await redis.execute("SET", f"author:followers:{author_id}", followers_payload)
-
-
-# Предварительное кеширование подписок автора
-async def precache_authors_follows(author_id, session):
-    follows_topics_query = select(TopicFollower.topic).where(TopicFollower.follower == author_id)
-    follows_authors_query = select(AuthorFollower.author).where(AuthorFollower.follower == author_id)
-    follows_shouts_query = select(ShoutReactionsFollower.shout).where(ShoutReactionsFollower.follower == author_id)
-
-    follows_topics = {row[0] for row in session.execute(follows_topics_query) if row[0]}
-    follows_authors = {row[0] for row in session.execute(follows_authors_query) if row[0]}
-    follows_shouts = {row[0] for row in session.execute(follows_shouts_query) if row[0]}
-
-    topics_payload = json.dumps(list(follows_topics), cls=CustomJSONEncoder)
-    authors_payload = json.dumps(list(follows_authors), cls=CustomJSONEncoder)
-    shouts_payload = json.dumps(list(follows_shouts), cls=CustomJSONEncoder)
-
-    await asyncio.gather(
-        redis.execute("SET", f"author:follows-topics:{author_id}", topics_payload),
-        redis.execute("SET", f"author:follows-authors:{author_id}", authors_payload),
-        redis.execute("SET", f"author:follows-shouts:{author_id}", shouts_payload),
-    )
-
-
-# Предварительное кеширование авторов тем
-async def precache_topics_authors(topic_id: int, session):
-    topic_authors_query = (
-        select(ShoutAuthor.author)
-        .select_from(join(ShoutTopic, Shout, ShoutTopic.shout == Shout.id))
-        .join(ShoutAuthor, ShoutAuthor.shout == Shout.id)
-        .filter(
-            and_(
-                ShoutTopic.topic == topic_id,
-                Shout.published_at.is_not(None),
-                Shout.deleted_at.is_(None),
-            )
-        )
-    )
-    topic_authors = {row[0] for row in session.execute(topic_authors_query) if row[0]}
-
-    authors_payload = json.dumps(list(topic_authors), cls=CustomJSONEncoder)
-    await redis.execute("SET", f"topic:authors:{topic_id}", authors_payload)
-
-
-# Предварительное кеширование подписчиков тем
-async def precache_topics_followers(topic_id: int, session):
-    followers_query = select(TopicFollower.follower).where(TopicFollower.topic == topic_id)
-    topic_followers = {row[0] for row in session.execute(followers_query) if row[0]}
-
-    followers_payload = json.dumps(list(topic_followers), cls=CustomJSONEncoder)
-    await redis.execute("SET", f"topic:followers:{topic_id}", followers_payload)
-
-
-async def precache_data():
-    logger.info("precaching...")
-    try:
-        key = "authorizer_env"
-        # cache reset
-        value = await redis.execute("HGETALL", key)
-        await redis.execute("FLUSHDB")
-        logger.info("redis: FLUSHDB")
-
-        # Преобразуем словарь в список аргументов для HSET
-        if value:
-            # Если значение - словарь, преобразуем его в плоский список для HSET
-            if isinstance(value, dict):
-                flattened = []
-                for field, val in value.items():
-                    flattened.extend([field, val])
-                await redis.execute("HSET", key, *flattened)
-            else:
-                # Предполагаем, что значение уже содержит список
-                await redis.execute("HSET", key, *value)
-            logger.info(f"redis hash '{key}' was restored")
-
-        with local_session() as session:
-            # topics
-            q = select(Topic).where(Topic.community == 1)
-            topics = get_with_stat(q)
-            for topic in topics:
-                topic_dict = topic.dict() if hasattr(topic, "dict") else topic
-                await cache_topic(topic_dict)
-                await asyncio.gather(
-                    precache_topics_followers(topic_dict["id"], session),
-                    precache_topics_authors(topic_dict["id"], session),
-                )
-            logger.info(f"{len(topics)} topics and their followings precached")
-
-            # authors
-            authors = get_with_stat(select(Author).where(Author.user.is_not(None)))
-            logger.info(f"{len(authors)} authors found in database")
-            for author in authors:
-                if isinstance(author, Author):
-                    profile = author.dict()
-                    author_id = profile.get("id")
-                    user_id = profile.get("user", "").strip()
-                    if author_id and user_id:
-                        await cache_author(profile)
-                        await asyncio.gather(
-                            precache_authors_followers(author_id, session), precache_authors_follows(author_id, session)
-                        )
-                else:
-                    logger.error(f"fail caching {author}")
-            logger.info(f"{len(authors)} authors and their followings precached")
-    except Exception as exc:
-        import traceback
-
-        traceback.print_exc()
-        logger.error(f"Error in precache_data: {exc}")
--- a/cache/revalidator.py
+++ b/cache/revalidator.py
@@ -1,172 +0,0 @@
-import asyncio
-
-from cache.cache import (
-    cache_author,
-    cache_topic,
-    get_cached_author,
-    get_cached_topic,
-    invalidate_cache_by_prefix,
-)
-from resolvers.stat import get_with_stat
-from services.redis import redis
-from utils.logger import root_logger as logger
-
-CACHE_REVALIDATION_INTERVAL = 300  # 5 minutes
-
-
-class CacheRevalidationManager:
-    def __init__(self, interval=CACHE_REVALIDATION_INTERVAL):
-        """Инициализация менеджера с заданным интервалом проверки (в секундах)."""
-        self.interval = interval
-        self.items_to_revalidate = {"authors": set(), "topics": set(), "shouts": set(), "reactions": set()}
-        self.lock = asyncio.Lock()
-        self.running = True
-        self.MAX_BATCH_SIZE = 10  # Максимальное количество элементов для поштучной обработки
-        self._redis = redis  # Добавлена инициализация _redis для доступа к Redis-клиенту
-
-    async def start(self):
-        """Запуск фонового воркера для ревалидации кэша."""
-        # Проверяем, что у нас есть соединение с Redis
-        if not self._redis._client:
-            logger.warning("Redis connection not established. Waiting for connection...")
-            try:
-                await self._redis.connect()
-                logger.info("Redis connection established for revalidation manager")
-            except Exception as e:
-                logger.error(f"Failed to connect to Redis: {e}")
-                
-        self.task = asyncio.create_task(self.revalidate_cache())
-
-    async def revalidate_cache(self):
-        """Циклическая проверка и ревалидация кэша каждые self.interval секунд."""
-        try:
-            while self.running:
-                await asyncio.sleep(self.interval)
-                await self.process_revalidation()
-        except asyncio.CancelledError:
-            logger.info("Revalidation worker was stopped.")
-        except Exception as e:
-            logger.error(f"An error occurred in the revalidation worker: {e}")
-
-    async def process_revalidation(self):
-        """Обновление кэша для всех сущностей, требующих ревалидации."""
-        # Проверяем соединение с Redis
-        if not self._redis._client:
-            return  # Выходим из метода, если не удалось подключиться
-                
-        async with self.lock:
-            # Ревалидация кэша авторов
-            if self.items_to_revalidate["authors"]:
-                logger.debug(f"Revalidating {len(self.items_to_revalidate['authors'])} authors")
-                for author_id in self.items_to_revalidate["authors"]:
-                    if author_id == "all":
-                        await invalidate_cache_by_prefix("authors")
-                        break
-                    author = await get_cached_author(author_id, get_with_stat)
-                    if author:
-                        await cache_author(author)
-                self.items_to_revalidate["authors"].clear()
-
-            # Ревалидация кэша тем
-            if self.items_to_revalidate["topics"]:
-                logger.debug(f"Revalidating {len(self.items_to_revalidate['topics'])} topics")
-                for topic_id in self.items_to_revalidate["topics"]:
-                    if topic_id == "all":
-                        await invalidate_cache_by_prefix("topics")
-                        break
-                    topic = await get_cached_topic(topic_id)
-                    if topic:
-                        await cache_topic(topic)
-                self.items_to_revalidate["topics"].clear()
-
-            # Ревалидация шаутов (публикаций)
-            if self.items_to_revalidate["shouts"]:
-                shouts_count = len(self.items_to_revalidate["shouts"])
-                logger.debug(f"Revalidating {shouts_count} shouts")
-
-                # Проверяем наличие специального флага 'all'
-                if "all" in self.items_to_revalidate["shouts"]:
-                    await invalidate_cache_by_prefix("shouts")
-                # Если элементов много, но не 'all', используем специфический подход
-                elif shouts_count > self.MAX_BATCH_SIZE:
-                    # Инвалидируем только collections keys, которые затрагивают много сущностей
-                    collection_keys = await asyncio.create_task(self._redis.execute("KEYS", "shouts:*"))
-                    if collection_keys:
-                        await self._redis.execute("DEL", *collection_keys)
-                        logger.debug(f"Удалено {len(collection_keys)} коллекционных ключей шаутов")
-
-                    # Обновляем кеш каждого конкретного шаута
-                    for shout_id in self.items_to_revalidate["shouts"]:
-                        if shout_id != "all":
-                            # Точечная инвалидация для каждого shout_id
-                            specific_keys = [f"shout:id:{shout_id}"]
-                            for key in specific_keys:
-                                await self._redis.execute("DEL", key)
-                                logger.debug(f"Удален ключ кеша {key}")
-                else:
-                    # Если элементов немного, обрабатываем каждый
-                    for shout_id in self.items_to_revalidate["shouts"]:
-                        if shout_id != "all":
-                            # Точечная инвалидация для каждого shout_id
-                            specific_keys = [f"shout:id:{shout_id}"]
-                            for key in specific_keys:
-                                await self._redis.execute("DEL", key)
-                                logger.debug(f"Удален ключ кеша {key}")
-
-                self.items_to_revalidate["shouts"].clear()
-
-            # Аналогично для реакций - точечная инвалидация
-            if self.items_to_revalidate["reactions"]:
-                reactions_count = len(self.items_to_revalidate["reactions"])
-                logger.debug(f"Revalidating {reactions_count} reactions")
-
-                if "all" in self.items_to_revalidate["reactions"]:
-                    await invalidate_cache_by_prefix("reactions")
-                elif reactions_count > self.MAX_BATCH_SIZE:
-                    # Инвалидируем только collections keys для реакций
-                    collection_keys = await asyncio.create_task(self._redis.execute("KEYS", "reactions:*"))
-                    if collection_keys:
-                        await self._redis.execute("DEL", *collection_keys)
-                        logger.debug(f"Удалено {len(collection_keys)} коллекционных ключей реакций")
-
-                    # Точечная инвалидация для каждой реакции
-                    for reaction_id in self.items_to_revalidate["reactions"]:
-                        if reaction_id != "all":
-                            specific_keys = [f"reaction:id:{reaction_id}"]
-                            for key in specific_keys:
-                                await self._redis.execute("DEL", key)
-                                logger.debug(f"Удален ключ кеша {key}")
-                else:
-                    # Точечная инвалидация для каждой реакции
-                    for reaction_id in self.items_to_revalidate["reactions"]:
-                        if reaction_id != "all":
-                            specific_keys = [f"reaction:id:{reaction_id}"]
-                            for key in specific_keys:
-                                await self._redis.execute("DEL", key)
-                                logger.debug(f"Удален ключ кеша {key}")
-
-                self.items_to_revalidate["reactions"].clear()
-
-    def mark_for_revalidation(self, entity_id, entity_type):
-        """Отметить сущность для ревалидации."""
-        if entity_id and entity_type:
-            self.items_to_revalidate[entity_type].add(entity_id)
-
-    def invalidate_all(self, entity_type):
-        """Пометить для инвалидации все элементы указанного типа."""
-        logger.debug(f"Marking all {entity_type} for invalidation")
-        # Особый флаг для полной инвалидации
-        self.items_to_revalidate[entity_type].add("all")
-
-    async def stop(self):
-        """Остановка фонового воркера."""
-        self.running = False
-        if hasattr(self, "task"):
-            self.task.cancel()
-            try:
-                await self.task
-            except asyncio.CancelledError:
-                pass
-
-
-revalidation_manager = CacheRevalidationManager()
--- a/cache/triggers.py
+++ b/cache/triggers.py
@@ -1,131 +0,0 @@
-from sqlalchemy import event
-
-from cache.revalidator import revalidation_manager
-from orm.author import Author, AuthorFollower
-from orm.reaction import Reaction, ReactionKind
-from orm.shout import Shout, ShoutAuthor, ShoutReactionsFollower
-from orm.topic import Topic, TopicFollower
-from services.db import local_session
-from utils.logger import root_logger as logger
-
-
-def mark_for_revalidation(entity, *args):
-    """Отметка сущности для ревалидации."""
-    entity_type = (
-        "authors"
-        if isinstance(entity, Author)
-        else "topics"
-        if isinstance(entity, Topic)
-        else "reactions"
-        if isinstance(entity, Reaction)
-        else "shouts"
-        if isinstance(entity, Shout)
-        else None
-    )
-    if entity_type:
-        revalidation_manager.mark_for_revalidation(entity.id, entity_type)
-
-
-def after_follower_handler(mapper, connection, target, is_delete=False):
-    """Обработчик добавления, обновления или удаления подписки."""
-    entity_type = None
-    if isinstance(target, AuthorFollower):
-        entity_type = "authors"
-    elif isinstance(target, TopicFollower):
-        entity_type = "topics"
-    elif isinstance(target, ShoutReactionsFollower):
-        entity_type = "shouts"
-
-    if entity_type:
-        revalidation_manager.mark_for_revalidation(
-            target.author if entity_type == "authors" else target.topic, entity_type
-        )
-        if not is_delete:
-            revalidation_manager.mark_for_revalidation(target.follower, "authors")
-
-
-def after_shout_handler(mapper, connection, target):
-    """Обработчик изменения статуса публикации"""
-    if not isinstance(target, Shout):
-        return
-
-    # Проверяем изменение статуса публикации
-    # was_published = target.published_at is not None and target.deleted_at is None
-
-    # Всегда обновляем счетчики для авторов и тем при любом изменении поста
-    for author in target.authors:
-        revalidation_manager.mark_for_revalidation(author.id, "authors")
-
-    for topic in target.topics:
-        revalidation_manager.mark_for_revalidation(topic.id, "topics")
-
-    # Обновляем сам пост
-    revalidation_manager.mark_for_revalidation(target.id, "shouts")
-
-
-def after_reaction_handler(mapper, connection, target):
-    """Обработчик для комментариев"""
-    if not isinstance(target, Reaction):
-        return
-
-    # Проверяем что это комментарий
-    is_comment = target.kind == ReactionKind.COMMENT.value
-
-    # Получаем связанный пост
-    shout_id = target.shout if isinstance(target.shout, int) else target.shout.id
-    if not shout_id:
-        return
-
-    # Обновляем счетчики для автора комментария
-    if target.created_by:
-        revalidation_manager.mark_for_revalidation(target.created_by, "authors")
-
-    # Обновляем счетчики для поста
-    revalidation_manager.mark_for_revalidation(shout_id, "shouts")
-
-    if is_comment:
-        # Для комментариев обновляем также авторов и темы
-        with local_session() as session:
-            shout = (
-                session.query(Shout)
-                .filter(Shout.id == shout_id, Shout.published_at.is_not(None), Shout.deleted_at.is_(None))
-                .first()
-            )
-
-            if shout:
-                for author in shout.authors:
-                    revalidation_manager.mark_for_revalidation(author.id, "authors")
-
-                for topic in shout.topics:
-                    revalidation_manager.mark_for_revalidation(topic.id, "topics")
-
-
-def events_register():
-    """Регистрация обработчиков событий для всех сущностей."""
-    event.listen(ShoutAuthor, "after_insert", mark_for_revalidation)
-    event.listen(ShoutAuthor, "after_update", mark_for_revalidation)
-    event.listen(ShoutAuthor, "after_delete", mark_for_revalidation)
-
-    event.listen(AuthorFollower, "after_insert", after_follower_handler)
-    event.listen(AuthorFollower, "after_update", after_follower_handler)
-    event.listen(AuthorFollower, "after_delete", lambda *args: after_follower_handler(*args, is_delete=True))
-
-    event.listen(TopicFollower, "after_insert", after_follower_handler)
-    event.listen(TopicFollower, "after_update", after_follower_handler)
-    event.listen(TopicFollower, "after_delete", lambda *args: after_follower_handler(*args, is_delete=True))
-
-    event.listen(ShoutReactionsFollower, "after_insert", after_follower_handler)
-    event.listen(ShoutReactionsFollower, "after_update", after_follower_handler)
-    event.listen(ShoutReactionsFollower, "after_delete", lambda *args: after_follower_handler(*args, is_delete=True))
-
-    event.listen(Reaction, "after_update", mark_for_revalidation)
-    event.listen(Author, "after_update", mark_for_revalidation)
-    event.listen(Topic, "after_update", mark_for_revalidation)
-    event.listen(Shout, "after_update", after_shout_handler)
-    event.listen(Shout, "after_delete", after_shout_handler)
-
-    event.listen(Reaction, "after_insert", after_reaction_handler)
-    event.listen(Reaction, "after_update", after_reaction_handler)
-    event.listen(Reaction, "after_delete", after_reaction_handler)
-
-    logger.info("Event handlers registered successfully.")
--- a/checks.sh
+++ b/checks.sh
@@ -0,0 +1,10 @@
+#!/usr/bin/env bash
+
+echo "> isort"
+isort .
+echo "> black"
+black .
+echo "> flake8"
+flake8 .
+# echo "> mypy"
+# mypy .
--- a/docs/caching.md
+++ b/docs/caching.md
@@ -1,295 +0,0 @@
-# Система кеширования Discours
-
-## Общее описание
-
-Система кеширования Discours - это комплексное решение для повышения производительности платформы. Она использует Redis для хранения часто запрашиваемых данных и уменьшения нагрузки на основную базу данных.
-
-Кеширование реализовано как многоуровневая система, состоящая из нескольких модулей:
-
- `cache.py` - основной модуль с функциями кеширования
- `revalidator.py` - асинхронный менеджер ревалидации кеша
- `triggers.py` - триггеры событий SQLAlchemy для автоматической ревалидации
- `precache.py` - предварительное кеширование данных при старте приложения
-
-## Ключевые компоненты
-
-### 1. Форматы ключей кеша
-
-Система поддерживает несколько форматов ключей для обеспечения совместимости и удобства использования:
-
- **Ключи сущностей**: `entity:property:value` (например, `author:id:123`)
- **Ключи коллекций**: `entity:collection:params` (например, `authors:stats:limit=10:offset=0`)
- **Специальные ключи**: для обратной совместимости (например, `topic_shouts_123`)
-
-Все стандартные форматы ключей хранятся в словаре `CACHE_KEYS`:
-
-```python
-CACHE_KEYS = {
-    "TOPIC_ID": "topic:id:{}",
-    "TOPIC_SLUG": "topic:slug:{}",
-    "AUTHOR_ID": "author:id:{}",
-    # и другие...
-}
-```
-
-### 2. Основные функции кеширования
-
-#### Структура ключей
-
-Вместо генерации ключей через вспомогательные функции, система следует строгим конвенциям формирования ключей:
-
-1. **Ключи для отдельных сущностей** строятся по шаблону:
-   ```
-   entity:property:value
-   ```
-   Например:
-   - `topic:id:123` - тема с ID 123
-   - `author:slug:john-doe` - автор со слагом "john-doe"
-   - `shout:id:456` - публикация с ID 456
-
-2. **Ключи для коллекций** строятся по шаблону:
-   ```
-   entity:collection[:filter1=value1:filter2=value2:...]
-   ```
-   Например:
-   - `topics:all:basic` - базовый список всех тем
-   - `authors:stats:limit=10:offset=0:sort=name` - отсортированный список авторов с пагинацией
-   - `shouts:feed:limit=20:community=1` - лента публикаций с фильтром по сообществу
-
-3. **Специальные форматы ключей** для обратной совместимости:
-   ```
-   entity_action_id
-   ```
-   Например:
-   - `topic_shouts_123` - публикации для темы с ID 123
-
-Во всех модулях системы разработчики должны явно формировать ключи в соответствии с этими конвенциями, что обеспечивает единообразие и предсказуемость кеширования.
-
-#### Работа с данными в кеше
-
-```python
-async def cache_data(key, data, ttl=None)
-async def get_cached_data(key)
-```
-
-Эти функции предоставляют универсальный интерфейс для сохранения и получения данных из кеша. Они напрямую используют Redis через вызовы `redis.execute()`.
-
-#### Высокоуровневое кеширование запросов
-
-```python
-async def cached_query(cache_key, query_func, ttl=None, force_refresh=False, **query_params)
-```
-
-Функция `cached_query` объединяет получение данных из кеша и выполнение запроса в случае отсутствия данных в кеше. Это основная функция, которую следует использовать в резолверах для кеширования результатов запросов.
-
-### 3. Кеширование сущностей
-
-Для основных типов сущностей реализованы специальные функции:
-
-```python
-async def cache_topic(topic: dict)
-async def cache_author(author: dict)
-async def get_cached_topic(topic_id: int)
-async def get_cached_author(author_id: int, get_with_stat)
-```
-
-Эти функции упрощают работу с часто используемыми типами данных и обеспечивают единообразный подход к их кешированию.
-
-### 4. Работа со связями
-
-Для работы со связями между сущностями предназначены функции:
-
-```python
-async def cache_follows(follower_id, entity_type, entity_id, is_insert=True)
-async def get_cached_topic_followers(topic_id)
-async def get_cached_author_followers(author_id)
-async def get_cached_follower_topics(author_id)
-```
-
-Они позволяют эффективно кешировать и получать информацию о подписках, связях между авторами, темами и публикациями.
-
-## Система инвалидации кеша
-
-### 1. Прямая инвалидация
-
-Система поддерживает два типа инвалидации кеша:
-
-#### 1.1. Инвалидация по префиксу
-
-```python
-async def invalidate_cache_by_prefix(prefix)
-```
-
-Позволяет инвалидировать все ключи кеша, начинающиеся с указанного префикса. Используется в резолверах для инвалидации группы кешей при массовых изменениях.
-
-#### 1.2. Точечная инвалидация
-
-```python
-async def invalidate_authors_cache(author_id=None)
-async def invalidate_topics_cache(topic_id=None)
-```
-
-Эти функции позволяют инвалидировать кеш только для конкретной сущности, что снижает нагрузку на Redis и предотвращает ненужную потерю кешированных данных. Если ID сущности не указан, используется инвалидация по префиксу.
-
-Примеры использования точечной инвалидации:
-
-```python
-# Инвалидация кеша только для автора с ID 123
-await invalidate_authors_cache(123)
-
-# Инвалидация кеша только для темы с ID 456
-await invalidate_topics_cache(456)
-```
-
-### 2. Отложенная инвалидация
-
-Модуль `revalidator.py` реализует систему отложенной инвалидации кеша через класс `CacheRevalidationManager`:
-
-```python
-class CacheRevalidationManager:
-    def __init__(self, interval=CACHE_REVALIDATION_INTERVAL):
-        # ...
-        self._redis = redis  # Прямая ссылка на сервис Redis
-        
-    async def start(self):
-        # Проверка и установка соединения с Redis
-        # ...
-        
-    async def process_revalidation(self):
-        # Обработка элементов для ревалидации
-        # ...
-        
-    def mark_for_revalidation(self, entity_id, entity_type):
-        # Добавляет сущность в очередь на ревалидацию
-        # ...
-```
-
-Менеджер ревалидации работает как асинхронный фоновый процесс, который периодически (по умолчанию каждые 5 минут) проверяет наличие сущностей для ревалидации.
-
-**Взаимодействие с Redis:**
- CacheRevalidationManager хранит прямую ссылку на сервис Redis через атрибут `_redis`
- При запуске проверяется наличие соединения с Redis и при необходимости устанавливается новое
- Включена автоматическая проверка соединения перед каждой операцией ревалидации
- Система самостоятельно восстанавливает соединение при его потере
-
-**Особенности реализации:**
- Для авторов и тем используется поштучная ревалидация каждой записи
- Для шаутов и реакций используется батчевая обработка, с порогом в 10 элементов
- При достижении порога система переключается на инвалидацию коллекций вместо поштучной обработки
- Специальный флаг `all` позволяет запустить полную инвалидацию всех записей типа
-
-### 3. Автоматическая инвалидация через триггеры
-
-Модуль `triggers.py` регистрирует обработчики событий SQLAlchemy, которые автоматически отмечают сущности для ревалидации при изменении данных в базе:
-
-```python
-def events_register():
-    event.listen(Author, "after_update", mark_for_revalidation)
-    event.listen(Topic, "after_update", mark_for_revalidation)
-    # и другие...
-```
-
-Триггеры имеют следующие особенности:
- Реагируют на события вставки, обновления и удаления
- Отмечают затронутые сущности для отложенной ревалидации
- Учитывают связи между сущностями (например, при изменении темы обновляются связанные шауты)
-
-## Предварительное кеширование
-
-Модуль `precache.py` реализует предварительное кеширование часто используемых данных при старте приложения:
-
-```python
-async def precache_data():
-    # ...
-```
-
-Эта функция выполняется при запуске приложения и заполняет кеш данными, которые будут часто запрашиваться пользователями.
-
-## Примеры использования
-
-### Простое кеширование результата запроса
-
-```python
-async def get_topics_with_stats(limit=10, offset=0, by="title"):
-    # Формирование ключа кеша по конвенции
-    cache_key = f"topics:stats:limit={limit}:offset={offset}:sort={by}"
-    
-    cached_data = await get_cached_data(cache_key)
-    if cached_data:
-        return cached_data
-        
-    # Выполнение запроса к базе данных
-    result = ... # логика получения данных
-    
-    await cache_data(cache_key, result, ttl=300)
-    return result
-```
-
-### Использование обобщенной функции cached_query
-
-```python
-async def get_topics_with_stats(limit=10, offset=0, by="title"):
-    async def fetch_data(limit, offset, by):
-        # Логика получения данных
-        return result
-        
-    # Формирование ключа кеша по конвенции
-    cache_key = f"topics:stats:limit={limit}:offset={offset}:sort={by}"
-    
-    return await cached_query(
-        cache_key, 
-        fetch_data, 
-        ttl=300, 
-        limit=limit, 
-        offset=offset, 
-        by=by
-    )
-```
-
-### Точечная инвалидация кеша при изменении данных
-
-```python
-async def update_topic(topic_id, new_data):
-    # Обновление данных в базе
-    # ...
-    
-    # Точечная инвалидация кеша только для измененной темы
-    await invalidate_topics_cache(topic_id)
-    
-    return updated_topic
-```
-
-## Отладка и мониторинг
-
-Система кеширования использует логгер для отслеживания операций:
-
-```python
-logger.debug(f"Данные получены из кеша по ключу {key}")
-logger.debug(f"Удалено {len(keys)} ключей кеша с префиксом {prefix}")
-logger.error(f"Ошибка при инвалидации кеша: {e}")
-```
-
-Это позволяет отслеживать работу кеша и выявлять возможные проблемы на ранних стадиях.
-
-## Рекомендации по использованию
-
-1. **Следуйте конвенциям формирования ключей** - это критически важно для консистентности и предсказуемости кеша.
-2. **Не создавайте собственные форматы ключей** - используйте существующие шаблоны для обеспечения единообразия.
-3. **Не забывайте об инвалидации** - всегда инвалидируйте кеш при изменении данных.
-4. **Используйте точечную инвалидацию** - вместо инвалидации по префиксу для снижения нагрузки на Redis.
-5. **Устанавливайте разумные TTL** - используйте разные значения TTL в зависимости от частоты изменения данных.
-6. **Не кешируйте большие объемы данных** - кешируйте только то, что действительно необходимо для повышения производительности.
-
-## Технические детали реализации
-
- **Сериализация данных**: используется `orjson` для эффективной сериализации и десериализации данных.
- **Форматирование даты и времени**: для корректной работы с датами используется `CustomJSONEncoder`.
- **Асинхронность**: все операции кеширования выполняются асинхронно для минимального влияния на производительность API.
- **Прямое взаимодействие с Redis**: все операции выполняются через прямые вызовы `redis.execute()` с обработкой ошибок.
- **Батчевая обработка**: для массовых операций используется пороговое значение, после которого применяются оптимизированные стратегии.
-
-## Известные ограничения
-
-1. **Согласованность данных** - система не гарантирует абсолютную согласованность данных в кеше и базе данных.
-2. **Память** - необходимо следить за объемом данных в кеше, чтобы избежать проблем с памятью Redis.
-3. **Производительность Redis** - при большом количестве операций с кешем может стать узким местом.
--- a/docs/comments-pagination.md
+++ b/docs/comments-pagination.md
@@ -1,165 +0,0 @@
-# Пагинация комментариев
-
-## Обзор
-
-Реализована система пагинации комментариев по веткам, которая позволяет эффективно загружать и отображать вложенные ветки обсуждений. Основные преимущества:
-
-1. Загрузка только необходимых комментариев, а не всего дерева
-2. Снижение нагрузки на сервер и клиент
-3. Возможность эффективной навигации по большим обсуждениям
-4. Предзагрузка первых N ответов для улучшения UX
-
-## API для иерархической загрузки комментариев
-
-### GraphQL запрос `load_comments_branch`
-
-```graphql
-query LoadCommentsBranch(
-  $shout: Int!,
-  $parentId: Int,
-  $limit: Int,
-  $offset: Int,
-  $sort: ReactionSort,
-  $childrenLimit: Int,
-  $childrenOffset: Int
-) {
-  load_comments_branch(
-    shout: $shout,
-    parent_id: $parentId,
-    limit: $limit,
-    offset: $offset,
-    sort: $sort,
-    children_limit: $childrenLimit,
-    children_offset: $childrenOffset
-  ) {
-    id
-    body
-    created_at
-    created_by {
-      id
-      name
-      slug
-      pic
-    }
-    kind
-    reply_to
-    stat {
-      rating
-      comments_count
-    }
-    first_replies {
-      id
-      body
-      created_at
-      created_by {
-        id
-        name
-        slug
-        pic
-      }
-      kind
-      reply_to
-      stat {
-        rating
-        comments_count
-      }
-    }
-  }
-}
-```
-
-### Параметры запроса
-
-| Параметр | Тип | По умолчанию | Описание |
-|----------|-----|--------------|----------|
-| shout | Int! | - | ID статьи, к которой относятся комментарии |
-| parent_id | Int | null | ID родительского комментария. Если null, загружаются корневые комментарии |
-| limit | Int | 10 | Максимальное количество комментариев для загрузки |
-| offset | Int | 0 | Смещение для пагинации |
-| sort | ReactionSort | newest | Порядок сортировки: newest, oldest, like |
-| children_limit | Int | 3 | Максимальное количество дочерних комментариев для каждого родительского |
-| children_offset | Int | 0 | Смещение для пагинации дочерних комментариев |
-
-### Поля в ответе
-
-Каждый комментарий содержит следующие основные поля:
-
- `id`: ID комментария
- `body`: Текст комментария
- `created_at`: Время создания
- `created_by`: Информация об авторе
- `kind`: Тип реакции (COMMENT)
- `reply_to`: ID родительского комментария (null для корневых)
- `first_replies`: Первые N дочерних комментариев
- `stat`: Статистика комментария, включающая:
-  - `comments_count`: Количество ответов на комментарий
-  - `rating`: Рейтинг комментария
-
-## Примеры использования
-
-### Загрузка корневых комментариев с первыми ответами
-
-```javascript
-const { data } = await client.query({
-  query: LOAD_COMMENTS_BRANCH,
-  variables: {
-    shout: 222,
-    limit: 10,
-    offset: 0,
-    sort: "newest",
-    childrenLimit: 3
-  }
-});
-```
-
-### Загрузка ответов на конкретный комментарий
-
-```javascript
-const { data } = await client.query({
-  query: LOAD_COMMENTS_BRANCH,
-  variables: {
-    shout: 222,
-    parentId: 123, // ID комментария, для которого загружаем ответы
-    limit: 10,
-    offset: 0,
-    sort: "oldest" // Сортируем ответы от старых к новым
-  }
-});
-```
-
-### Пагинация дочерних комментариев
-
-Для загрузки дополнительных ответов на комментарий:
-
-```javascript
-const { data } = await client.query({
-  query: LOAD_COMMENTS_BRANCH,
-  variables: {
-    shout: 222,
-    parentId: 123,
-    limit: 10,
-    offset: 0,
-    childrenLimit: 5,
-    childrenOffset: 3 // Пропускаем первые 3 комментария (уже загруженные)
-  }
-});
-```
-
-## Рекомендации по клиентской реализации
-
-1. Для эффективной работы со сложными ветками обсуждений рекомендуется:
-
-   - Сначала загружать только корневые комментарии с первыми N ответами
-   - При наличии дополнительных ответов (когда `stat.comments_count > first_replies.length`) 
-     добавить кнопку "Показать все ответы"
-   - При нажатии на кнопку загружать дополнительные ответы с помощью запроса с указанным `parentId`
-
-2. Для сортировки:
-   - По умолчанию использовать `newest` для отображения свежих обсуждений
-   - Предусмотреть переключатель сортировки для всего дерева комментариев
-   - При изменении сортировки перезагружать данные с новым параметром `sort`
-
-3. Для улучшения производительности:
-   - Кешировать результаты запросов на клиенте
-   - Использовать оптимистичные обновления при добавлении/редактировании комментариев
-   - При необходимости загружать комментарии порциями (ленивая загрузка) 
--- a/docs/features.md
+++ b/docs/features.md
@@ -1,48 +0,0 @@
-## Просмотры публикаций
-
- Интеграция с Google Analytics для отслеживания просмотров публикаций
- Подсчет уникальных пользователей и общего количества просмотров
- Автоматическое обновление статистики при запросе данных публикации 
-
-## Мультидоменная авторизация
-
- Поддержка авторизации для разных доменов
- Автоматическое определение сервера авторизации
- Корректная обработка CORS для всех поддерживаемых доменов
-
-## Система кеширования
-
- Redis используется в качестве основного механизма кеширования
- Поддержка как синхронных, так и асинхронных функций в декораторе cache_on_arguments
- Автоматическая сериализация/десериализация данных в JSON с использованием CustomJSONEncoder
- Резервная сериализация через pickle для сложных объектов
- Генерация уникальных ключей кеша на основе сигнатуры функции и переданных аргументов
- Настраиваемое время жизни кеша (TTL)
- Возможность ручной инвалидации кеша для конкретных функций и аргументов
-
-## Webhooks
-
- Автоматическая регистрация вебхука для события user.login
- Предотвращение создания дублирующихся вебхуков
- Автоматическая очистка устаревших вебхуков
- Поддержка авторизации вебхуков через WEBHOOK_SECRET
- Обработка ошибок при операциях с вебхуками
- Динамическое определение endpoint'а на основе окружения
-
-## CORS Configuration
-
- Поддерживаемые методы: GET, POST, OPTIONS
- Настроена поддержка credentials
- Разрешенные заголовки: Authorization, Content-Type, X-Requested-With, DNT, Cache-Control
- Настроено кэширование preflight-ответов на 20 дней (1728000 секунд)
-
-## Пагинация комментариев по веткам
-
- Эффективная загрузка комментариев с учетом их иерархической структуры
- Отдельный запрос `load_comments_branch` для оптимизированной загрузки ветки комментариев
- Возможность загрузки корневых комментариев статьи с первыми ответами на них
- Гибкая пагинация как для корневых, так и для дочерних комментариев
- Использование поля `stat.comments_count` для отображения количества ответов на комментарий
- Добавление специального поля `first_replies` для хранения первых ответов на комментарий
- Поддержка различных методов сортировки (новые, старые, популярные)
- Оптимизированные SQL запросы для минимизации нагрузки на базу данных
--- a/docs/follower.md
+++ b/docs/follower.md
@@ -1,94 +0,0 @@
-# Following System
-
-## Overview
-System supports following different entity types:
- Authors
- Topics
- Communities
- Shouts (Posts)
-
-## GraphQL API
-
-### Mutations
-
-#### follow
-Follow an entity (author/topic/community/shout).
-
-**Parameters:**
- `what: String!` - Entity type (`AUTHOR`, `TOPIC`, `COMMUNITY`, `SHOUT`)
- `slug: String` - Entity slug
- `entity_id: Int` - Optional entity ID
-
-**Returns:**
-```typescript
-{
-  authors?: Author[]        // For AUTHOR type
-  topics?: Topic[]          // For TOPIC type
-  communities?: Community[] // For COMMUNITY type
-  shouts?: Shout[]          // For SHOUT type
-  error?: String            // Error message if any
-}
-```
-
-#### unfollow
-Unfollow an entity.
-
-**Parameters:** Same as `follow`
-
-**Returns:** Same as `follow`
-
-### Queries
-
-#### get_shout_followers
-Get list of users who reacted to a shout.
-
-**Parameters:**
- `slug: String` - Shout slug
- `shout_id: Int` - Optional shout ID
-
-**Returns:**
-```typescript
-Author[] // List of authors who reacted
-```
-
-## Caching System
-
-### Supported Entity Types
- Authors: `cache_author`, `get_cached_follower_authors`
- Topics: `cache_topic`, `get_cached_follower_topics`
- Communities: No cache
- Shouts: No cache
-
-### Cache Flow
-1. On follow/unfollow:
-   - Update entity in cache
-   - Update follower's following list
-2. Cache is updated before notifications
-
-## Notifications
-
- Sent when author is followed/unfollowed
- Contains:
-  - Follower info
-  - Author ID
-  - Action type ("follow"/"unfollow")
-
-## Error Handling
-
- Unauthorized access check
- Entity existence validation
- Duplicate follow prevention
- Full error logging
- Transaction safety with `local_session()`
-
-## Database Schema
-
-### Follower Tables
- `AuthorFollower`
- `TopicFollower`
- `CommunityFollower`
- `ShoutReactionsFollower`
-
-Each table contains:
- `follower` - ID of following user
- `{entity_type}` - ID of followed entity 
--- a/docs/load_shouts.md
+++ b/docs/load_shouts.md
@@ -1,80 +0,0 @@
-# Система загрузки публикаций
-
-## Особенности реализации
-
-### Базовый запрос
- Автоматически подгружает основного автора
- Добавляет основную тему публикации
- Поддерживает гибкую систему фильтрации
- Оптимизирует запросы на основе запрошенных полей
-
-### Статистика
- Подсчёт лайков/дислайков
- Количество комментариев
- Дата последней реакции
- Статистика подгружается только при запросе поля `stat`
-
-### Оптимизация производительности
- Ленивая загрузка связанных данных
- Кэширование результатов на 5 минут
- Пакетная загрузка авторов и тем
- Использование подзапросов для сложных выборок
-
-## Типы лент
-
-### Случайные топовые посты (load_shouts_random_top)
-**Преимущества:**
- Разнообразный контент
- Быстрая выборка из кэша топовых постов
- Настраиваемый размер пула для выборки
-
-**Ограничения:**
- Обновление раз в 5 минут
- Максимальный размер пула: 100 постов
- Учитываются только лайки/дислайки (без комментариев)
-
-### Неоцененные посты (load_shouts_unrated)
-**Преимущества:**
- Помогает найти новый контент
- Равномерное распределение оценок
- Случайный порядок выдачи
-
-**Ограничения:**
- Только посты с менее чем 3 реакциями
- Не учитываются комментарии
- Без сортировки по рейтингу
-
-### Закладки (load_shouts_bookmarked)
-**Преимущества:**
- Персонализированная выборка
- Быстрый доступ к сохраненному
- Поддержка всех фильтров
-
-**Ограничения:**
- Требует авторизации
- Ограничение на количество закладок
- Кэширование отключено
-
-## Важные моменты
-
-### Пагинация
- Стандартный размер страницы: 10
- Максимальный размер: 100
- Поддержка курсор-пагинации
-
-### Кэширование
- TTL: 5 минут
- Инвалидация при изменении поста
- Отдельный кэш для каждого типа сортировки
-
-### Сортировка
- По рейтингу (лайки минус дислайки)
- По количеству комментариев
- По дате последней реакции
- По дате публикации (по умолчанию)
-
-### Безопасность
- Проверка прав доступа
- Фильтрация удаленного контента
- Защита от SQL-инъекций
- Валидация входных данных
--- a/docs/rating.md
+++ b/docs/rating.md
@@ -1,82 +0,0 @@
-# Rating System
-
-## GraphQL Resolvers
-
-### Queries
-
-#### get_my_rates_shouts
-Get user's reactions (LIKE/DISLIKE) for specified posts.
-
-**Parameters:**
- `shouts: [Int!]!` - array of shout IDs
-
-**Returns:**
-```typescript
-[{
-  shout_id: Int
-  my_rate: ReactionKind // LIKE or DISLIKE
-}]
-```
-
-#### get_my_rates_comments
-Get user's reactions (LIKE/DISLIKE) for specified comments.
-
-**Parameters:**
- `comments: [Int!]!` - array of comment IDs
-
-**Returns:**
-```typescript
-[{
-  comment_id: Int
-  my_rate: ReactionKind // LIKE or DISLIKE
-}]
-```
-
-### Mutations
-
-#### rate_author
-Rate another author (karma system).
-
-**Parameters:**
- `rated_slug: String!` - author's slug
- `value: Int!` - rating value (positive/negative)
-
-## Rating Calculation
-
-### Author Rating Components
-
-#### Shouts Rating
- Calculated from LIKE/DISLIKE reactions on author's posts
- Each LIKE: +1
- Each DISLIKE: -1
- Excludes deleted reactions
- Excludes comment reactions
-
-#### Comments Rating  
- Calculated from LIKE/DISLIKE reactions on author's comments
- Each LIKE: +1
- Each DISLIKE: -1
- Only counts reactions to COMMENT type reactions
- Excludes deleted reactions
-
-#### Legacy Karma
- Based on direct author ratings via `rate_author` mutation
- Stored in `AuthorRating` table
- Each positive rating: +1
- Each negative rating: -1
-
-### Helper Functions
-
- `count_author_comments_rating()` - Calculate comment rating
- `count_author_shouts_rating()` - Calculate posts rating
- `get_author_rating_old()` - Get legacy karma rating
- `get_author_rating_shouts()` - Get posts rating (optimized)
- `get_author_rating_comments()` - Get comments rating (optimized)
- `add_author_rating_columns()` - Add rating columns to author query
-
-## Notes
-
- All ratings exclude deleted content
- Reactions are unique per user/content
- Rating calculations are optimized with SQLAlchemy
- System supports both direct author rating and content-based rating 
--- a/generate_gql_types.sh
+++ b/generate_gql_types.sh
@@ -0,0 +1 @@
+python -m gql_schema_codegen -p ./schema.graphql -t ./schema_types.py
--- a/main.py
+++ b/main.py
@@ -1,95 +1,94 @@
 import asyncio
 import os
-import sys
 from importlib import import_module
 from os.path import exists

 from ariadne import load_schema_from_path, make_executable_schema
 from ariadne.asgi import GraphQL
 from starlette.applications import Starlette
-from starlette.middleware.cors import CORSMiddleware
-from starlette.requests import Request
-from starlette.responses import JSONResponse, Response
+from starlette.middleware import Middleware
+from starlette.middleware.authentication import AuthenticationMiddleware
+from starlette.middleware.sessions import SessionMiddleware
 from starlette.routing import Route

-from cache.precache import precache_data
-from cache.revalidator import revalidation_manager
-from services.exception import ExceptionHandlerMiddleware
-from services.redis import redis
-from services.schema import create_all_tables, resolvers
-from services.search import search_service
-from services.viewed import ViewedStorage
-from services.webhook import WebhookEndpoint, create_webhook_endpoint
-from settings import DEV_SERVER_PID_FILE_NAME, MODE
+from auth.authenticate import JWTAuthenticate
+from auth.oauth import oauth_authorize, oauth_login
+from base.redis import redis
+from base.resolvers import resolvers
+from orm import init_tables
+from resolvers.upload import upload_handler
+from services.main import storages_init
+from services.notifications.notification_service import notification_service
+from services.notifications.sse import sse_subscribe_handler
+from services.stat.viewed import ViewedStorage
+
+# from services.zine.gittask import GitTask
+from settings import DEV_SERVER_PID_FILE_NAME, SENTRY_DSN, SESSION_SECRET_KEY

 import_module("resolvers")
-schema = make_executable_schema(load_schema_from_path("schema/"), resolvers)
+schema = make_executable_schema(load_schema_from_path("schema.graphql"), resolvers)
+
+middleware = [
+    Middleware(AuthenticationMiddleware, backend=JWTAuthenticate()),
+    Middleware(SessionMiddleware, secret_key=SESSION_SECRET_KEY),
+]


-async def start():
-    if MODE == "development":
-        if not exists(DEV_SERVER_PID_FILE_NAME):
-            # pid file management
-            with open(DEV_SERVER_PID_FILE_NAME, "w", encoding="utf-8") as f:
-                f.write(str(os.getpid()))
-    print(f"[main] process started in {MODE} mode")
-
-
-async def lifespan(_app):
-    try:
-        create_all_tables()
-        await asyncio.gather(
-            redis.connect(),
-            precache_data(),
-            ViewedStorage.init(),
-            create_webhook_endpoint(),
-            search_service.info(),
-            start(),
-            revalidation_manager.start(),
-        )
-        yield
-    finally:
-        tasks = [redis.disconnect(), ViewedStorage.stop(), revalidation_manager.stop()]
-        await asyncio.gather(*tasks, return_exceptions=True)
-
-
-# Создаем экземпляр GraphQL
-graphql_app = GraphQL(schema, debug=True)
-
-
-# Оборачиваем GraphQL-обработчик для лучшей обработки ошибок
-async def graphql_handler(request: Request):
-    if request.method not in ["GET", "POST"]:
-        return JSONResponse({"error": "Method Not Allowed"}, status_code=405)
+async def start_up():
+    init_tables()
+    await redis.connect()
+    await storages_init()
+    views_stat_task = asyncio.create_task(ViewedStorage().worker())
+    print(views_stat_task)
+    # git_task = asyncio.create_task(GitTask.git_task_worker())
+    # print(git_task)
+    notification_service_task = asyncio.create_task(notification_service.worker())
+    print(notification_service_task)

    try:
-        result = await graphql_app.handle_request(request)
-        if isinstance(result, Response):
-            return result
-        return JSONResponse(result)
-    except asyncio.CancelledError:
-        return JSONResponse({"error": "Request cancelled"}, status_code=499)
+        import sentry_sdk
+
+        sentry_sdk.init(SENTRY_DSN)
    except Exception as e:
-        print(f"GraphQL error: {str(e)}")
-        return JSONResponse({"error": str(e)}, status_code=500)
+        print("[sentry] init error")
+        print(e)


-# Обновляем маршрут в Starlette
+async def dev_start_up():
+    if exists(DEV_SERVER_PID_FILE_NAME):
+        await redis.connect()
+        return
+    else:
+        with open(DEV_SERVER_PID_FILE_NAME, "w", encoding="utf-8") as f:
+            f.write(str(os.getpid()))
+
+    await start_up()
+
+
+async def shutdown():
+    await redis.disconnect()
+
+
+routes = [
+    Route("/oauth/{provider}", endpoint=oauth_login),
+    Route("/oauth-authorize", endpoint=oauth_authorize),
+    Route("/upload", endpoint=upload_handler, methods=["POST"]),
+    Route("/subscribe/{user_id}", endpoint=sse_subscribe_handler),
+]
+
 app = Starlette(
-    routes=[
-        Route("/", graphql_handler, methods=["GET", "POST"]),
-        Route("/new-author", WebhookEndpoint),
-    ],
-    lifespan=lifespan,
-    debug=True,
+    on_startup=[start_up],
+    on_shutdown=[shutdown],
+    middleware=middleware,
+    routes=routes,
 )
+app.mount("/", GraphQL(schema))

-app.add_middleware(ExceptionHandlerMiddleware)
-if "dev" in sys.argv:
-    app.add_middleware(
-        CORSMiddleware,
-        allow_origins=["https://localhost:3000"],
-        allow_credentials=True,
-        allow_methods=["*"],
-        allow_headers=["*"],
-    )
+dev_app = Starlette(
+    debug=True,
+    on_startup=[dev_start_up],
+    on_shutdown=[shutdown],
+    middleware=middleware,
+    routes=routes,
+)
+dev_app.mount("/", GraphQL(schema, debug=True))
--- a/migrate.sh
+++ b/migrate.sh
@@ -0,0 +1,18 @@
+database_name="discoursio"
+
+echo "DATABASE MIGRATION STARTED"
+
+echo "Dropping database $database_name"
+dropdb $database_name --force
+if [ $? -ne 0 ]; then { echo "Failed to drop database, aborting." ; exit 1; } fi
+echo "Database $database_name dropped"
+
+echo "Creating database $database_name"
+createdb $database_name
+if [ $? -ne 0 ]; then { echo "Failed to create database, aborting." ; exit 1; } fi
+echo "Database $database_name successfully created"
+
+echo "Start migration"
+python3 server.py migrate
+if [ $? -ne 0 ]; then { echo "Migration failed, aborting." ; exit 1; } fi
+echo 'Done!'
--- a/migration/init.py
+++ b/migration/init.py
@@ -0,0 +1,279 @@
+""" cmd managed migration """
+import asyncio
+import gc
+import json
+import sys
+from datetime import datetime, timezone
+
+import bs4
+
+from migration.export import export_mdx
+from migration.tables.comments import migrate as migrateComment
+from migration.tables.comments import migrate_2stage as migrateComment_2stage
+from migration.tables.content_items import get_shout_slug
+from migration.tables.content_items import migrate as migrateShout
+
+# from migration.tables.remarks import migrate as migrateRemark
+from migration.tables.topics import migrate as migrateTopic
+from migration.tables.users import migrate as migrateUser
+from migration.tables.users import migrate_2stage as migrateUser_2stage
+from migration.tables.users import post_migrate as users_post_migrate
+from orm import init_tables
+from orm.reaction import Reaction
+
+TODAY = datetime.strftime(datetime.now(tz=timezone.utc), "%Y%m%d")
+OLD_DATE = "2016-03-05 22:22:00.350000"
+
+
+async def users_handle(storage):
+    """migrating users first"""
+    counter = 0
+    id_map = {}
+    print("[migration] migrating %d users" % (len(storage["users"]["data"])))
+    for entry in storage["users"]["data"]:
+        oid = entry["_id"]
+        user = migrateUser(entry)
+        storage["users"]["by_oid"][oid] = user  # full
+        del user["password"]
+        del user["emailConfirmed"]
+        del user["username"]
+        del user["email"]
+        storage["users"]["by_slug"][user["slug"]] = user  # public
+        id_map[user["oid"]] = user["slug"]
+        counter += 1
+    ce = 0
+    for entry in storage["users"]["data"]:
+        ce += migrateUser_2stage(entry, id_map)
+    users_post_migrate()
+
+
+async def topics_handle(storage):
+    """topics from categories and tags"""
+    counter = 0
+    for t in storage["topics"]["tags"] + storage["topics"]["cats"]:
+        if t["slug"] in storage["replacements"]:
+            t["slug"] = storage["replacements"][t["slug"]]
+            topic = migrateTopic(t)
+            storage["topics"]["by_oid"][t["_id"]] = topic
+            storage["topics"]["by_slug"][t["slug"]] = topic
+            counter += 1
+        else:
+            print("[migration] topic " + t["slug"] + " ignored")
+    for oldslug, newslug in storage["replacements"].items():
+        if oldslug != newslug and oldslug in storage["topics"]["by_slug"]:
+            oid = storage["topics"]["by_slug"][oldslug]["_id"]
+            del storage["topics"]["by_slug"][oldslug]
+            storage["topics"]["by_oid"][oid] = storage["topics"]["by_slug"][newslug]
+    print("[migration] " + str(counter) + " topics migrated")
+    print("[migration] " + str(len(storage["topics"]["by_oid"].values())) + " topics by oid")
+    print("[migration] " + str(len(storage["topics"]["by_slug"].values())) + " topics by slug")
+
+
+async def shouts_handle(storage, args):
+    """migrating content items one by one"""
+    counter = 0
+    discours_author = 0
+    anonymous_author = 0
+    pub_counter = 0
+    ignored = 0
+    topics_dataset_bodies = []
+    topics_dataset_tlist = []
+    for entry in storage["shouts"]["data"]:
+        gc.collect()
+        # slug
+        slug = get_shout_slug(entry)
+
+        # single slug mode
+        if "-" in args and slug not in args:
+            continue
+
+        # migrate
+        shout_dict = await migrateShout(entry, storage)
+        if shout_dict:
+            storage["shouts"]["by_oid"][entry["_id"]] = shout_dict
+            storage["shouts"]["by_slug"][shout_dict["slug"]] = shout_dict
+            # shouts.topics
+            if not shout_dict["topics"]:
+                print("[migration] no topics!")
+
+            # with author
+            author = shout_dict["authors"][0]
+            if author["slug"] == "discours":
+                discours_author += 1
+            if author["slug"] == "anonymous":
+                anonymous_author += 1
+            # print('[migration] ' + shout['slug'] + ' with author ' + author)
+
+            if entry.get("published"):
+                if "mdx" in args:
+                    export_mdx(shout_dict)
+                pub_counter += 1
+
+            # print main counter
+            counter += 1
+            print(
+                "[migration] shouts_handle %d: %s @%s"
+                % ((counter + 1), shout_dict["slug"], author["slug"])
+            )
+
+            b = bs4.BeautifulSoup(shout_dict["body"], "html.parser")
+            texts = [shout_dict["title"].lower().replace(r"[^а-яА-Яa-zA-Z]", "")]
+            texts = texts + b.findAll(text=True)
+            topics_dataset_bodies.append(" ".join([x.strip().lower() for x in texts]))
+            topics_dataset_tlist.append(shout_dict["topics"])
+        else:
+            ignored += 1
+
+    # np.savetxt('topics_dataset.csv', (topics_dataset_bodies, topics_dataset_tlist), delimiter=',
+    # ', fmt='%s')
+
+    print("[migration] " + str(counter) + " content items were migrated")
+    print("[migration] " + str(pub_counter) + " have been published")
+    print("[migration] " + str(discours_author) + " authored by @discours")
+    print("[migration] " + str(anonymous_author) + " authored by @anonymous")
+
+
+# async def remarks_handle(storage):
+#     print("[migration] comments")
+#     c = 0
+#     for entry_remark in storage["remarks"]["data"]:
+#         remark = await migrateRemark(entry_remark, storage)
+#         c += 1
+#     print("[migration] " + str(c) + " remarks migrated")
+
+
+async def comments_handle(storage):
+    print("[migration] comments")
+    id_map = {}
+    ignored_counter = 0
+    missed_shouts = {}
+    for oldcomment in storage["reactions"]["data"]:
+        if not oldcomment.get("deleted"):
+            reaction = await migrateComment(oldcomment, storage)
+            if isinstance(reaction, str):
+                missed_shouts[reaction] = oldcomment
+            elif isinstance(reaction, Reaction):
+                reaction = reaction.dict()
+                rid = reaction["id"]
+                oid = reaction["oid"]
+                id_map[oid] = rid
+            else:
+                ignored_counter += 1
+
+    for reaction in storage["reactions"]["data"]:
+        migrateComment_2stage(reaction, id_map)
+    print("[migration] " + str(len(id_map)) + " comments migrated")
+    print("[migration] " + str(ignored_counter) + " comments ignored")
+    print("[migration] " + str(len(missed_shouts.keys())) + " commented shouts missed")
+    missed_counter = 0
+    for missed in missed_shouts.values():
+        missed_counter += len(missed)
+    print("[migration] " + str(missed_counter) + " comments dropped")
+
+
+async def all_handle(storage, args):
+    print("[migration] handle everything")
+    await users_handle(storage)
+    await topics_handle(storage)
+    print("[migration] users and topics are migrated")
+    await shouts_handle(storage, args)
+    # print("[migration] remarks...")
+    # await remarks_handle(storage)
+    print("[migration] migrating comments")
+    await comments_handle(storage)
+    # export_email_subscriptions()
+    print("[migration] done!")
+
+
+def data_load():
+    storage = {
+        "content_items": {
+            "by_oid": {},
+            "by_slug": {},
+        },
+        "shouts": {"by_oid": {}, "by_slug": {}, "data": []},
+        "reactions": {"by_oid": {}, "by_slug": {}, "by_content": {}, "data": []},
+        "topics": {
+            "by_oid": {},
+            "by_slug": {},
+            "cats": [],
+            "tags": [],
+        },
+        "remarks": {"data": []},
+        "users": {"by_oid": {}, "by_slug": {}, "data": []},
+        "replacements": json.loads(open("migration/tables/replacements.json").read()),
+    }
+    try:
+        users_data = json.loads(open("migration/data/users.json").read())
+        print("[migration.load] " + str(len(users_data)) + " users ")
+        tags_data = json.loads(open("migration/data/tags.json").read())
+        storage["topics"]["tags"] = tags_data
+        print("[migration.load] " + str(len(tags_data)) + " tags ")
+        cats_data = json.loads(open("migration/data/content_item_categories.json").read())
+        storage["topics"]["cats"] = cats_data
+        print("[migration.load] " + str(len(cats_data)) + " cats ")
+        comments_data = json.loads(open("migration/data/comments.json").read())
+        storage["reactions"]["data"] = comments_data
+        print("[migration.load] " + str(len(comments_data)) + " comments ")
+        content_data = json.loads(open("migration/data/content_items.json").read())
+        storage["shouts"]["data"] = content_data
+        print("[migration.load] " + str(len(content_data)) + " content items ")
+
+        remarks_data = json.loads(open("migration/data/remarks.json").read())
+        storage["remarks"]["data"] = remarks_data
+        print("[migration.load] " + str(len(remarks_data)) + " remarks data ")
+
+        # fill out storage
+        for x in users_data:
+            storage["users"]["by_oid"][x["_id"]] = x
+            # storage['users']['by_slug'][x['slug']] = x
+        # no user.slug yet
+        print("[migration.load] " + str(len(storage["users"]["by_oid"].keys())) + " users by oid")
+        for x in tags_data:
+            storage["topics"]["by_oid"][x["_id"]] = x
+            storage["topics"]["by_slug"][x["slug"]] = x
+        for x in cats_data:
+            storage["topics"]["by_oid"][x["_id"]] = x
+            storage["topics"]["by_slug"][x["slug"]] = x
+        print(
+            "[migration.load] " + str(len(storage["topics"]["by_slug"].keys())) + " topics by slug"
+        )
+        for item in content_data:
+            slug = get_shout_slug(item)
+            storage["content_items"]["by_slug"][slug] = item
+            storage["content_items"]["by_oid"][item["_id"]] = item
+        print("[migration.load] " + str(len(content_data)) + " content items")
+        for x in comments_data:
+            storage["reactions"]["by_oid"][x["_id"]] = x
+            cid = x["contentItem"]
+            storage["reactions"]["by_content"][cid] = x
+            ci = storage["content_items"]["by_oid"].get(cid, {})
+            if "slug" in ci:
+                storage["reactions"]["by_slug"][ci["slug"]] = x
+        print(
+            "[migration.load] "
+            + str(len(storage["reactions"]["by_content"].keys()))
+            + " with comments"
+        )
+        storage["users"]["data"] = users_data
+        storage["topics"]["tags"] = tags_data
+        storage["topics"]["cats"] = cats_data
+        storage["shouts"]["data"] = content_data
+        storage["reactions"]["data"] = comments_data
+    except Exception as e:
+        raise e
+    return storage
+
+
+async def handling_migration():
+    init_tables()
+    await all_handle(data_load(), sys.argv)
+
+
+def process():
+    loop = asyncio.get_event_loop()
+    loop.run_until_complete(handling_migration())
+
+
+if __name__ == "__main__":
+    process()
--- a/migration/bson2json.py
+++ b/migration/bson2json.py
@@ -0,0 +1,33 @@
+import gc
+import json
+import os
+
+import bson
+
+from .utils import DateTimeEncoder
+
+
+def json_tables():
+    print("[migration] unpack dump/discours/*.bson to migration/data/*.json")
+    data = {
+        "content_items": [],
+        "content_item_categories": [],
+        "tags": [],
+        "email_subscriptions": [],
+        "users": [],
+        "comments": [],
+        "remarks": [],
+    }
+    for table in data.keys():
+        print("[migration] bson2json for " + table)
+        gc.collect()
+        lc = []
+        bs = open("dump/discours/" + table + ".bson", "rb").read()
+        base = 0
+        while base < len(bs):
+            base, d = bson.decode_document(bs, base)
+            lc.append(d)
+        data[table] = lc
+        open(os.getcwd() + "/migration/data/" + table + ".json", "w").write(
+            json.dumps(lc, cls=DateTimeEncoder)
+        )
--- a/migration/data/.gitkeep
+++ b/migration/data/.gitkeep
--- a/migration/export.py
+++ b/migration/export.py
@@ -0,0 +1,137 @@
+import json
+import os
+from datetime import datetime, timezone
+
+import frontmatter
+
+from .extract import extract_html, extract_media
+from .utils import DateTimeEncoder
+
+OLD_DATE = "2016-03-05 22:22:00.350000"
+EXPORT_DEST = "../discoursio-web/data/"
+parentDir = "/".join(os.getcwd().split("/")[:-1])
+contentDir = parentDir + "/discoursio-web/content/"
+ts = datetime.now(tz=timezone.utc)
+
+
+def get_metadata(r):
+    authors = []
+    for a in r["authors"]:
+        authors.append(
+            {  # a short version for public listings
+                "slug": a.slug or "discours",
+                "name": a.name or "Дискурс",
+                "userpic": a.userpic or "https://discours.io/static/img/discours.png",
+            }
+        )
+    metadata = {}
+    metadata["title"] = r.get("title", "").replace("{", "(").replace("}", ")")
+    metadata["authors"] = authors
+    metadata["createdAt"] = r.get("createdAt", ts)
+    metadata["layout"] = r["layout"]
+    metadata["topics"] = [topic for topic in r["topics"]]
+    metadata["topics"].sort()
+    if r.get("cover", False):
+        metadata["cover"] = r.get("cover")
+    return metadata
+
+
+def export_mdx(r):
+    # print('[export] mdx %s' % r['slug'])
+    content = ""
+    metadata = get_metadata(r)
+    content = frontmatter.dumps(frontmatter.Post(r["body"], **metadata))
+    ext = "mdx"
+    filepath = contentDir + r["slug"]
+    bc = bytes(content, "utf-8").decode("utf-8", "ignore")
+    open(filepath + "." + ext, "w").write(bc)
+
+
+def export_body(shout, storage):
+    entry = storage["content_items"]["by_oid"][shout["oid"]]
+    if entry:
+        body = extract_html(entry)
+        media = extract_media(entry)
+        shout["body"] = body  # prepare_html_body(entry)  # prepare_md_body(entry)
+        shout["media"] = media
+        export_mdx(shout)
+        print("[export] html for %s" % shout["slug"])
+        open(contentDir + shout["slug"] + ".html", "w").write(body)
+    else:
+        raise Exception("no content_items entry found")
+
+
+def export_slug(slug, storage):
+    shout = storage["shouts"]["by_slug"][slug]
+    shout = storage["shouts"]["by_slug"].get(slug)
+    assert shout, "[export] no shout found by slug: %s " % slug
+    author = shout["authors"][0]
+    assert author, "[export] no author error"
+    export_body(shout, storage)
+
+
+def export_email_subscriptions():
+    email_subscriptions_data = json.loads(open("migration/data/email_subscriptions.json").read())
+    for data in email_subscriptions_data:
+        # TODO: migrate to mailgun list manually
+        # migrate_email_subscription(data)
+        pass
+    print("[migration] " + str(len(email_subscriptions_data)) + " email subscriptions exported")
+
+
+def export_shouts(storage):
+    # update what was just migrated or load json again
+    if len(storage["users"]["by_slugs"].keys()) == 0:
+        storage["users"]["by_slugs"] = json.loads(open(EXPORT_DEST + "authors.json").read())
+        print("[migration] " + str(len(storage["users"]["by_slugs"].keys())) + " exported authors ")
+    if len(storage["shouts"]["by_slugs"].keys()) == 0:
+        storage["shouts"]["by_slugs"] = json.loads(open(EXPORT_DEST + "articles.json").read())
+        print(
+            "[migration] " + str(len(storage["shouts"]["by_slugs"].keys())) + " exported articles "
+        )
+    for slug in storage["shouts"]["by_slugs"].keys():
+        export_slug(slug, storage)
+
+
+def export_json(export_articles={}, export_authors={}, export_topics={}, export_comments={}):
+    open(EXPORT_DEST + "authors.json", "w").write(
+        json.dumps(
+            export_authors,
+            cls=DateTimeEncoder,
+            indent=4,
+            sort_keys=True,
+            ensure_ascii=False,
+        )
+    )
+    print("[migration] " + str(len(export_authors.items())) + " authors exported")
+    open(EXPORT_DEST + "topics.json", "w").write(
+        json.dumps(
+            export_topics,
+            cls=DateTimeEncoder,
+            indent=4,
+            sort_keys=True,
+            ensure_ascii=False,
+        )
+    )
+    print("[migration] " + str(len(export_topics.keys())) + " topics exported")
+
+    open(EXPORT_DEST + "articles.json", "w").write(
+        json.dumps(
+            export_articles,
+            cls=DateTimeEncoder,
+            indent=4,
+            sort_keys=True,
+            ensure_ascii=False,
+        )
+    )
+    print("[migration] " + str(len(export_articles.items())) + " articles exported")
+    open(EXPORT_DEST + "comments.json", "w").write(
+        json.dumps(
+            export_comments,
+            cls=DateTimeEncoder,
+            indent=4,
+            sort_keys=True,
+            ensure_ascii=False,
+        )
+    )
+    print("[migration] " + str(len(export_comments.items())) + " exported articles with comments")
--- a/migration/extract.py
+++ b/migration/extract.py
@@ -0,0 +1,276 @@
+import os
+import re
+
+from bs4 import BeautifulSoup
+
+TOOLTIP_REGEX = r"(\/\/\/(.+)\/\/\/)"
+contentDir = os.path.join(
+    os.path.dirname(os.path.realpath(__file__)), "..", "..", "discoursio-web", "content"
+)
+
+cdn = "https://images.discours.io"
+
+
+def replace_tooltips(body):
+    # change if you prefer regexp
+    newbody = body
+    matches = list(re.finditer(TOOLTIP_REGEX, body, re.IGNORECASE | re.MULTILINE))[1:]
+    for match in matches:
+        newbody = body.replace(
+            match.group(1), '<Tooltip text="' + match.group(2) + '" />'
+        )  # NOTE: doesn't work
+    if len(matches) > 0:
+        print("[extract] found %d tooltips" % len(matches))
+    return newbody
+
+
+# def extract_footnotes(body, shout_dict):
+#     parts = body.split("&&&")
+#     lll = len(parts)
+#     newparts = list(parts)
+#     placed = False
+#     if lll & 1:
+#         if lll > 1:
+#             i = 1
+#             print("[extract] found %d footnotes in body" % (lll - 1))
+#             for part in parts[1:]:
+#                 if i & 1:
+#                     placed = True
+#                     if 'a class="footnote-url" href=' in part:
+#                         print("[extract] footnote: " + part)
+#                         fn = 'a class="footnote-url" href="'
+#                         exxtracted_link = part.split(fn, 1)[1].split('"', 1)[0]
+#                         extracted_body = part.split(fn, 1)[1].split(">", 1)[1].split("</a>", 1)[0]
+#                         print("[extract] footnote link: " + extracted_link)
+#                         with local_session() as session:
+#                             Reaction.create(
+#                                 {
+#                                     "shout": shout_dict["id"],
+#                                     "kind": ReactionKind.FOOTNOTE,
+#                                     "body": extracted_body,
+#                                     "range": str(body.index(fn + link) - len("<"))
+#                                     + ":"
+#                                     + str(body.index(extracted_body) + len("</a>")),
+#                                 }
+#                             )
+#                         newparts[i] = "<a href='#'>ℹ️</a>"
+#                 else:
+#                     newparts[i] = part
+#                 i += 1
+#     return ("".join(newparts), placed)
+
+
+# def place_tooltips(body):
+#     parts = body.split("&&&")
+#     lll = len(parts)
+#     newparts = list(parts)
+#     placed = False
+#     if lll & 1:
+#         if lll > 1:
+#             i = 1
+#             print("[extract] found %d tooltips" % (lll - 1))
+#             for part in parts[1:]:
+#                 if i & 1:
+#                     placed = True
+#                     if 'a class="footnote-url" href=' in part:
+#                         print("[extract] footnote: " + part)
+#                         fn = 'a class="footnote-url" href="'
+#                         link = part.split(fn, 1)[1].split('"', 1)[0]
+#                         extracted_part = part.split(fn, 1)[0] + " " + part.split("/", 1)[-1]
+#                         newparts[i] = (
+#                             "<Tooltip"
+#                             + (' link="' + link + '" ' if link else "")
+#                             + ">"
+#                             + extracted_part
+#                             + "</Tooltip>"
+#                         )
+#                     else:
+#                         newparts[i] = "<Tooltip>%s</Tooltip>" % part
+#                         # print('[extract] ' + newparts[i])
+#                 else:
+#                     # print('[extract] ' + part[:10] + '..')
+#                     newparts[i] = part
+#                 i += 1
+#     return ("".join(newparts), placed)
+
+
+IMG_REGEX = (
+    r"\!\[(.*?)\]\((data\:image\/(png|jpeg|jpg);base64\,((?:[A-Za-z\d+\/]{4})*(?:[A-Za-z\d+\/]{3}="
+)
+IMG_REGEX += r"|[A-Za-z\d+\/]{2}==)))\)"
+
+parentDir = "/".join(os.getcwd().split("/")[:-1])
+public = parentDir + "/discoursio-web/public"
+cache = {}
+
+
+# def reextract_images(body, oid):
+#     # change if you prefer regexp
+#     matches = list(re.finditer(IMG_REGEX, body, re.IGNORECASE | re.MULTILINE))[1:]
+#     i = 0
+#     for match in matches:
+#         print("[extract] image " + match.group(1))
+#         ext = match.group(3)
+#         name = oid + str(i)
+#         link = public + "/upload/image-" + name + "." + ext
+#         img = match.group(4)
+#         title = match.group(1)  # NOTE: this is not the title
+#         if img not in cache:
+#             content = base64.b64decode(img + "==")
+#             print(str(len(img)) + " image bytes been written")
+#             open("../" + link, "wb").write(content)
+#             cache[img] = name
+#             i += 1
+#         else:
+#             print("[extract] image cached " + cache[img])
+#         body.replace(
+#             str(match), "![" + title + "](" + cdn + link + ")"
+#         )  # WARNING: this does not work
+#     return body
+
+
+IMAGES = {
+    "data:image/png": "png",
+    "data:image/jpg": "jpg",
+    "data:image/jpeg": "jpg",
+}
+
+b64 = ";base64,"
+
+di = "data:image"
+
+
+def extract_media(entry):
+    """normalized media extraction method"""
+    # media [ { title pic url body } ]}
+    kind = entry.get("type")
+    if not kind:
+        print(entry)
+        raise Exception("shout no layout")
+    media = []
+    for m in entry.get("media") or []:
+        # title
+        title = m.get("title", "").replace("\n", " ").replace("&nbsp;", " ")
+        artist = m.get("performer") or m.get("artist")
+        if artist:
+            title = artist + " - " + title
+
+        # pic
+        url = m.get("fileUrl") or m.get("url", "")
+        pic = ""
+        if m.get("thumborId"):
+            pic = cdn + "/unsafe/" + m["thumborId"]
+
+        # url
+        if not url:
+            if kind == "Image":
+                url = pic
+            elif "youtubeId" in m:
+                url = "https://youtube.com/?watch=" + m["youtubeId"]
+            elif "vimeoId" in m:
+                url = "https://vimeo.com/" + m["vimeoId"]
+        # body
+        body = m.get("body") or m.get("literatureBody") or ""
+        media.append({"url": url, "pic": pic, "title": title, "body": body})
+    return media
+
+
+def prepare_html_body(entry):
+    # body modifications
+    body = ""
+    kind = entry.get("type")
+    addon = ""
+    if kind == "Video":
+        addon = ""
+        for m in entry.get("media") or []:
+            if "youtubeId" in m:
+                addon += '<iframe width="420" height="345" src="http://www.youtube.com/embed/'
+                addon += m["youtubeId"]
+                addon += '?autoplay=1" frameborder="0" allowfullscreen></iframe>\n'
+            elif "vimeoId" in m:
+                addon += '<iframe src="https://player.vimeo.com/video/'
+                addon += m["vimeoId"]
+                addon += ' width="420" height="345" frameborder="0" allow="autoplay; fullscreen"'
+                addon += " allowfullscreen></iframe>"
+            else:
+                print("[extract] media is not supported")
+                print(m)
+        body += addon
+
+    elif kind == "Music":
+        addon = ""
+        for m in entry.get("media") or []:
+            artist = m.get("performer")
+            trackname = ""
+            if artist:
+                trackname += artist + " - "
+            if "title" in m:
+                trackname += m.get("title", "")
+            addon += "<figure><figcaption>"
+            addon += trackname
+            addon += '</figcaption><audio controls src="'
+            addon += m.get("fileUrl", "")
+            addon += '"></audio></figure>'
+        body += addon
+
+    body = extract_html(entry)
+    # if body_orig: body += extract_md(html2text(body_orig), entry['_id'])
+    return body
+
+
+def cleanup_html(body: str) -> str:
+    new_body = body
+    regex_remove = [
+        r"style=\"width:\s*\d+px;height:\s*\d+px;\"",
+        r"style=\"width:\s*\d+px;\"",
+        r"style=\"color: #000000;\"",
+        r"style=\"float: none;\"",
+        r"style=\"background: white;\"",
+        r"class=\"Apple-interchange-newline\"",
+        r"class=\"MsoNormalCxSpMiddle\"",
+        r"class=\"MsoNormal\"",
+        r"lang=\"EN-US\"",
+        r"id=\"docs-internal-guid-[\w-]+\"",
+        r"<p>\s*</p>",
+        r"<span></span>",
+        r"<i>\s*</i>",
+        r"<b>\s*</b>",
+        r"<h1>\s*</h1>",
+        r"<h2>\s*</h2>",
+        r"<h3>\s*</h3>",
+        r"<h4>\s*</h4>",
+        r"<div>\s*</div>",
+    ]
+    regex_replace = {r"<br>\s*</p>": "</p>"}
+    changed = True
+    while changed:
+        # we need several iterations to clean nested tags this way
+        changed = False
+        new_body_iteration = new_body
+        for regex in regex_remove:
+            new_body = re.sub(regex, "", new_body)
+        for regex, replace in regex_replace.items():
+            new_body = re.sub(regex, replace, new_body)
+        if new_body_iteration != new_body:
+            changed = True
+    return new_body
+
+
+def extract_html(entry, shout_id=None, cleanup=False):
+    body_orig = (entry.get("body") or "").replace(r"\(", "(").replace(r"\)", ")")
+    if cleanup:
+        # we do that before bs parsing to catch the invalid html
+        body_clean = cleanup_html(body_orig)
+        if body_clean != body_orig:
+            print(f"[migration] html cleaned for slug {entry.get('slug', None)}")
+        body_orig = body_clean
+    # if shout_id:
+    #     extract_footnotes(body_orig, shout_id)
+    body_html = str(BeautifulSoup(body_orig, features="html.parser"))
+    if cleanup:
+        # we do that after bs parsing because it can add dummy tags
+        body_clean_html = cleanup_html(body_html)
+        if body_clean_html != body_html:
+            print(f"[migration] html cleaned after bs4 for slug {entry.get('slug', None)}")
+        body_html = body_clean_html
+    return body_html
--- a/migration/html2text/init.py
+++ b/migration/html2text/init.py
--- a/migration/html2text/main.py
+++ b/migration/html2text/main.py
@@ -0,0 +1,3 @@
+from .cli import main
+
+main()
--- a/migration/html2text/cli.py
+++ b/migration/html2text/cli.py
@@ -0,0 +1,318 @@
+import argparse
+import sys
+
+from . import HTML2Text, __version__, config
+
+
+# noinspection DuplicatedCode
+def main() -> None:
+    baseurl = ""
+
+    class bcolors:
+        HEADER = "\033[95m"
+        OKBLUE = "\033[94m"
+        OKGREEN = "\033[92m"
+        WARNING = "\033[93m"
+        FAIL = "\033[91m"
+        ENDC = "\033[0m"
+        BOLD = "\033[1m"
+        UNDERLINE = "\033[4m"
+
+    p = argparse.ArgumentParser()
+    p.add_argument(
+        "--default-image-alt",
+        dest="default_image_alt",
+        default=config.DEFAULT_IMAGE_ALT,
+        help="The default alt string for images with missing ones",
+    )
+    p.add_argument(
+        "--pad-tables",
+        dest="pad_tables",
+        action="store_true",
+        default=config.PAD_TABLES,
+        help="pad the cells to equal column width in tables",
+    )
+    p.add_argument(
+        "--no-wrap-links",
+        dest="wrap_links",
+        action="store_false",
+        default=config.WRAP_LINKS,
+        help="don't wrap links during conversion",
+    )
+    p.add_argument(
+        "--wrap-list-items",
+        dest="wrap_list_items",
+        action="store_true",
+        default=config.WRAP_LIST_ITEMS,
+        help="wrap list items during conversion",
+    )
+    p.add_argument(
+        "--wrap-tables",
+        dest="wrap_tables",
+        action="store_true",
+        default=config.WRAP_TABLES,
+        help="wrap tables",
+    )
+    p.add_argument(
+        "--ignore-emphasis",
+        dest="ignore_emphasis",
+        action="store_true",
+        default=config.IGNORE_EMPHASIS,
+        help="don't include any formatting for emphasis",
+    )
+    p.add_argument(
+        "--reference-links",
+        dest="inline_links",
+        action="store_false",
+        default=config.INLINE_LINKS,
+        help="use reference style links instead of inline links",
+    )
+    p.add_argument(
+        "--ignore-links",
+        dest="ignore_links",
+        action="store_true",
+        default=config.IGNORE_ANCHORS,
+        help="don't include any formatting for links",
+    )
+    p.add_argument(
+        "--ignore-mailto-links",
+        action="store_true",
+        dest="ignore_mailto_links",
+        default=config.IGNORE_MAILTO_LINKS,
+        help="don't include mailto: links",
+    )
+    p.add_argument(
+        "--protect-links",
+        dest="protect_links",
+        action="store_true",
+        default=config.PROTECT_LINKS,
+        help="protect links from line breaks surrounding them with angle brackets",
+    )
+    p.add_argument(
+        "--ignore-images",
+        dest="ignore_images",
+        action="store_true",
+        default=config.IGNORE_IMAGES,
+        help="don't include any formatting for images",
+    )
+    p.add_argument(
+        "--images-as-html",
+        dest="images_as_html",
+        action="store_true",
+        default=config.IMAGES_AS_HTML,
+        help=(
+            "Always write image tags as raw html; preserves `height`, `width` and "
+            "`alt` if possible."
+        ),
+    )
+    p.add_argument(
+        "--images-to-alt",
+        dest="images_to_alt",
+        action="store_true",
+        default=config.IMAGES_TO_ALT,
+        help="Discard image data, only keep alt text",
+    )
+    p.add_argument(
+        "--images-with-size",
+        dest="images_with_size",
+        action="store_true",
+        default=config.IMAGES_WITH_SIZE,
+        help=("Write image tags with height and width attrs as raw html to retain " "dimensions"),
+    )
+    p.add_argument(
+        "-g",
+        "--google-doc",
+        action="store_true",
+        dest="google_doc",
+        default=False,
+        help="convert an html-exported Google Document",
+    )
+    p.add_argument(
+        "-d",
+        "--dash-unordered-list",
+        action="store_true",
+        dest="ul_style_dash",
+        default=False,
+        help="use a dash rather than a star for unordered list items",
+    )
+    p.add_argument(
+        "-e",
+        "--asterisk-emphasis",
+        action="store_true",
+        dest="em_style_asterisk",
+        default=False,
+        help="use an asterisk rather than an underscore for emphasized text",
+    )
+    p.add_argument(
+        "-b",
+        "--body-width",
+        dest="body_width",
+        type=int,
+        default=config.BODY_WIDTH,
+        help="number of characters per output line, 0 for no wrap",
+    )
+    p.add_argument(
+        "-i",
+        "--google-list-indent",
+        dest="list_indent",
+        type=int,
+        default=config.GOOGLE_LIST_INDENT,
+        help="number of pixels Google indents nested lists",
+    )
+    p.add_argument(
+        "-s",
+        "--hide-strikethrough",
+        action="store_true",
+        dest="hide_strikethrough",
+        default=False,
+        help="hide strike-through text. only relevant when -g is " "specified as well",
+    )
+    p.add_argument(
+        "--escape-all",
+        action="store_true",
+        dest="escape_snob",
+        default=False,
+        help=(
+            "Escape all special characters.  Output is less readable, but avoids "
+            "corner case formatting issues."
+        ),
+    )
+    p.add_argument(
+        "--bypass-tables",
+        action="store_true",
+        dest="bypass_tables",
+        default=config.BYPASS_TABLES,
+        help="Format tables in HTML rather than Markdown syntax.",
+    )
+    p.add_argument(
+        "--ignore-tables",
+        action="store_true",
+        dest="ignore_tables",
+        default=config.IGNORE_TABLES,
+        help="Ignore table-related tags (table, th, td, tr) " "while keeping rows.",
+    )
+    p.add_argument(
+        "--single-line-break",
+        action="store_true",
+        dest="single_line_break",
+        default=config.SINGLE_LINE_BREAK,
+        help=(
+            "Use a single line break after a block element rather than two line "
+            "breaks. NOTE: Requires --body-width=0"
+        ),
+    )
+    p.add_argument(
+        "--unicode-snob",
+        action="store_true",
+        dest="unicode_snob",
+        default=config.UNICODE_SNOB,
+        help="Use unicode throughout document",
+    )
+    p.add_argument(
+        "--no-automatic-links",
+        action="store_false",
+        dest="use_automatic_links",
+        default=config.USE_AUTOMATIC_LINKS,
+        help="Do not use automatic links wherever applicable",
+    )
+    p.add_argument(
+        "--no-skip-internal-links",
+        action="store_false",
+        dest="skip_internal_links",
+        default=config.SKIP_INTERNAL_LINKS,
+        help="Do not skip internal links",
+    )
+    p.add_argument(
+        "--links-after-para",
+        action="store_true",
+        dest="links_each_paragraph",
+        default=config.LINKS_EACH_PARAGRAPH,
+        help="Put links after each paragraph instead of document",
+    )
+    p.add_argument(
+        "--mark-code",
+        action="store_true",
+        dest="mark_code",
+        default=config.MARK_CODE,
+        help="Mark program code blocks with [code]...[/code]",
+    )
+    p.add_argument(
+        "--decode-errors",
+        dest="decode_errors",
+        default=config.DECODE_ERRORS,
+        help=(
+            "What to do in case of decode errors.'ignore', 'strict' and 'replace' are "
+            "acceptable values"
+        ),
+    )
+    p.add_argument(
+        "--open-quote",
+        dest="open_quote",
+        default=config.OPEN_QUOTE,
+        help="The character used to open quotes",
+    )
+    p.add_argument(
+        "--close-quote",
+        dest="close_quote",
+        default=config.CLOSE_QUOTE,
+        help="The character used to close quotes",
+    )
+    p.add_argument("--version", action="version", version=".".join(map(str, __version__)))
+    p.add_argument("filename", nargs="?")
+    p.add_argument("encoding", nargs="?", default="utf-8")
+    args = p.parse_args()
+
+    if args.filename and args.filename != "-":
+        with open(args.filename, "rb") as fp:
+            data = fp.read()
+    else:
+        data = sys.stdin.buffer.read()
+
+    try:
+        html = data.decode(args.encoding, args.decode_errors)
+    except UnicodeDecodeError as err:
+        warning = bcolors.WARNING + "Warning:" + bcolors.ENDC
+        warning += " Use the " + bcolors.OKGREEN
+        warning += "--decode-errors=ignore" + bcolors.ENDC + " flag."
+        print(warning)
+        raise err
+
+    h = HTML2Text(baseurl=baseurl)
+    # handle options
+    if args.ul_style_dash:
+        h.ul_item_mark = "-"
+    if args.em_style_asterisk:
+        h.emphasis_mark = "*"
+        h.strong_mark = "__"
+
+    h.body_width = args.body_width
+    h.google_list_indent = args.list_indent
+    h.ignore_emphasis = args.ignore_emphasis
+    h.ignore_links = args.ignore_links
+    h.ignore_mailto_links = args.ignore_mailto_links
+    h.protect_links = args.protect_links
+    h.ignore_images = args.ignore_images
+    h.images_as_html = args.images_as_html
+    h.images_to_alt = args.images_to_alt
+    h.images_with_size = args.images_with_size
+    h.google_doc = args.google_doc
+    h.hide_strikethrough = args.hide_strikethrough
+    h.escape_snob = args.escape_snob
+    h.bypass_tables = args.bypass_tables
+    h.ignore_tables = args.ignore_tables
+    h.single_line_break = args.single_line_break
+    h.inline_links = args.inline_links
+    h.unicode_snob = args.unicode_snob
+    h.use_automatic_links = args.use_automatic_links
+    h.skip_internal_links = args.skip_internal_links
+    h.links_each_paragraph = args.links_each_paragraph
+    h.mark_code = args.mark_code
+    h.wrap_links = args.wrap_links
+    h.wrap_list_items = args.wrap_list_items
+    h.wrap_tables = args.wrap_tables
+    h.pad_tables = args.pad_tables
+    h.default_image_alt = args.default_image_alt
+    h.open_quote = args.open_quote
+    h.close_quote = args.close_quote
+
+    sys.stdout.write(h.handle(html))
--- a/migration/html2text/config.py
+++ b/migration/html2text/config.py
@@ -0,0 +1,164 @@
+import re
+
+# Use Unicode characters instead of their ascii pseudo-replacements
+UNICODE_SNOB = True
+
+# Marker to use for marking tables for padding post processing
+TABLE_MARKER_FOR_PAD = "special_marker_for_table_padding"
+# Escape all special characters.  Output is less readable, but avoids
+# corner case formatting issues.
+ESCAPE_SNOB = True
+
+# Put the links after each paragraph instead of at the end.
+LINKS_EACH_PARAGRAPH = False
+
+# Wrap long lines at position. 0 for no wrapping.
+BODY_WIDTH = 0
+
+# Don't show internal links (href="#local-anchor") -- corresponding link
+# targets won't be visible in the plain text file anyway.
+SKIP_INTERNAL_LINKS = False
+
+# Use inline, rather than reference, formatting for images and links
+INLINE_LINKS = True
+
+# Protect links from line breaks surrounding them with angle brackets (in
+# addition to their square brackets)
+PROTECT_LINKS = True
+WRAP_LINKS = True
+
+# Wrap list items.
+WRAP_LIST_ITEMS = False
+
+# Wrap tables
+WRAP_TABLES = False
+
+# Number of pixels Google indents nested lists
+GOOGLE_LIST_INDENT = 36
+
+# Values Google and others may use to indicate bold text
+BOLD_TEXT_STYLE_VALUES = ("bold", "700", "800", "900")
+
+IGNORE_ANCHORS = False
+IGNORE_MAILTO_LINKS = False
+IGNORE_IMAGES = False
+IMAGES_AS_HTML = False
+IMAGES_TO_ALT = False
+IMAGES_WITH_SIZE = False
+IGNORE_EMPHASIS = False
+MARK_CODE = True
+DECODE_ERRORS = "strict"
+DEFAULT_IMAGE_ALT = ""
+PAD_TABLES = True
+
+# Convert links with same href and text to <href> format
+# if they are absolute links
+USE_AUTOMATIC_LINKS = True
+
+# For checking space-only lines on line 771
+RE_SPACE = re.compile(r"\s\+")
+
+RE_ORDERED_LIST_MATCHER = re.compile(r"\d+\.\s")
+RE_UNORDERED_LIST_MATCHER = re.compile(r"[-\*\+]\s")
+RE_MD_CHARS_MATCHER = re.compile(r"([\\\[\]\(\)])")
+RE_MD_CHARS_MATCHER_ALL = re.compile(r"([`\*_{}\[\]\(\)#!])")
+
+# to find links in the text
+RE_LINK = re.compile(r"(\[.*?\] ?\(.*?\))|(\[.*?\]:.*?)")
+
+# to find table separators
+RE_TABLE = re.compile(r" \| ")
+
+RE_MD_DOT_MATCHER = re.compile(
+    r"""
+    ^             # start of line
+    (\s*\d+)      # optional whitespace and a number
+    (\.)          # dot
+    (?=\s)        # lookahead assert whitespace
+    """,
+    re.MULTILINE | re.VERBOSE,
+)
+RE_MD_PLUS_MATCHER = re.compile(
+    r"""
+    ^
+    (\s*)
+    (\+)
+    (?=\s)
+    """,
+    flags=re.MULTILINE | re.VERBOSE,
+)
+RE_MD_DASH_MATCHER = re.compile(
+    r"""
+    ^
+    (\s*)
+    (-)
+    (?=\s|\-)     # followed by whitespace (bullet list, or spaced out hr)
+                  # or another dash (header or hr)
+    """,
+    flags=re.MULTILINE | re.VERBOSE,
+)
+RE_SLASH_CHARS = r"\`*_{}[]()#+-.!"
+RE_MD_BACKSLASH_MATCHER = re.compile(
+    r"""
+    (\\)          # match one slash
+    (?=[%s])      # followed by a char that requires escaping
+    """
+    % re.escape(RE_SLASH_CHARS),
+    flags=re.VERBOSE,
+)
+
+UNIFIABLE = {
+    "rsquo": "'",
+    "lsquo": "'",
+    "rdquo": '"',
+    "ldquo": '"',
+    "copy": "(C)",
+    "mdash": "--",
+    "nbsp": " ",
+    "rarr": "->",
+    "larr": "<-",
+    "middot": "*",
+    "ndash": "-",
+    "oelig": "oe",
+    "aelig": "ae",
+    "agrave": "a",
+    "aacute": "a",
+    "acirc": "a",
+    "atilde": "a",
+    "auml": "a",
+    "aring": "a",
+    "egrave": "e",
+    "eacute": "e",
+    "ecirc": "e",
+    "euml": "e",
+    "igrave": "i",
+    "iacute": "i",
+    "icirc": "i",
+    "iuml": "i",
+    "ograve": "o",
+    "oacute": "o",
+    "ocirc": "o",
+    "otilde": "o",
+    "ouml": "o",
+    "ugrave": "u",
+    "uacute": "u",
+    "ucirc": "u",
+    "uuml": "u",
+    "lrm": "",
+    "rlm": "",
+}
+
+# Format tables in HTML rather than Markdown syntax
+BYPASS_TABLES = False
+# Ignore table-related tags (table, th, td, tr) while keeping rows
+IGNORE_TABLES = False
+
+
+# Use a single line break after a block element rather than two line breaks.
+# NOTE: Requires body width setting to be 0.
+SINGLE_LINE_BREAK = False
+
+
+# Use double quotation marks when converting the <q> tag.
+OPEN_QUOTE = '"'
+CLOSE_QUOTE = '"'
--- a/migration/html2text/elements.py
+++ b/migration/html2text/elements.py
@@ -0,0 +1,18 @@
+from typing import Dict, Optional
+
+
+class AnchorElement:
+    __slots__ = ["attrs", "count", "outcount"]
+
+    def __init__(self, attrs: Dict[str, Optional[str]], count: int, outcount: int):
+        self.attrs = attrs
+        self.count = count
+        self.outcount = outcount
+
+
+class ListElement:
+    __slots__ = ["name", "num"]
+
+    def __init__(self, name: str, num: int):
+        self.name = name
+        self.num = num
--- a/migration/html2text/py.typed
+++ b/migration/html2text/py.typed
--- a/migration/html2text/typing.py
+++ b/migration/html2text/typing.py
@@ -0,0 +1,3 @@
+class OutCallback:
+    def __call__(self, s: str) -> None:
+        ...
--- a/migration/html2text/utils.py
+++ b/migration/html2text/utils.py
@@ -0,0 +1,282 @@
+import html.entities
+from typing import Dict, List, Optional
+
+from . import config
+
+unifiable_n = {
+    html.entities.name2codepoint[k]: v for k, v in config.UNIFIABLE.items() if k != "nbsp"
+}
+
+
+def hn(tag: str) -> int:
+    if tag[0] == "h" and len(tag) == 2:
+        n = tag[1]
+        if "0" < n <= "9":
+            return int(n)
+    return 0
+
+
+def dumb_property_dict(style: str) -> Dict[str, str]:
+    """
+    :returns: A hash of css attributes
+    """
+    return {
+        x.strip().lower(): y.strip().lower()
+        for x, y in [z.split(":", 1) for z in style.split(";") if ":" in z]
+    }
+
+
+def dumb_css_parser(data: str) -> Dict[str, Dict[str, str]]:
+    """
+    :type data: str
+
+    :returns: A hash of css selectors, each of which contains a hash of
+    css attributes.
+    :rtype: dict
+    """
+    # remove @import sentences
+    data += ";"
+    importIndex = data.find("@import")
+    while importIndex != -1:
+        data = data[0:importIndex] + data[data.find(";", importIndex) + 1 :]
+        importIndex = data.find("@import")
+
+    # parse the css. reverted from dictionary comprehension in order to
+    # support older pythons
+    pairs = [x.split("{") for x in data.split("}") if "{" in x.strip()]
+    try:
+        elements = {a.strip(): dumb_property_dict(b) for a, b in pairs}
+    except ValueError:
+        elements = {}  # not that important
+
+    return elements
+
+
+def element_style(
+    attrs: Dict[str, Optional[str]],
+    style_def: Dict[str, Dict[str, str]],
+    parent_style: Dict[str, str],
+) -> Dict[str, str]:
+    """
+    :type attrs: dict
+    :type style_def: dict
+    :type style_def: dict
+
+    :returns: A hash of the 'final' style attributes of the element
+    :rtype: dict
+    """
+    style = parent_style.copy()
+    attrs_class = attrs.get("class")
+    if attrs_class:
+        for css_class in attrs_class.split():
+            css_style = style_def.get("." + css_class, {})
+            style.update(css_style)
+    attrs_style = attrs.get("style")
+    if attrs_style:
+        immediate_style = dumb_property_dict(attrs_style)
+        style.update(immediate_style)
+
+    return style
+
+
+def google_list_style(style: Dict[str, str]) -> str:
+    """
+    Finds out whether this is an ordered or unordered list
+
+    :type style: dict
+
+    :rtype: str
+    """
+    if "list-style-type" in style:
+        list_style = style["list-style-type"]
+        if list_style in ["disc", "circle", "square", "none"]:
+            return "ul"
+
+    return "ol"
+
+
+def google_has_height(style: Dict[str, str]) -> bool:
+    """
+    Check if the style of the element has the 'height' attribute
+    explicitly defined
+
+    :type style: dict
+
+    :rtype: bool
+    """
+    return "height" in style
+
+
+def google_text_emphasis(style: Dict[str, str]) -> List[str]:
+    """
+    :type style: dict
+
+    :returns: A list of all emphasis modifiers of the element
+    :rtype: list
+    """
+    emphasis = []
+    if "text-decoration" in style:
+        emphasis.append(style["text-decoration"])
+    if "font-style" in style:
+        emphasis.append(style["font-style"])
+    if "font-weight" in style:
+        emphasis.append(style["font-weight"])
+
+    return emphasis
+
+
+def google_fixed_width_font(style: Dict[str, str]) -> bool:
+    """
+    Check if the css of the current element defines a fixed width font
+
+    :type style: dict
+
+    :rtype: bool
+    """
+    font_family = ""
+    if "font-family" in style:
+        font_family = style["font-family"]
+    return "courier new" == font_family or "consolas" == font_family
+
+
+def list_numbering_start(attrs: Dict[str, Optional[str]]) -> int:
+    """
+    Extract numbering from list element attributes
+
+    :type attrs: dict
+
+    :rtype: int or None
+    """
+    attrs_start = attrs.get("start")
+    if attrs_start:
+        try:
+            return int(attrs_start) - 1
+        except ValueError:
+            pass
+
+    return 0
+
+
+def skipwrap(para: str, wrap_links: bool, wrap_list_items: bool, wrap_tables: bool) -> bool:
+    # If it appears to contain a link
+    # don't wrap
+    if not wrap_links and config.RE_LINK.search(para):
+        return True
+    # If the text begins with four spaces or one tab, it's a code block;
+    # don't wrap
+    if para[0:4] == "    " or para[0] == "\t":
+        return True
+
+    # If the text begins with only two "--", possibly preceded by
+    # whitespace, that's an emdash; so wrap.
+    stripped = para.lstrip()
+    if stripped[0:2] == "--" and len(stripped) > 2 and stripped[2] != "-":
+        return False
+
+    # I'm not sure what this is for; I thought it was to detect lists,
+    # but there's a <br>-inside-<span> case in one of the tests that
+    # also depends upon it.
+    if stripped[0:1] in ("-", "*") and not stripped[0:2] == "**":
+        return not wrap_list_items
+
+    # If text contains a pipe character it is likely a table
+    if not wrap_tables and config.RE_TABLE.search(para):
+        return True
+
+    # If the text begins with a single -, *, or +, followed by a space,
+    # or an integer, followed by a ., followed by a space (in either
+    # case optionally proceeded by whitespace), it's a list; don't wrap.
+    return bool(
+        config.RE_ORDERED_LIST_MATCHER.match(stripped)
+        or config.RE_UNORDERED_LIST_MATCHER.match(stripped)
+    )
+
+
+def escape_md(text: str) -> str:
+    """
+    Escapes markdown-sensitive characters within other markdown
+    constructs.
+    """
+    return config.RE_MD_CHARS_MATCHER.sub(r"\\\1", text)
+
+
+def escape_md_section(text: str, snob: bool = False) -> str:
+    """
+    Escapes markdown-sensitive characters across whole document sections.
+    """
+    text = config.RE_MD_BACKSLASH_MATCHER.sub(r"\\\1", text)
+
+    if snob:
+        text = config.RE_MD_CHARS_MATCHER_ALL.sub(r"\\\1", text)
+
+    text = config.RE_MD_DOT_MATCHER.sub(r"\1\\\2", text)
+    text = config.RE_MD_PLUS_MATCHER.sub(r"\1\\\2", text)
+    text = config.RE_MD_DASH_MATCHER.sub(r"\1\\\2", text)
+
+    return text
+
+
+def reformat_table(lines: List[str], right_margin: int) -> List[str]:
+    """
+    Given the lines of a table
+    padds the cells and returns the new lines
+    """
+    # find the maximum width of the columns
+    max_width = [len(x.rstrip()) + right_margin for x in lines[0].split("|")]
+    max_cols = len(max_width)
+    for line in lines:
+        cols = [x.rstrip() for x in line.split("|")]
+        num_cols = len(cols)
+
+        # don't drop any data if colspan attributes result in unequal lengths
+        if num_cols < max_cols:
+            cols += [""] * (max_cols - num_cols)
+        elif max_cols < num_cols:
+            max_width += [len(x) + right_margin for x in cols[-(num_cols - max_cols) :]]
+            max_cols = num_cols
+
+        max_width = [max(len(x) + right_margin, old_len) for x, old_len in zip(cols, max_width)]
+
+    # reformat
+    new_lines = []
+    for line in lines:
+        cols = [x.rstrip() for x in line.split("|")]
+        if set(line.strip()) == set("-|"):
+            filler = "-"
+            new_cols = [
+                x.rstrip() + (filler * (M - len(x.rstrip()))) for x, M in zip(cols, max_width)
+            ]
+            new_lines.append("|-" + "|".join(new_cols) + "|")
+        else:
+            filler = " "
+            new_cols = [
+                x.rstrip() + (filler * (M - len(x.rstrip()))) for x, M in zip(cols, max_width)
+            ]
+            new_lines.append("| " + "|".join(new_cols) + "|")
+    return new_lines
+
+
+def pad_tables_in_text(text: str, right_margin: int = 1) -> str:
+    """
+    Provide padding for tables in the text
+    """
+    lines = text.split("\n")
+    table_buffer = []  # type: List[str]
+    table_started = False
+    new_lines = []
+    for line in lines:
+        # Toggle table started
+        if config.TABLE_MARKER_FOR_PAD in line:
+            table_started = not table_started
+            if not table_started:
+                table = reformat_table(table_buffer, right_margin)
+                new_lines.extend(table)
+                table_buffer = []
+                new_lines.append("")
+            continue
+        # Process lines
+        if table_started:
+            table_buffer.append(line)
+        else:
+            new_lines.append(line)
+    return "\n".join(new_lines)
--- a/migration/tables/comments.py
+++ b/migration/tables/comments.py
@@ -0,0 +1,196 @@
+from datetime import datetime, timezone
+
+from dateutil.parser import parse as date_parse
+
+from base.orm import local_session
+from migration.html2text import html2text
+from orm.reaction import Reaction, ReactionKind
+from orm.shout import Shout, ShoutReactionsFollower
+from orm.topic import TopicFollower
+from orm.user import User
+
+ts = datetime.now(tz=timezone.utc)
+
+
+def auto_followers(session, topics, reaction_dict):
+    # creating shout's reactions following for reaction author
+    following1 = (
+        session.query(ShoutReactionsFollower)
+        .where(ShoutReactionsFollower.follower == reaction_dict["createdBy"])
+        .filter(ShoutReactionsFollower.shout == reaction_dict["shout"])
+        .first()
+    )
+    if not following1:
+        following1 = ShoutReactionsFollower.create(
+            follower=reaction_dict["createdBy"], shout=reaction_dict["shout"], auto=True
+        )
+        session.add(following1)
+    # creating topics followings for reaction author
+    for t in topics:
+        tf = (
+            session.query(TopicFollower)
+            .where(TopicFollower.follower == reaction_dict["createdBy"])
+            .filter(TopicFollower.topic == t["id"])
+            .first()
+        )
+        if not tf:
+            topic_following = TopicFollower.create(
+                follower=reaction_dict["createdBy"], topic=t["id"], auto=True
+            )
+            session.add(topic_following)
+
+
+def migrate_ratings(session, entry, reaction_dict):
+    for comment_rating_old in entry.get("ratings", []):
+        rater = session.query(User).filter(User.oid == comment_rating_old["createdBy"]).first()
+        re_reaction_dict = {
+            "shout": reaction_dict["shout"],
+            "replyTo": reaction_dict["id"],
+            "kind": ReactionKind.LIKE if comment_rating_old["value"] > 0 else ReactionKind.DISLIKE,
+            "createdBy": rater.id if rater else 1,
+        }
+        cts = comment_rating_old.get("createdAt")
+        if cts:
+            re_reaction_dict["createdAt"] = date_parse(cts)
+        try:
+            # creating reaction from old rating
+            rr = Reaction.create(**re_reaction_dict)
+            following2 = (
+                session.query(ShoutReactionsFollower)
+                .where(ShoutReactionsFollower.follower == re_reaction_dict["createdBy"])
+                .filter(ShoutReactionsFollower.shout == rr.shout)
+                .first()
+            )
+            if not following2:
+                following2 = ShoutReactionsFollower.create(
+                    follower=re_reaction_dict["createdBy"], shout=rr.shout, auto=True
+                )
+                session.add(following2)
+            session.add(rr)
+
+        except Exception as e:
+            print("[migration] comment rating error: %r" % re_reaction_dict)
+            raise e
+    session.commit()
+
+
+async def migrate(entry, storage):
+    """
+    {
+      "_id": "hdtwS8fSyFLxXCgSC",
+      "body": "<p>",
+      "contentItem": "mnK8KsJHPRi8DrybQ",
+      "createdBy": "bMFPuyNg6qAD2mhXe",
+      "thread": "01/",
+      "createdAt": "2016-04-19 04:33:53+00:00",
+      "ratings": [
+            { "createdBy": "AqmRukvRiExNpAe8C", "value": 1 },
+            { "createdBy": "YdE76Wth3yqymKEu5", "value": 1 }
+      ],
+      "rating": 2,
+      "updatedAt": "2020-05-27 19:22:57.091000+00:00",
+      "updatedBy": "0"
+    }
+    ->
+    type Reaction {
+            id: Int!
+            shout: Shout!
+            createdAt: DateTime!
+            createdBy: User!
+            updatedAt: DateTime
+            deletedAt: DateTime
+            deletedBy: User
+            range: String # full / 0:2340
+            kind: ReactionKind!
+            body: String
+            replyTo: Reaction
+            stat: Stat
+            old_id: String
+            old_thread: String
+            }
+    """
+    old_ts = entry.get("createdAt")
+    reaction_dict = {
+        "createdAt": (ts if not old_ts else date_parse(old_ts)),
+        "body": html2text(entry.get("body", "")),
+        "oid": entry["_id"],
+    }
+    shout_oid = entry.get("contentItem")
+    if shout_oid not in storage["shouts"]["by_oid"]:
+        if len(storage["shouts"]["by_oid"]) > 0:
+            return shout_oid
+        else:
+            print("[migration] no shouts migrated yet")
+            raise Exception
+        return
+    else:
+        stage = "started"
+        reaction = None
+        with local_session() as session:
+            author = session.query(User).filter(User.oid == entry["createdBy"]).first()
+            old_shout = storage["shouts"]["by_oid"].get(shout_oid)
+            if not old_shout:
+                raise Exception("no old shout in storage")
+            else:
+                stage = "author and old id found"
+                try:
+                    shout = session.query(Shout).where(Shout.slug == old_shout["slug"]).one()
+                    if shout:
+                        reaction_dict["shout"] = shout.id
+                        reaction_dict["createdBy"] = author.id if author else 1
+                        reaction_dict["kind"] = ReactionKind.COMMENT
+
+                        # creating reaction from old comment
+                        reaction = Reaction.create(**reaction_dict)
+                        session.add(reaction)
+                        # session.commit()
+                        stage = "new reaction commited"
+                        reaction_dict = reaction.dict()
+                        topics = [t.dict() for t in shout.topics]
+                        auto_followers(session, topics, reaction_dict)
+
+                        migrate_ratings(session, entry, reaction_dict)
+
+                        return reaction
+                except Exception as e:
+                    print(e)
+                    print(reaction)
+                    raise Exception(stage)
+    return
+
+
+def migrate_2stage(old_comment, idmap):
+    if old_comment.get("body"):
+        new_id = idmap.get(old_comment.get("oid"))
+        new_id = idmap.get(old_comment.get("_id"))
+        if new_id:
+            new_replyto_id = None
+            old_replyto_id = old_comment.get("replyTo")
+            if old_replyto_id:
+                new_replyto_id = int(idmap.get(old_replyto_id, "0"))
+            with local_session() as session:
+                comment = session.query(Reaction).where(Reaction.id == new_id).first()
+                try:
+                    if new_replyto_id:
+                        new_reply = (
+                            session.query(Reaction).where(Reaction.id == new_replyto_id).first()
+                        )
+                        if not new_reply:
+                            print(new_replyto_id)
+                            raise Exception("cannot find reply by id!")
+                        comment.replyTo = new_reply.id
+                        session.add(comment)
+                    srf = (
+                        session.query(ShoutReactionsFollower)
+                        .where(ShoutReactionsFollower.shout == comment.shout)
+                        .filter(ShoutReactionsFollower.follower == comment.createdBy)
+                        .first()
+                    )
+                    if not srf:
+                        srf = ShoutReactionsFollower.create(
+                            shout=comment.shout, follower=comment.createdBy, auto=True
+                        )
+                        session.add(srf)
+                    session.commit()
+                except Exception:
+                    raise Exception("cannot find a comment by oldid")
--- a/migration/tables/content_items.py
+++ b/migration/tables/content_items.py
@@ -0,0 +1,399 @@
+import json
+import re
+from datetime import datetime, timezone
+
+from dateutil.parser import parse as date_parse
+from sqlalchemy.exc import IntegrityError
+from transliterate import translit
+
+from base.orm import local_session
+from migration.extract import extract_html, extract_media
+from orm.reaction import Reaction, ReactionKind
+from orm.shout import Shout, ShoutReactionsFollower, ShoutTopic
+from orm.topic import Topic, TopicFollower
+from orm.user import User
+from services.stat.viewed import ViewedStorage
+
+OLD_DATE = "2016-03-05 22:22:00.350000"
+ts = datetime.now(tz=timezone.utc)
+type2layout = {
+    "Article": "article",
+    "Literature": "literature",
+    "Music": "music",
+    "Video": "video",
+    "Image": "image",
+}
+
+anondict = {"slug": "anonymous", "id": 1, "name": "Аноним"}
+discours = {"slug": "discours", "id": 2, "name": "Дискурс"}
+
+
+def get_shout_slug(entry):
+    slug = entry.get("slug", "")
+    if not slug:
+        for friend in entry.get("friendlySlugs", []):
+            slug = friend.get("slug", "")
+            if slug:
+                break
+    slug = re.sub("[^0-9a-zA-Z]+", "-", slug)
+    return slug
+
+
+def create_author_from_app(app):
+    user = None
+    userdata = None
+    # check if email is used
+    if app["email"]:
+        with local_session() as session:
+            user = session.query(User).where(User.email == app["email"]).first()
+            if not user:
+                # print('[migration] app %r' % app)
+                name = app.get("name")
+                if name:
+                    slug = translit(name, "ru", reversed=True).lower()
+                    slug = re.sub("[^0-9a-zA-Z]+", "-", slug)
+                    print("[migration] created slug %s" % slug)
+                    # check if slug is used
+                    if slug:
+                        user = session.query(User).where(User.slug == slug).first()
+
+                        # get slug from email
+                        if user:
+                            slug = app["email"].split("@")[0]
+                            user = session.query(User).where(User.slug == slug).first()
+                            # one more try
+                            if user:
+                                slug += "-author"
+                                user = session.query(User).where(User.slug == slug).first()
+
+                # create user with application data
+                if not user:
+                    userdata = {
+                        "username": app["email"],
+                        "email": app["email"],
+                        "name": app.get("name", ""),
+                        "emailConfirmed": False,
+                        "slug": slug,
+                        "createdAt": ts,
+                        "lastSeen": ts,
+                    }
+                    # print('[migration] userdata %r' % userdata)
+                    user = User.create(**userdata)
+                    session.add(user)
+                    session.commit()
+                    userdata["id"] = user.id
+
+            userdata = user.dict()
+        return userdata
+    else:
+        raise Exception("app is not ok", app)
+
+
+async def create_shout(shout_dict):
+    s = Shout.create(**shout_dict)
+    author = s.authors[0]
+    with local_session() as session:
+        srf = (
+            session.query(ShoutReactionsFollower)
+            .where(ShoutReactionsFollower.shout == s.id)
+            .filter(ShoutReactionsFollower.follower == author.id)
+            .first()
+        )
+        if not srf:
+            srf = ShoutReactionsFollower.create(shout=s.id, follower=author.id, auto=True)
+            session.add(srf)
+        session.commit()
+    return s
+
+
+async def get_user(entry, storage):
+    app = entry.get("application")
+    userdata = None
+    user_oid = None
+    if app:
+        userdata = create_author_from_app(app)
+    else:
+        user_oid = entry.get("createdBy")
+        if user_oid == "0":
+            userdata = discours
+        elif user_oid:
+            userdata = storage["users"]["by_oid"].get(user_oid)
+        if not userdata:
+            print("no userdata by oid, anonymous")
+            userdata = anondict
+            print(app)
+    # cleanup slug
+    if userdata:
+        slug = userdata.get("slug", "")
+        if slug:
+            slug = re.sub("[^0-9a-zA-Z]+", "-", slug)
+            userdata["slug"] = slug
+    else:
+        userdata = anondict
+
+    user = await process_user(userdata, storage, user_oid)
+    return user, user_oid
+
+
+async def migrate(entry, storage):
+    author, user_oid = await get_user(entry, storage)
+    r = {
+        "layout": type2layout[entry["type"]],
+        "title": entry["title"],
+        "authors": [
+            author,
+        ],
+        "slug": get_shout_slug(entry),
+        "cover": (
+            "https://images.discours.io/unsafe/" + entry["thumborId"]
+            if entry.get("thumborId")
+            else entry.get("image", {}).get("url")
+        ),
+        "visibility": "public" if entry.get("published") else "community",
+        "publishedAt": date_parse(entry.get("publishedAt")) if entry.get("published") else None,
+        "deletedAt": date_parse(entry.get("deletedAt")) if entry.get("deletedAt") else None,
+        "createdAt": date_parse(entry.get("createdAt", OLD_DATE)),
+        "updatedAt": date_parse(entry["updatedAt"]) if "updatedAt" in entry else ts,
+        "createdBy": author.id,
+        "topics": await add_topics_follower(entry, storage, author),
+        "body": extract_html(entry, cleanup=True),
+    }
+
+    # main topic patch
+    r["mainTopic"] = r["topics"][0]
+
+    # published author auto-confirm
+    if entry.get("published"):
+        with local_session() as session:
+            # update user.emailConfirmed if published
+            author.emailConfirmed = True
+            session.add(author)
+            session.commit()
+
+    # media
+    media = extract_media(entry)
+    r["media"] = json.dumps(media, ensure_ascii=True) if media else None
+
+    # ----------------------------------- copy
+    shout_dict = r.copy()
+    del shout_dict["topics"]
+
+    try:
+        # save shout to db
+        shout_dict["oid"] = entry.get("_id", "")
+        shout = await create_shout(shout_dict)
+    except IntegrityError as e:
+        print("[migration] create_shout integrity error", e)
+        shout = await resolve_create_shout(shout_dict)
+    except Exception as e:
+        raise Exception(e)
+
+    # udpate data
+    shout_dict = shout.dict()
+    shout_dict["authors"] = [
+        author.dict(),
+    ]
+
+    # shout topics aftermath
+    shout_dict["topics"] = await topics_aftermath(r, storage)
+
+    # content_item ratings to reactions
+    await content_ratings_to_reactions(entry, shout_dict["slug"])
+
+    # shout views
+    await ViewedStorage.increment(
+        shout_dict["slug"], amount=entry.get("views", 1), viewer="old-discours"
+    )
+    # del shout_dict['ratings']
+
+    storage["shouts"]["by_oid"][entry["_id"]] = shout_dict
+    storage["shouts"]["by_slug"][shout_dict["slug"]] = shout_dict
+    return shout_dict
+
+
+async def add_topics_follower(entry, storage, user):
+    topics = set([])
+    category = entry.get("category")
+    topics_by_oid = storage["topics"]["by_oid"]
+    oids = [
+        category,
+    ] + entry.get("tags", [])
+    for toid in oids:
+        tslug = topics_by_oid.get(toid, {}).get("slug")
+        if tslug:
+            topics.add(tslug)
+    ttt = list(topics)
+    # add author as TopicFollower
+    with local_session() as session:
+        for tpcslug in topics:
+            try:
+                tpc = session.query(Topic).where(Topic.slug == tpcslug).first()
+                if tpc:
+                    tf = (
+                        session.query(TopicFollower)
+                        .where(TopicFollower.follower == user.id)
+                        .filter(TopicFollower.topic == tpc.id)
+                        .first()
+                    )
+                    if not tf:
+                        tf = TopicFollower.create(topic=tpc.id, follower=user.id, auto=True)
+                        session.add(tf)
+                        session.commit()
+            except IntegrityError:
+                print("[migration.shout] hidden by topic " + tpc.slug)
+    # main topic
+    maintopic = storage["replacements"].get(topics_by_oid.get(category, {}).get("slug"))
+    if maintopic in ttt:
+        ttt.remove(maintopic)
+    ttt.insert(0, maintopic)
+    return ttt
+
+
+async def process_user(userdata, storage, oid):
+    with local_session() as session:
+        uid = userdata.get("id")  # anonymous as
+        if not uid:
+            print(userdata)
+            print("has no id field, set it @anonymous")
+            userdata = anondict
+            uid = 1
+        user = session.query(User).filter(User.id == uid).first()
+        if not user:
+            try:
+                slug = userdata["slug"].lower().strip()
+                slug = re.sub("[^0-9a-zA-Z]+", "-", slug)
+                userdata["slug"] = slug
+                user = User.create(**userdata)
+                session.add(user)
+                session.commit()
+            except IntegrityError:
+                print(f"[migration] user creating with slug {userdata['slug']}")
+                print("[migration] from userdata")
+                print(userdata)
+                raise Exception("[migration] cannot create user in content_items.get_user()")
+        if user.id == 946:
+            print("[migration] ***************** ALPINA")
+        if user.id == 2:
+            print("[migration] +++++++++++++++++ DISCOURS")
+        userdata["id"] = user.id
+        userdata["createdAt"] = user.createdAt
+        storage["users"]["by_slug"][userdata["slug"]] = userdata
+        storage["users"]["by_oid"][oid] = userdata
+        if not user:
+            raise Exception("could not get a user")
+        return user
+
+
+async def resolve_create_shout(shout_dict):
+    with local_session() as session:
+        s = session.query(Shout).filter(Shout.slug == shout_dict["slug"]).first()
+        bump = False
+        if s:
+            if s.createdAt != shout_dict["createdAt"]:
+                # create new with different slug
+                shout_dict["slug"] += "-" + shout_dict["layout"]
+                try:
+                    await create_shout(shout_dict)
+                except IntegrityError as e:
+                    print(e)
+                    bump = True
+            else:
+                # update old
+                for key in shout_dict:
+                    if key in s.__dict__:
+                        if s.__dict__[key] != shout_dict[key]:
+                            print("[migration] shout already exists, but differs in %s" % key)
+                            bump = True
+                    else:
+                        print("[migration] shout already exists, but lacks %s" % key)
+                        bump = True
+                if bump:
+                    s.update(shout_dict)
+        else:
+            print("[migration] something went wrong with shout: \n%r" % shout_dict)
+            raise Exception("")
+        session.commit()
+        return s
+
+
+async def topics_aftermath(entry, storage):
+    r = []
+    for tpc in filter(lambda x: bool(x), entry["topics"]):
+        oldslug = tpc
+        newslug = storage["replacements"].get(oldslug, oldslug)
+
+        if newslug:
+            with local_session() as session:
+                shout = session.query(Shout).where(Shout.slug == entry["slug"]).first()
+                new_topic = session.query(Topic).where(Topic.slug == newslug).first()
+
+                shout_topic_old = (
+                    session.query(ShoutTopic)
+                    .join(Shout)
+                    .join(Topic)
+                    .filter(Shout.slug == entry["slug"])
+                    .filter(Topic.slug == oldslug)
+                    .first()
+                )
+                if shout_topic_old:
+                    shout_topic_old.update({"topic": new_topic.id})
+                else:
+                    shout_topic_new = (
+                        session.query(ShoutTopic)
+                        .join(Shout)
+                        .join(Topic)
+                        .filter(Shout.slug == entry["slug"])
+                        .filter(Topic.slug == newslug)
+                        .first()
+                    )
+                    if not shout_topic_new:
+                        try:
+                            ShoutTopic.create(**{"shout": shout.id, "topic": new_topic.id})
+                        except Exception:
+                            print("[migration] shout topic error: " + newslug)
+                session.commit()
+            if newslug not in r:
+                r.append(newslug)
+        else:
+            print("[migration] ignored topic slug: \n%r" % tpc["slug"])
+            # raise Exception
+    return r
+
+
+async def content_ratings_to_reactions(entry, slug):
+    try:
+        with local_session() as session:
+            for content_rating in entry.get("ratings", []):
+                rater = (
+                    session.query(User).filter(User.oid == content_rating["createdBy"]).first()
+                ) or User.default_user
+                shout = session.query(Shout).where(Shout.slug == slug).first()
+                cts = content_rating.get("createdAt")
+                reaction_dict = {
+                    "createdAt": date_parse(cts) if cts else None,
+                    "kind": ReactionKind.LIKE
+                    if content_rating["value"] > 0
+                    else ReactionKind.DISLIKE,
+                    "createdBy": rater.id,
+                    "shout": shout.id,
+                }
+                reaction = (
+                    session.query(Reaction)
+                    .filter(Reaction.shout == reaction_dict["shout"])
+                    .filter(Reaction.createdBy == reaction_dict["createdBy"])
+                    .filter(Reaction.kind == reaction_dict["kind"])
+                    .first()
+                )
+                if reaction:
+                    k = ReactionKind.AGREE if content_rating["value"] > 0 else ReactionKind.DISAGREE
+                    reaction_dict["kind"] = k
+                    reaction.update(reaction_dict)
+                    session.add(reaction)
+                else:
+                    rea = Reaction.create(**reaction_dict)
+                    session.add(rea)
+                # shout_dict['ratings'].append(reaction_dict)
+
+            session.commit()
+    except Exception:
+        print("[migration] content_item.ratings error: \n%r" % content_rating)
--- a/migration/tables/remarks.py
+++ b/migration/tables/remarks.py
@@ -0,0 +1,35 @@
+# from base.orm import local_session
+
+# from migration.extract import extract_md
+# from migration.html2text import html2text
+# from orm.reaction import Reaction, ReactionKind
+
+
+# def migrate(entry, storage):
+#     post_oid = entry["contentItem"]
+#     print(post_oid)
+#     shout_dict = storage["shouts"]["by_oid"].get(post_oid)
+#     if shout_dict:
+#         print(shout_dict["body"])
+#         remark = {
+#             "shout": shout_dict["id"],
+#             "body": extract_md(html2text(entry["body"]), shout_dict),
+#             "kind": ReactionKind.REMARK,
+#         }
+#
+#         if entry.get("textBefore"):
+#             remark["range"] = (
+#                 str(shout_dict["body"].index(entry["textBefore"] or ""))
+#                 + ":"
+#                 + str(
+#                     shout_dict["body"].index(entry["textAfter"] or "")
+#                     + len(entry["textAfter"] or "")
+#                 )
+#             )
+#
+#         with local_session() as session:
+#             rmrk = Reaction.create(**remark)
+#             session.commit()
+#             del rmrk["_sa_instance_state"]
+#             return rmrk
+#     return
--- a/migration/tables/replacements.json
+++ b/migration/tables/replacements.json
@@ -0,0 +1,828 @@
+{
+    "207": "207",
+    "1990-e": "90s",
+    "2000-e": "2000s",
+    "90-e": "90s",
+    "Georgia": "georgia",
+    "Japan": "japan",
+    "Sweden": "sweden",
+    "abstraktsiya": "abstract",
+    "absurdism": "absurdism",
+    "acclimatization": "acclimatisation",
+    "activism": "activism",
+    "adolf-gitler": "adolf-hitler",
+    "afrika": "africa",
+    "agata-kristi": "agatha-christie",
+    "agressivnoe-povedenie": "agression",
+    "agressiya": "agression",
+    "aktsii": "actions",
+    "aktsionizm": "actionism",
+    "alber-kamyu": "albert-kamus",
+    "albomy": "albums",
+    "aleksandr-griboedov": "aleksander-griboedov",
+    "aleksandr-pushkin": "aleksander-pushkin",
+    "aleksandr-solzhenitsyn": "aleksander-solzhenitsyn",
+    "aleksandr-vvedenskiy": "aleksander-vvedensky",
+    "aleksey-navalnyy": "alexey-navalny",
+    "alfavit": "alphabet",
+    "alkogol": "alcohol",
+    "alternativa": "alternative",
+    "alternative": "alternative",
+    "alternativnaya-istoriya": "alternative-history",
+    "amerika": "america",
+    "anarhizm": "anarchism",
+    "anatoliy-mariengof": "anatoly-mariengof",
+    "ancient-russia": "ancient-russia",
+    "andegraund": "underground",
+    "andrey-platonov": "andrey-platonov",
+    "andrey-rodionov": "andrey-rodionov",
+    "andrey-tarkovskiy": "andrey-tarkovsky",
+    "angliyskie-istorii": "english-stories",
+    "angliyskiy-yazyk": "english-langugae",
+    "ango": "ango",
+    "animation": "animation",
+    "animatsiya": "animation",
+    "anime": "anime",
+    "anri-volohonskiy": "anri-volohonsky",
+    "antifashizm": "anti-faschism",
+    "antiquity": "antiquity",
+    "antiutopiya": "dystopia",
+    "anton-dolin": "anton-dolin",
+    "antropology": "antropology",
+    "antropotsen": "antropocenus",
+    "architecture": "architecture",
+    "arheologiya": "archeology",
+    "arhetipy": "archetypes",
+    "arhiv": "archive",
+    "aristokraty": "aristocracy",
+    "aristotel": "aristotle",
+    "arktika": "arctic",
+    "armiya": "army",
+    "armiya-1": "army",
+    "art": "art",
+    "art-is": "art-is",
+    "artists": "artists",
+    "ateizm": "atheism",
+    "audio-poetry": "audio-poetry",
+    "audiopoeziya": "audio-poetry",
+    "audiospektakl": "audio-spectacles",
+    "auktsyon": "auktsyon",
+    "avangard": "avantgarde",
+    "avtofikshn": "autofiction",
+    "avtorskaya-pesnya": "bardsongs",
+    "azbuka-immigratsii": "immigration-basics",
+    "aziatskiy-kinematograf": "asian-cinema",
+    "b-movie": "b-movie",
+    "bannye-chteniya": "sauna-reading",
+    "bardsongs": "bardsongs",
+    "bdsm": "bdsm",
+    "beecake": "beecake",
+    "belarus": "belarus",
+    "belgiya": "belgium",
+    "bertold-breht": "berttold-brecht",
+    "bezumie": "madness",
+    "biography": "biography",
+    "biologiya": "biology",
+    "bipolyarnoe-rasstroystvo": "bipolar-disorder",
+    "bitniki": "beatnics",
+    "biznes": "business",
+    "blizhniy-vostok": "middle-east",
+    "blizost": "closeness",
+    "blocked-in-russia": "blocked-in-russia",
+    "blokada": "blockade",
+    "bob-dilan": "bob-dylan",
+    "bog": "god",
+    "bol": "pain",
+    "bolotnoe-delo": "bolotnaya-case",
+    "books": "books",
+    "boris-eltsin": "boris-eltsin",
+    "boris-godunov": "boris-godunov",
+    "boris-grebenschikov": "boris-grebenschikov",
+    "boris-nemtsov": "boris-nemtsov",
+    "boris-pasternak": "boris-pasternak",
+    "brak": "marriage",
+    "bret-iston-ellis": "bret-iston-ellis",
+    "buddizm": "buddhism",
+    "bullying": "bullying",
+    "bunt": "riot",
+    "burning-man": "burning-man",
+    "bytie": "being",
+    "byurokratiya": "bureaucracy",
+    "capitalism": "capitalism",
+    "censored-in-russia": "censored-in-russia",
+    "ch-rno-beloe": "black-and-white",
+    "ch-rnyy-yumor": "black-humour",
+    "chapters": "chapters",
+    "charity": "charity",
+    "chayldfri": "childfree",
+    "chechenskaya-voyna": "chechen-war",
+    "chechnya": "chechnya",
+    "chelovek": "male",
+    "chernobyl": "chernobyl",
+    "chernyy-yumor": "black-humour",
+    "children": "children",
+    "china": "china",
+    "chinovniki": "bureaucracy",
+    "chukotka": "chukotka",
+    "chuma": "plague",
+    "church": "church",
+    "cinema": "cinema",
+    "city": "city",
+    "civil-position": "civil-position",
+    "clips": "clips",
+    "collage": "collage",
+    "comics": "comics",
+    "conspiracy-theory": "conspiracy-theory",
+    "contemporary-art": "contemporary-art",
+    "contemporary-poetry": "poetry",
+    "contemporary-prose": "prose",
+    "coronavirus": "coronavirus",
+    "corruption": "corruption",
+    "creative-writing-school": "creative-writing-school",
+    "crime": "crime",
+    "criticism": "criticism",
+    "critiques": "reviews",
+    "culture": "culture",
+    "dadaizm": "dadaism",
+    "daniel-defo": "daniel-defoe",
+    "daniil-harms": "daniil-kharms",
+    "dante-aligeri": "dante-alighieri",
+    "darkveyv": "darkwave",
+    "death": "death",
+    "debaty": "debats",
+    "delo-seti": "seti-case",
+    "democracy": "democracy",
+    "demografiya": "demographics",
+    "demonstrations": "demonstrations",
+    "depression": "depression",
+    "derevnya": "village",
+    "derrida": "derrida",
+    "design": "design",
+    "detskie-doma": "orphanages",
+    "detstvo": "childhood",
+    "devid-linch": "david-linch",
+    "devyanostye": "90s",
+    "dialog": "dialogue",
+    "digital": "digital",
+    "digital-art": "digital-art",
+    "dinozavry": "dinosaurs",
+    "directing": "directing",
+    "diskurs": "discours",
+    "diskurs-1": "discourse",
+    "diskurs-analiz": "discourse-analytics",
+    "dissidenty": "dissidents",
+    "diy": "diy",
+    "dmitriy-donskoy": "dmitriy-donskoy",
+    "dmitriy-prigov": "dmitriy-prigov",
+    "dnevnik-1": "dairy",
+    "dnevniki": "dairies",
+    "documentary": "documentary",
+    "dokumentalnaya-poema": "documentary-poem",
+    "dokumentalnaya-poeziya": "documentary-poetry",
+    "dokumenty": "doсuments",
+    "domashnee-nasilie": "home-terror",
+    "donald-tramp": "donald-trump",
+    "donbass": "donbass",
+    "donbass-diary": "donbass-diary",
+    "donorstvo": "donation",
+    "dozhd": "rain",
+    "drama": "drama",
+    "dramaturgy": "dramaturgy",
+    "drawing": "drawing",
+    "drevo-zhizni": "tree-of-life",
+    "drugs": "drugs",
+    "duh": "spirit",
+    "dzhaz": "jazz",
+    "dzhek-keruak": "jack-keruak",
+    "dzhim-morrison": "jim-morrison",
+    "dzhordzh-romero": "george-romero",
+    "dzhordzho-agamben": "giorgio-agamben",
+    "ecology": "ecology",
+    "economics": "economics",
+    "eda": "food",
+    "editorial-statements": "editorial-statements",
+    "eduard-limonov": "eduard-limonov",
+    "education": "education",
+    "egor-letov": "egor-letov",
+    "ekspat": "expat",
+    "eksperiment": "experiments",
+    "eksperimentalnaya-muzyka": "experimental-music",
+    "ekspressionizm": "expressionism",
+    "ekstremizm": "extremism",
+    "ekzistentsializm-1": "existentialism",
+    "ekzistentsiya": "existence",
+    "elections": "elections",
+    "electronic": "electronics",
+    "electronics": "electronics",
+    "elena-glinskaya": "elena-glinskaya",
+    "elena-guro": "elena-guro",
+    "elizaveta-mnatsakanova": "elizaveta-mnatsakanova",
+    "embient": "ambient",
+    "emigration": "emigration",
+    "emil-dyurkgeym": "emile-durkheim",
+    "emotsii": "emotions",
+    "empiric": "empiric",
+    "epidemiya": "pandemic",
+    "erich-von-neff": "erich-von-neff",
+    "erotika": "erotics",
+    "essay": "essay",
+    "estetika": "aestetics",
+    "etika": "ethics",
+    "etno": "ethno",
+    "etnos": "ethnics",
+    "everyday-life": "everyday-life",
+    "evgeniy-onegin": "eugene-onegin",
+    "evolyutsiya": "evolution",
+    "exhibitions": "exhibitions",
+    "experience": "experiences",
+    "experimental": "experimental",
+    "experimental-music": "experimental-music",
+    "explanation": "explanation",
+    "faktcheking": "fact-checking",
+    "falsifikatsii": "falsifications",
+    "family": "family",
+    "fanfiki": "fan-fiction",
+    "fantastika": "sci-fi",
+    "fatalizm": "fatalism",
+    "fedor-dostoevskiy": "fedor-dostoevsky",
+    "fedor-ioannovich": "fedor-ioannovich",
+    "feleton": "feuilleton",
+    "feminism": "feminism",
+    "fenomenologiya": "phenomenology",
+    "fentezi": "fantasy",
+    "festival": "festival",
+    "festival-territoriya": "festival-territory",
+    "folk": "folk",
+    "folklor": "folklore",
+    "fotoreportazh": "photoreports",
+    "france": "france",
+    "frants-kafka": "franz-kafka",
+    "frederik-begbeder": "frederick-begbeder",
+    "freedom": "freedom",
+    "friendship": "friendship",
+    "fsb": "fsb",
+    "futbol": "footbool",
+    "future": "future",
+    "futuristy": "futurists",
+    "futurizm": "futurism",
+    "galereya": "gallery",
+    "galereya-anna-nova": "gallery-anna-nova",
+    "gdr": "gdr",
+    "gender": "gender",
+    "gendernyy-diskurs": "gender",
+    "gennadiy-aygi": "gennadiy-aygi",
+    "gerhard-rihter": "gerhard-rihter",
+    "germaniya": "germany",
+    "germenevtika": "hermeneutics",
+    "geroi": "heroes",
+    "girls": "girls",
+    "gkchp": "gkchp",
+    "glitch": "glitch",
+    "globalizatsiya": "globalisation",
+    "gollivud": "hollywood",
+    "gonzo": "gonzo",
+    "gore-ot-uma": "woe-from-wit",
+    "graffiti": "graffiti",
+    "graficheskaya-novella": "graphic-novell",
+    "graphics": "graphics",
+    "gravyura": "engraving",
+    "grazhdanskaya-oborona": "grazhdanskaya-oborona",
+    "gretsiya": "greece",
+    "griby": "mushrooms",
+    "gruziya-2": "georgia",
+    "gulag": "gulag",
+    "han-batyy": "khan-batyy",
+    "hayku": "haiku",
+    "health": "health",
+    "himiya": "chemistry",
+    "hip-hop": "hip-hop",
+    "history": "history",
+    "history-of-russia": "history-of-russia",
+    "holokost": "holocaust",
+    "horeografiya": "choreography",
+    "horror": "horror",
+    "hospis": "hospice",
+    "hristianstvo": "christianity",
+    "humans": "humans",
+    "humour": "humour",
+    "ideologiya": "ideology",
+    "idm": "idm",
+    "igil": "isis",
+    "igor-pomerantsev": "igor-pomerantsev",
+    "igra": "game",
+    "igra-prestolov": "game-of-throne",
+    "igry": "games",
+    "iisus-hristos": "jesus-christ",
+    "illness": "illness",
+    "illustration-history": "illustration-history",
+    "illustrations": "illustrations",
+    "imazhinizm": "imagism",
+    "immanuil-kant": "immanuel-kant",
+    "impressionizm": "impressionism",
+    "improvizatsiya": "improvisation",
+    "indi": "indie",
+    "individualizm": "individualism",
+    "infografika": "infographics",
+    "informatsiya": "information",
+    "ingmar-bergman": "ingmar-bergman",
+    "inklyuziya": "inclusion",
+    "installyatsiya": "installation",
+    "internet": "internet",
+    "interview": "interview",
+    "invalidnost": "disability",
+    "investigations": "investigations",
+    "iosif-brodskiy": "joseph-brodsky",
+    "iosif-stalin": "joseph-stalin",
+    "iskusstvennyy-intellekt": "artificial-intelligence",
+    "islam": "islam",
+    "istoriya-moskvy": "moscow-history",
+    "istoriya-nauki": "history-of-sceince",
+    "istoriya-o-medsestre": "nurse-story",
+    "istoriya-teatra": "theatre-history",
+    "italiya": "italy",
+    "italyanskiy-yazyk": "italian-language",
+    "iudaika": "judaica",
+    "ivan-groznyy": "ivan-grozny",
+    "ivan-iii-gorbatyy": "ivan-iii-gorbaty",
+    "ivan-kalita": "ivan-kalita",
+    "ivan-krylov": "ivan-krylov",
+    "izobreteniya": "inventions",
+    "izrail-1": "israel",
+    "jazz": "jazz",
+    "john-lennon": "john-lennon",
+    "journalism": "journalism",
+    "justice": "justice",
+    "k-pop": "k-pop",
+    "kalligrafiya": "calligraphy",
+    "karikatura": "caricatures",
+    "kartochki-rubinshteyna": "rubinstein-cards",
+    "katrin-nenasheva": "katrin-nenasheva",
+    "kavarga": "kavarga",
+    "kavkaz": "caucasus",
+    "kazan": "kazan",
+    "kiberbezopasnost": "cybersecurity",
+    "kinoklub": "cinema-club",
+    "kinokritika": "film-criticism",
+    "kirill-serebrennikov": "kirill-serebrennikov",
+    "kladbische": "cemetery",
+    "klassika": "classic",
+    "kollektivnoe-bessoznatelnoe": "сollective-unconscious",
+    "komediya": "comedy",
+    "kommunikatsii": "communications",
+    "kommunizm": "communism",
+    "kommuny": "communes",
+    "kompyuternye-igry": "computer-games",
+    "konets-vesny": "end-of-spring",
+    "konservatizm": "conservatism",
+    "kontrkultura": "counter-culture",
+    "kontseptualizm": "conceptualism",
+    "korotkometrazhka": "cinema-shorts",
+    "kosmos": "cosmos",
+    "kraudfanding": "crowdfunding",
+    "kriptovalyuty": "cryptocurrencies",
+    "krizis": "crisis",
+    "krov": "blood",
+    "krym": "crimea",
+    "kulturologiya": "culturology",
+    "kulty": "cults",
+    "kurdistan": "kurdistan",
+    "kurt-kobeyn": "kurt-cobain",
+    "kurt-vonnegut": "kurt-vonnegut",
+    "kvir": "queer",
+    "laboratoriya": "lab",
+    "language": "languages",
+    "lars-fon-trier": "lars-fon-trier",
+    "laws": "laws",
+    "lectures": "lectures",
+    "leto": "summer",
+    "lev-tolstoy": "leo-tolstoy",
+    "lgbt": "lgbt",
+    "liberalizm": "liberalism",
+    "libertarianstvo": "libertarianism",
+    "life": "life",
+    "likbez": "likbez",
+    "lingvistika": "linguistics",
+    "lirika": "lirics",
+    "literary-studies": "literary-studies",
+    "literature": "literature",
+    "literaturnyykaver": "literature-cover",
+    "lo-fi": "lo-fi",
+    "lomonosov": "lomonosov",
+    "love": "love",
+    "luzha-goluboy-krovi": "luzha-goluboy-krovi",
+    "lyudvig-vitgenshteyn": "ludwig-wittgenstein",
+    "lzhedmitriy": "false-dmitry",
+    "lzhenauka": "pseudoscience",
+    "magiya": "magic",
+    "maks-veber": "max-weber",
+    "manifests": "manifests",
+    "manipulyatsii-soznaniem": "mind-manipulation",
+    "marina-abramovich": "marina-abramovich",
+    "marketing": "marketing",
+    "marksizm": "marxism",
+    "marsel-dyushan": "marchel-duchamp",
+    "marsel-prust": "marcel-proust",
+    "martin-haydegger": "martin-hidegger",
+    "matematika": "maths",
+    "mayakovskiy": "vladimir-mayakovsky",
+    "media": "media",
+    "medicine": "medicine",
+    "memuary": "memoirs",
+    "menedzhment": "management",
+    "menty": "police",
+    "merab-mamardashvili": "merab-mamardashvili",
+    "mest": "revenge",
+    "metamodernizm": "metamodern",
+    "metavselennaya": "metaverse",
+    "metro": "metro",
+    "mifologiya": "mythology",
+    "mify": "myth",
+    "mihael-haneke": "michael-haneke",
+    "mihail-baryshnikov": "mihail-baryshnikov",
+    "mihail-bulgakov": "mihail-bulgakov",
+    "mikrotonalnaya-muzyka": "mikrotone-muzyka",
+    "minimalizm": "minimalism",
+    "minkult-privet": "minkult-privet",
+    "mir": "world",
+    "mirovozzrenie": "mindsets",
+    "mishel-fuko": "michel-foucault",
+    "mistika": "mystics",
+    "mitropolit-makariy": "mitropolit-makariy",
+    "mlm": "mlm",
+    "mobilizatsiya": "mobilisation",
+    "moda": "fashion",
+    "modernizm": "modernism",
+    "mokyumentari": "mockumentary",
+    "molodezh": "youth",
+    "moloko-plus": "moloko-plus",
+    "money": "money",
+    "monologs": "monologues",
+    "monstratsiya": "monstration",
+    "moralnaya-otvetstvennost": "moral-responsibility",
+    "more": "sea",
+    "moscow": "moscow",
+    "moshennichestvo": "frauds",
+    "moskovskiy-romanticheskiy-kontseptualizm": "moscow-romantic-conceptualism",
+    "moskovskoe-delo": "moscow-case",
+    "movies": "movies",
+    "mozg": "brain",
+    "multiplikatsiya": "animation",
+    "music": "music",
+    "musulmanstvo": "islam",
+    "muzei": "museum",
+    "muzey": "museum",
+    "muzhchiny": "man",
+    "myshlenie": "thinking",
+    "nagornyy-karabah": "nagorno-karabakh",
+    "nasilie-1": "violence",
+    "natsionalizm": "nationalism",
+    "natsionalnaya-ideya": "national-idea",
+    "natsizm": "nazism",
+    "natyurmort": "nature-morte",
+    "nauchpop": "pop-science",
+    "nbp": "nbp",
+    "nenavist": "hate",
+    "neofitsialnaya-literatura": "unofficial-literature",
+    "neoklassika": "neoclassic",
+    "neprozrachnye-smysly": "hidden-meanings",
+    "neravenstvo": "inequality",
+    "net-voyne": "no-war",
+    "new-year": "new-year",
+    "neyronauka": "neuro-science",
+    "neyroseti": "neural-networks",
+    "niu-vshe": "hse",
+    "nizhniy-novgorod": "nizhny-novgorod",
+    "nko": "nonprofits",
+    "nlo": "ufo",
+    "nobelevskaya-premiya": "nobel-prize",
+    "noize-mc": "noize-mc",
+    "nonkonformizm": "nonconformism",
+    "notforall": "notforall",
+    "novaya-drama": "new-drama",
+    "novosti": "news",
+    "noyz": "noise",
+    "nuar": "noir",
+    "oberiu": "oberiu",
+    "ocherk": "etudes",
+    "ochevidnyy-nuar": "ochevidnyy-nuar",
+    "odinochestvo": "loneliness",
+    "odna-kniga-odna-istoriya": "one-book-one-story",
+    "okrainy": "outskirts",
+    "omon": "swat",
+    "opinions": "opinions",
+    "oppozitsiya": "opposition",
+    "orhan-pamuk": "orhan-pamuk",
+    "ornitologiya": "ornitology",
+    "osen": "autumn",
+    "osip-mandelshtam": "osip-mandelshtam",
+    "oskar-uayld": "oscar-wilde",
+    "osoznanie": "awareness",
+    "otnosheniya": "relationship",
+    "pablo-pikasso": "pablo-picasso",
+    "painting": "painting",
+    "paintings": "painting",
+    "pamyat": "memory",
+    "pandemiya": "pandemic",
+    "parizh": "paris",
+    "patriotizm": "patriotism",
+    "patsifizm": "pacifism",
+    "paul-tselan": "paul-tselan",
+    "per-burd": "pierre-bourdieu",
+    "perezhivaniya": "worries",
+    "performance": "performance",
+    "peyzazh": "landscape",
+    "philology": "philology",
+    "philosophy": "philosophy",
+    "photo": "photography",
+    "photography": "photography",
+    "photoprojects": "photoprojects",
+    "plakaty": "posters",
+    "plastilin": "plasticine",
+    "plays": "plays",
+    "podrostki": "teenagers",
+    "poema": "poem",
+    "poems": "poems",
+    "poeticheskaya-proza": "poetic-prose",
+    "poetry": "poetry",
+    "poetry-of-squares": "poetry-of-squares",
+    "poetry-slam": "poetry-slam",
+    "pokoy": "peace",
+    "police": "police",
+    "politicheskoe-fentezi": "political-fantasy",
+    "politics": "politics",
+    "politzaklyuchennye": "political-prisoners",
+    "polsha": "poland",
+    "pomosch": "help",
+    "pop-art": "pop-art",
+    "pop-culture": "pop-culture",
+    "populyarnaya-psihologiya": "popular-psychology",
+    "pornografiya": "pornography",
+    "portret": "portrait",
+    "poslovitsy": "proverbs",
+    "post-pank": "post-punk",
+    "post-rok": "post-rock",
+    "postmodernism": "postmodernism",
+    "povest": "novells",
+    "povsednevnost": "everyday-life",
+    "power": "power",
+    "pravo": "right",
+    "pravoslavie": "orthodox",
+    "pravozaschitniki": "human-rights-activism",
+    "prazdnik": "holidays",
+    "predatelstvo": "betrayal",
+    "predprinimatelstvo": "entrepreneurship",
+    "premera": "premier",
+    "premiya-oskar": "oscar-prize",
+    "pribaltika-1": "baltic",
+    "priroda": "nature",
+    "prison": "prison",
+    "pritcha": "parable",
+    "privatnost": "privacy",
+    "progress": "progress",
+    "projects": "projects",
+    "prokrastinatsiya": "procrastination",
+    "propaganda": "propaganda",
+    "proschenie": "forgiveness",
+    "prose": "prose",
+    "proshloe": "past",
+    "prostitutsiya": "prostitution",
+    "prosveschenie": "enlightenment",
+    "protests": "protests",
+    "psalmy": "psalms",
+    "psihoanaliz": "psychoanalysis",
+    "psihodeliki": "psychodelics",
+    "pskov": "pskov",
+    "psychiatry": "psychiatry",
+    "psychology": "psychology",
+    "ptitsy": "birds",
+    "punk": "punk",
+    "r-b": "rnb",
+    "rasizm": "racism",
+    "realizm": "realism",
+    "redaktura": "editing",
+    "refleksiya": "reflection",
+    "reggi": "reggae",
+    "religion": "religion",
+    "rene-zhirar": "rene-girard",
+    "renesanss": "renessance",
+    "renovatsiya": "renovation",
+    "rep": "rap",
+    "reportage": "reportage",
+    "reportazh-1": "reportage",
+    "repressions": "repressions",
+    "research": "research",
+    "retroveyv": "retrowave",
+    "review": "review",
+    "revolution": "revolution",
+    "rezo-gabriadze": "rezo-gabriadze",
+    "risunki": "painting",
+    "roboty": "robots",
+    "rock": "rock",
+    "roditeli": "parents",
+    "romantizm": "romantism",
+    "romany": "novell",
+    "ronald-reygan": "ronald-reygan",
+    "roskomnadzor": "roskomnadzor",
+    "rossiyskoe-kino": "russian-cinema",
+    "rouling": "rowling",
+    "rozhava": "rojava",
+    "rpts": "rpts",
+    "rus-na-grani-sryva": "rus-na-grani-sryva",
+    "russia": "russia",
+    "russian-language": "russian-language",
+    "russian-literature": "russian-literature",
+    "russkaya-toska": "russian-toska",
+    "russkiy-mir": "russkiy-mir",
+    "salo": "lard",
+    "salvador-dali": "salvador-dali",
+    "samoidentifikatsiya": "self-identity",
+    "samoopredelenie": "self-definition",
+    "sankt-peterburg": "saint-petersburg",
+    "sasha-skochilenko": "sasha-skochilenko",
+    "satira": "satiric",
+    "saund-art": "sound-art",
+    "schaste": "happiness",
+    "school": "school",
+    "science": "science",
+    "sculpture": "sculpture",
+    "second-world-war": "second-world-war",
+    "sekond-hend": "second-hand",
+    "seksprosvet": "sex-education",
+    "seksualizirovannoe-nasilie": "sexualized-violence",
+    "seksualnoe-nasilie": "sexualized-violence",
+    "sekty": "sects",
+    "semi": "semi",
+    "semiotics": "semiotics",
+    "serbiya": "serbia",
+    "sergey-bodrov-mladshiy": "sergey-bodrov-junior",
+    "sergey-solov-v": "sergey-solovyov",
+    "serialy": "series",
+    "sever": "north",
+    "severnaya-koreya": "north-korea",
+    "sex": "sex",
+    "shotlandiya": "scotland",
+    "shugeyz": "shoegaze",
+    "siloviki": "siloviki",
+    "simeon-bekbulatovich": "simeon-bekbulatovich",
+    "simvolizm": "simbolism",
+    "siriya": "siria",
+    "skulptura": "sculpture",
+    "slavoy-zhizhek": "slavoj-zizek",
+    "smert-1": "death",
+    "smysl": "meaning",
+    "sny": "dreams",
+    "sobytiya": "events",
+    "social": "society",
+    "society": "society",
+    "sociology": "sociology",
+    "sofya-paleolog": "sofya-paleolog",
+    "sofya-vitovtovna": "sofya-vitovtovna",
+    "soobschestva": "communities",
+    "soprotivlenie": "resistence",
+    "sotsializm": "socialism",
+    "sotsialnaya-filosofiya": "social-philosophy",
+    "sotsiologiya-1": "sociology",
+    "sotsseti": "social-networks",
+    "sotvorenie-tretego-rima": "third-rome",
+    "sovremennost": "modernity",
+    "spaces": "spaces",
+    "spektakl": "spectacles",
+    "spetseffekty": "special-fx",
+    "spetsoperatsiya": "special-operation",
+    "spetssluzhby": "special-services",
+    "sport": "sport",
+    "srednevekove": "middle-age",
+    "state": "state",
+    "statistika": "statistics",
+    "stendap": "stand-up",
+    "stihi": "poetry",
+    "stoitsizm": "stoicism",
+    "stories": "stories",
+    "stoyanie-na-ugre": "stoyanie-na-ugre",
+    "strah": "fear",
+    "street-art": "street-art",
+    "stsenarii": "scenarios",
+    "sud": "court",
+    "summary": "summary",
+    "supergeroi": "superheroes",
+    "svetlana-aleksievich": "svetlana-aleksievich",
+    "svobodu-ivanu-golunovu": "free-ivan-golunov",
+    "syurrealizm": "surrealism",
+    "tales": "tales",
+    "tanets": "dance",
+    "tataro-mongolskoe-igo": "mongol-tatar-yoke",
+    "tatuirovki": "tattoo",
+    "technology": "technology",
+    "televidenie": "television",
+    "telo": "body",
+    "telo-kak-iskusstvo": "body-as-art",
+    "terrorizm": "terrorism",
+    "tests": "tests",
+    "text": "texts",
+    "the-beatles": "the-beatles",
+    "theater": "theater",
+    "theory": "theory",
+    "tokio": "tokio",
+    "torture": "torture",
+    "totalitarizm": "totalitarism",
+    "traditions": "traditions",
+    "tragicomedy": "tragicomedy",
+    "transgendernost": "transgender",
+    "translation": "translation",
+    "transport": "transport",
+    "travel": "travel",
+    "travma": "trauma",
+    "trendy": "trends",
+    "tretiy-reyh": "third-reich",
+    "triller": "thriller",
+    "tsar": "central-african-republic",
+    "tsar-edip": "oedipus",
+    "tsarevich-dmitriy": "tsarevich-dmitry",
+    "tsennosti": "values",
+    "tsenzura": "censorship",
+    "tseremonii": "ceremonies",
+    "turizm": "tourism",
+    "tvorchestvo": "creativity",
+    "ugnetennyy-zhilischnyy-klass": "oppressed-housing-class",
+    "uilyam-shekspir": "william-shakespeare",
+    "ukraina-2": "ukraine",
+    "ukraine": "ukraine",
+    "university": "university",
+    "urban-studies": "urban-studies",
+    "uroki-literatury": "literature-lessons",
+    "usa": "usa",
+    "ussr": "ussr",
+    "utopiya": "utopia",
+    "utrata": "loss",
+    "valter-benyamin": "valter-benyamin",
+    "varlam-shalamov": "varlam-shalamov",
+    "vasiliy-ii-temnyy": "basil-ii-temnyy",
+    "vasiliy-iii": "basil-iii",
+    "vdnh": "vdnh",
+    "vechnost": "ethernety",
+    "velikobritaniya": "great-britain",
+    "velimir-hlebnikov": "velimir-hlebnikov",
+    "velkom-tu-greyt-britn": "welcome-to-great-britain",
+    "venedikt-erofeev": "venedikt-erofeev",
+    "venetsiya": "veneece",
+    "vengriya": "hungary",
+    "verlibry": "free-verse",
+    "veschi": "things",
+    "vessels": "vessels",
+    "veterany": "veterans",
+    "video": "video",
+    "videoart": "videoart",
+    "videoklip": "clips",
+    "videopoeziya": "video-poetry",
+    "viktor-astafev": "viktor-astafev",
+    "viktor-pelevin": "viktor-pelevin",
+    "vilgelm-rayh": "wilhelm-reich",
+    "vinzavod": "vinzavod",
+    "violence": "violence",
+    "visual-culture": "visual-culture",
+    "vizualnaya-poeziya": "visual-poetry",
+    "vladimir-lenin": "vladimir-lenin",
+    "vladimir-mayakovskiy": "vladimir-mayakovsky",
+    "vladimir-nabokov": "vladimir-nabokov",
+    "vladimir-putin": "vladimir-putin",
+    "vladimir-sorokin": "vladimir-sorokin",
+    "vladimir-voynovich": "vladimir-voynovich",
+    "vnutrenniy-opyt": "inner-expirience",
+    "volga": "volga",
+    "volontery": "volonteurs",
+    "vong-karvay": "wong-karwai",
+    "vospominaniya": "memories",
+    "vostok": "east",
+    "voyna-na-ukraine": "war-in-ukraine",
+    "voyna-v-ukraine": "war-in-ukraine",
+    "vremya": "time",
+    "vudi-allen": "woody-allen",
+    "vynuzhdennye-otnosheniya": "forced-relationship",
+    "war": "war",
+    "war-in-ukraine-images": "war-in-ukrahine-images",
+    "women": "women",
+    "work": "work",
+    "writers": "writers",
+    "xx-century": "xx-century",
+    "yakob-yordans": "yakob-yordans",
+    "yan-vermeer": "yan-vermeer",
+    "yanka-dyagileva": "yanka-dyagileva",
+    "yaponskaya-literatura": "japan-literature",
+    "yazychestvo": "paganism",
+    "youth": "youth",
+    "yozef-rot": "yozef-rot",
+    "yurgen-habermas": "jorgen-habermas",
+    "za-liniey-mannergeyma": "behind-mannerheim-line",
+    "zabota": "care",
+    "zahar-prilepin": "zahar-prilepin",
+    "zakonodatelstvo": "laws",
+    "zakony-mira": "world-laws",
+    "zametki": "notes",
+    "zhelanie": "wish",
+    "zhivotnye": "animals",
+    "zhoze-saramago": "jose-saramago",
+    "zigmund-freyd": "sigmund-freud",
+    "zolotaya-orda": "golden-horde",
+    "zombi": "zombie",
+    "zombi-simpsony": "zombie-simpsons"
+}
--- a/migration/tables/topics.py
+++ b/migration/tables/topics.py
@@ -0,0 +1,31 @@
+from base.orm import local_session
+from migration.html2text import html2text
+from orm import Topic
+
+
+def migrate(entry):
+    body_orig = entry.get("description", "").replace("&nbsp;", " ")
+    topic_dict = {
+        "slug": entry["slug"],
+        "oid": entry["_id"],
+        "title": entry["title"].replace("&nbsp;", " "),
+        "body": html2text(body_orig),
+    }
+
+    with local_session() as session:
+        slug = topic_dict["slug"]
+        topic = session.query(Topic).filter(Topic.slug == slug).first() or Topic.create(
+            **topic_dict
+        )
+        if not topic:
+            raise Exception("no topic!")
+        if topic:
+            if len(topic.title) > len(topic_dict["title"]):
+                Topic.update(topic, {"title": topic_dict["title"]})
+            if len(topic.body) < len(topic_dict["body"]):
+                Topic.update(topic, {"body": topic_dict["body"]})
+            session.commit()
+    # print(topic.__dict__)
+    rt = topic.__dict__.copy()
+    del rt["_sa_instance_state"]
+    return rt
--- a/migration/tables/users.py
+++ b/migration/tables/users.py
@@ -0,0 +1,156 @@
+import re
+
+from bs4 import BeautifulSoup
+from dateutil.parser import parse
+from sqlalchemy.exc import IntegrityError
+
+from base.orm import local_session
+from orm.user import AuthorFollower, User, UserRating
+
+
+def migrate(entry):  # noqa: C901
+    if "subscribedTo" in entry:
+        del entry["subscribedTo"]
+    email = entry["emails"][0]["address"]
+    user_dict = {
+        "oid": entry["_id"],
+        "roles": [],
+        "ratings": [],
+        "username": email,
+        "email": email,
+        "createdAt": parse(entry["createdAt"]),
+        "emailConfirmed": ("@discours.io" in email) or bool(entry["emails"][0]["verified"]),
+        "muted": False,  # amnesty
+        "links": [],
+        "name": "anonymous",
+        "password": entry["services"]["password"].get("bcrypt"),
+    }
+
+    if "updatedAt" in entry:
+        user_dict["updatedAt"] = parse(entry["updatedAt"])
+    if "wasOnlineAt" in entry:
+        user_dict["lastSeen"] = parse(entry["wasOnlineAt"])
+    if entry.get("profile"):
+        # slug
+        slug = entry["profile"].get("path").lower()
+        slug = re.sub("[^0-9a-zA-Z]+", "-", slug).strip()
+        user_dict["slug"] = slug
+        bio = (
+            (entry.get("profile", {"bio": ""}).get("bio") or "")
+            .replace(r"\(", "(")
+            .replace(r"\)", ")")
+        )
+        bio_text = BeautifulSoup(bio, features="lxml").text
+
+        if len(bio_text) > 120:
+            user_dict["about"] = bio_text
+        else:
+            user_dict["bio"] = bio_text
+
+        # userpic
+        try:
+            user_dict["userpic"] = (
+                "https://images.discours.io/unsafe/" + entry["profile"]["thumborId"]
+            )
+        except KeyError:
+            try:
+                user_dict["userpic"] = entry["profile"]["image"]["url"]
+            except KeyError:
+                user_dict["userpic"] = ""
+
+        # name
+        fn = entry["profile"].get("firstName", "")
+        ln = entry["profile"].get("lastName", "")
+        name = fn if fn else ""
+        name = (name + " " + ln) if ln else name
+        if not name:
+            name = slug if slug else "anonymous"
+        name = entry["profile"]["path"].lower().strip().replace(" ", "-") if len(name) < 2 else name
+        user_dict["name"] = name
+
+        # links
+        fb = entry["profile"].get("facebook", False)
+        if fb:
+            user_dict["links"].append(fb)
+        vk = entry["profile"].get("vkontakte", False)
+        if vk:
+            user_dict["links"].append(vk)
+        tr = entry["profile"].get("twitter", False)
+        if tr:
+            user_dict["links"].append(tr)
+        ws = entry["profile"].get("website", False)
+        if ws:
+            user_dict["links"].append(ws)
+
+    # some checks
+    if not user_dict["slug"] and len(user_dict["links"]) > 0:
+        user_dict["slug"] = user_dict["links"][0].split("/")[-1]
+
+    user_dict["slug"] = user_dict.get("slug", user_dict["email"].split("@")[0])
+    oid = user_dict["oid"]
+    user_dict["slug"] = user_dict["slug"].lower().strip().replace(" ", "-")
+    try:
+        user = User.create(**user_dict.copy())
+    except IntegrityError:
+        print("[migration] cannot create user " + user_dict["slug"])
+        with local_session() as session:
+            old_user = session.query(User).filter(User.slug == user_dict["slug"]).first()
+            old_user.oid = oid
+            old_user.password = user_dict["password"]
+            session.commit()
+            user = old_user
+            if not user:
+                print("[migration] ERROR: cannot find user " + user_dict["slug"])
+                raise Exception
+    user_dict["id"] = user.id
+    return user_dict
+
+
+def post_migrate():
+    old_discours_dict = {
+        "slug": "old-discours",
+        "username": "old-discours",
+        "email": "old@discours.io",
+        "name": "Просмотры на старой версии сайта",
+    }
+
+    with local_session() as session:
+        old_discours_user = User.create(**old_discours_dict)
+        session.add(old_discours_user)
+        session.commit()
+
+
+def migrate_2stage(entry, id_map):
+    ce = 0
+    for rating_entry in entry.get("ratings", []):
+        rater_oid = rating_entry["createdBy"]
+        rater_slug = id_map.get(rater_oid)
+        if not rater_slug:
+            ce += 1
+            # print(rating_entry)
+            continue
+        oid = entry["_id"]
+        author_slug = id_map.get(oid)
+
+        with local_session() as session:
+            try:
+                rater = session.query(User).where(User.slug == rater_slug).one()
+                user = session.query(User).where(User.slug == author_slug).one()
+
+                user_rating_dict = {
+                    "value": rating_entry["value"],
+                    "rater": rater.id,
+                    "user": user.id,
+                }
+
+                user_rating = UserRating.create(**user_rating_dict)
+                if user_rating_dict["value"] > 0:
+                    af = AuthorFollower.create(author=user.id, follower=rater.id, auto=True)
+                    session.add(af)
+                session.add(user_rating)
+                session.commit()
+            except IntegrityError:
+                print("[migration] cannot rate " + author_slug + "`s by " + rater_slug)
+            except Exception as e:
+                print(e)
+    return ce
--- a/migration/utils.py
+++ b/migration/utils.py
@@ -0,0 +1,10 @@
+from datetime import datetime
+from json import JSONEncoder
+
+
+class DateTimeEncoder(JSONEncoder):
+    def default(self, z):
+        if isinstance(z, datetime):
+            return str(z)
+        else:
+            return super().default(z)
--- a/nginx.conf.sigil
+++ b/nginx.conf.sigil
@@ -1,14 +1,9 @@
-log_format custom '$remote_addr - $remote_user [$time_local] "$request" '
-                  'origin=$http_origin allow_origin=$allow_origin status=$status '
-                  '"$http_referer" "$http_user_agent"';
-
 {{ $proxy_settings := "proxy_http_version 1.1; proxy_set_header Upgrade $http_upgrade; proxy_set_header Connection $http_connection; proxy_set_header Host $http_host; proxy_set_header X-Request-Start $msec;" }}
 {{ $gzip_settings := "gzip on; gzip_min_length  1100; gzip_buffers  4 32k; gzip_types text/css text/javascript text/xml text/plain text/x-component application/javascript application/x-javascript application/json application/xml application/rss+xml font/truetype application/x-font-ttf font/opentype application/vnd.ms-fontobject image/svg+xml; gzip_vary on; gzip_comp_level  6;" }}

-proxy_cache_path /var/cache/nginx levels=1:2 keys_zone=my_cache:10m max_size=1g
-                 inactive=60m use_temp_path=off;
-limit_conn_zone $binary_remote_addr zone=addr:10m;
-limit_req_zone $binary_remote_addr zone=req_zone:10m rate=20r/s;
+{{ $cors_headers_options := "if ($request_method = 'OPTIONS') { add_header 'Access-Control-Allow-Origin' '$allow_origin' always; add_header 'Access-Control-Allow-Methods' 'GET, POST, OPTIONS'; add_header 'Access-Control-Allow-Headers' 'DNT,User-Agent,X-Requested-With,If-Modified-Since,Cache-Control,Content-Type,Range,Authorization'; add_header 'Access-Control-Allow-Credentials' 'true'; add_header 'Access-Control-Max-Age' 1728000; add_header 'Content-Type' 'text/plain; charset=utf-8'; add_header 'Content-Length' 0; return 204; }" }}
+{{ $cors_headers_post := "if ($request_method = 'POST') { add_header 'Access-Control-Allow-Origin' '$allow_origin' always; add_header 'Access-Control-Allow-Methods' 'GET, POST, OPTIONS' always; add_header 'Access-Control-Allow-Headers' 'DNT,User-Agent,X-Requested-With,If-Modified-Since,Cache-Control,Content-Type,Range,Authorization' always; add_header 'Access-Control-Expose-Headers' 'Content-Length,Content-Range' always; add_header 'Access-Control-Allow-Credentials' 'true' always; }" }}
+{{ $cors_headers_get := "if ($request_method = 'GET') { add_header 'Access-Control-Allow-Origin' '$allow_origin' always; add_header 'Access-Control-Allow-Methods' 'GET, POST, OPTIONS' always; add_header 'Access-Control-Allow-Headers' 'DNT,User-Agent,X-Requested-With,If-Modified-Since,Cache-Control,Content-Type,Range,Authorization' always; add_header 'Access-Control-Expose-Headers' 'Content-Length,Content-Range' always; add_header 'Access-Control-Allow-Credentials' 'true' always; }" }}

 {{ range $port_map := .PROXY_PORT_MAP | split " " }}
 {{ $port_map_list := $port_map | split ":" }}
@@ -21,15 +16,14 @@ server {
        listen [::]:{{ $listen_port }};
        listen {{ $listen_port }};
        server_name {{ $.NOSSL_SERVER_NAME }};
-        access_log  /var/log/nginx/{{ $.APP }}-access.log custom;
+        access_log  /var/log/nginx/{{ $.APP }}-access.log;
        error_log   /var/log/nginx/{{ $.APP }}-error.log;
-        client_max_body_size 100M;

    {{ else if eq $scheme "https" }}
        listen [::]:{{ $listen_port }} ssl http2;
        listen {{ $listen_port }} ssl http2;
        server_name {{ $.NOSSL_SERVER_NAME }};
-        access_log  /var/log/nginx/{{ $.APP }}-access.log custom;
+        access_log  /var/log/nginx/{{ $.APP }}-access.log;
        error_log   /var/log/nginx/{{ $.APP }}-error.log;
        ssl_certificate {{ $.APP_SSL_PATH }}/server.crt;
        ssl_certificate_key {{ $.APP_SSL_PATH }}/server.key;
@@ -37,72 +31,21 @@ server {
        ssl_prefer_server_ciphers off;

        keepalive_timeout   70;
-        keepalive_requests 500;
-        proxy_read_timeout 3600;
-        limit_conn addr 10000;
-        client_max_body_size 100M;
    {{ end }}


    location / {
-        proxy_pass http://{{ $.APP }}-{{ $upstream_port }};
+        proxy_pass  http://{{ $.APP }}-{{ $upstream_port }};
        {{ $proxy_settings }}
        {{ $gzip_settings }}
-
-        # Handle CORS for OPTIONS method
-        if ($request_method = 'OPTIONS') {
-            add_header 'Access-Control-Allow-Origin' $allow_origin always;
-            add_header 'Access-Control-Allow-Methods' 'POST, GET, OPTIONS';
-            add_header 'Access-Control-Allow-Headers' 'Content-Type, Authorization' always;
-            add_header 'Access-Control-Allow-Credentials' 'true' always;
-            add_header 'Access-Control-Max-Age' 1728000;
-            add_header 'Content-Type' 'text/plain; charset=utf-8';
-            add_header 'Content-Length' 0;
-            return 204;
-        }
-
-        # Handle CORS for POST method
-        if ($request_method = 'POST') {
-            add_header 'Access-Control-Allow-Origin' $allow_origin always;
-            add_header 'Access-Control-Allow-Methods' 'POST, GET, OPTIONS' always;
-            add_header 'Access-Control-Allow-Headers' 'Content-Type, Authorization' always;
-            add_header 'Access-Control-Allow-Credentials' 'true' always;
-        }
-
-        # Handle CORS for GET method
-        if ($request_method = 'GET') {
-            add_header 'Access-Control-Allow-Origin' $allow_origin always;
-            add_header 'Access-Control-Allow-Methods' 'POST, GET, OPTIONS' always;
-            add_header 'Access-Control-Allow-Headers' 'Content-Type, Authorization' always;
-            add_header 'Access-Control-Allow-Credentials' 'true' always;
-        }
-
-        proxy_cache my_cache;
-        proxy_cache_revalidate on;
-        proxy_cache_min_uses 2;
-        proxy_cache_use_stale error timeout updating http_500 http_502 http_503 http_504;
-        proxy_cache_background_update on;
-        proxy_cache_lock on;
-
-        # Connections and request limits increase (bad for DDos)
-        limit_req zone=req_zone burst=10 nodelay;
+        {{ $cors_headers_options }}
+        {{ $cors_headers_post }}
+        {{ $cors_headers_get }}
    }

    location ~* \.(jpg|jpeg|png|gif|ico|css|js)$ {
-        proxy_pass  http://{{ $.APP }}-{{ $upstream_port }};
-        expires 30d;
-        add_header Cache-Control "public, no-transform";
-    }
-
-    location ~* \.(mp3|wav|ogg|flac|aac|aif|webm)$ {
-      proxy_pass  http://{{ $.APP }}-{{ $upstream_port }};
-      if ($request_method = 'GET') {
-        add_header 'Access-Control-Allow-Origin' $allow_origin always;
-        add_header 'Access-Control-Allow-Methods' 'GET, POST, OPTIONS' always;
-        add_header 'Access-Control-Allow-Headers' 'DNT,User-Agent,X-Requested-With,If-Modified-Since,Cache-Control,Content-Type,Range,Authorization' always;
-        add_header 'Access-Control-Expose-Headers' 'Content-Length,Content-Range' always;
-        add_header 'Access-Control-Allow-Credentials' 'true' always;
-      }
+    expires 30d;  # This means that the client can cache these resources for 30 days.
+    add_header Cache-Control "public, no-transform";
    }


@@ -130,6 +73,7 @@ server {
    internal;
  }

+  # include /home/dokku/gateway/nginx.conf.d/*.conf;
  include {{ $.DOKKU_ROOT }}/{{ $.APP }}/nginx.conf.d/*.conf;
 }
 {{ end }}
--- a/orm/init.py
+++ b/orm/init.py
@@ -0,0 +1,36 @@
+from base.orm import Base, engine
+from orm.community import Community
+from orm.notification import Notification
+from orm.rbac import Operation, Permission, Resource, Role
+from orm.reaction import Reaction
+from orm.shout import Shout
+from orm.topic import Topic, TopicFollower
+from orm.user import User, UserRating
+
+
+def init_tables():
+    Base.metadata.create_all(engine)
+    Operation.init_table()
+    Resource.init_table()
+    User.init_table()
+    Community.init_table()
+    Role.init_table()
+    UserRating.init_table()
+    Shout.init_table()
+    print("[orm] tables initialized")
+
+
+__all__ = [
+    "User",
+    "Role",
+    "Operation",
+    "Permission",
+    "Community",
+    "Shout",
+    "Topic",
+    "TopicFollower",
+    "Notification",
+    "Reaction",
+    "UserRating",
+    "init_tables",
+]
--- a/orm/author.py
+++ b/orm/author.py
@@ -1,137 +0,0 @@
-import time
-
-from sqlalchemy import JSON, Boolean, Column, ForeignKey, Index, Integer, String
-
-from services.db import Base
-
-# from sqlalchemy_utils import TSVectorType
-
-
-class AuthorRating(Base):
-    """
-    Рейтинг автора от другого автора.
-
-    Attributes:
-        rater (int): ID оценивающего автора
-        author (int): ID оцениваемого автора
-        plus (bool): Положительная/отрицательная оценка
-    """
-
-    __tablename__ = "author_rating"
-
-    id = None  # type: ignore
-    rater = Column(ForeignKey("author.id"), primary_key=True)
-    author = Column(ForeignKey("author.id"), primary_key=True)
-    plus = Column(Boolean)
-
-    # Определяем индексы
-    __table_args__ = (
-        # Индекс для быстрого поиска всех оценок конкретного автора
-        Index("idx_author_rating_author", "author"),
-        # Индекс для быстрого поиска всех оценок, оставленных конкретным автором
-        Index("idx_author_rating_rater", "rater"),
-    )
-
-
-class AuthorFollower(Base):
-    """
-    Подписка одного автора на другого.
-
-    Attributes:
-        follower (int): ID подписчика
-        author (int): ID автора, на которого подписываются
-        created_at (int): Время создания подписки
-        auto (bool): Признак автоматической подписки
-    """
-
-    __tablename__ = "author_follower"
-
-    id = None  # type: ignore
-    follower = Column(ForeignKey("author.id"), primary_key=True)
-    author = Column(ForeignKey("author.id"), primary_key=True)
-    created_at = Column(Integer, nullable=False, default=lambda: int(time.time()))
-    auto = Column(Boolean, nullable=False, default=False)
-
-    # Определяем индексы
-    __table_args__ = (
-        # Индекс для быстрого поиска всех подписчиков автора
-        Index("idx_author_follower_author", "author"),
-        # Индекс для быстрого поиска всех авторов, на которых подписан конкретный автор
-        Index("idx_author_follower_follower", "follower"),
-    )
-
-
-class AuthorBookmark(Base):
-    """
-    Закладка автора на публикацию.
-
-    Attributes:
-        author (int): ID автора
-        shout (int): ID публикации
-    """
-
-    __tablename__ = "author_bookmark"
-
-    id = None  # type: ignore
-    author = Column(ForeignKey("author.id"), primary_key=True)
-    shout = Column(ForeignKey("shout.id"), primary_key=True)
-
-    # Определяем индексы
-    __table_args__ = (
-        # Индекс для быстрого поиска всех закладок автора
-        Index("idx_author_bookmark_author", "author"),
-        # Индекс для быстрого поиска всех авторов, добавивших публикацию в закладки
-        Index("idx_author_bookmark_shout", "shout"),
-    )
-
-
-class Author(Base):
-    """
-    Модель автора в системе.
-
-    Attributes:
-        user (str): Идентификатор пользователя в системе авторизации
-        name (str): Отображаемое имя
-        slug (str): Уникальный строковый идентификатор
-        bio (str): Краткая биография/статус
-        about (str): Полное описание
-        pic (str): URL изображения профиля
-        links (dict): Ссылки на социальные сети и сайты
-        created_at (int): Время создания профиля
-        last_seen (int): Время последнего посещения
-        updated_at (int): Время последнего обновления
-        deleted_at (int): Время удаления (если профиль удален)
-    """
-
-    __tablename__ = "author"
-
-    user = Column(String)  # unbounded link with authorizer's User type
-
-    name = Column(String, nullable=True, comment="Display name")
-    slug = Column(String, unique=True, comment="Author's slug")
-    bio = Column(String, nullable=True, comment="Bio")  # status description
-    about = Column(String, nullable=True, comment="About")  # long and formatted
-    pic = Column(String, nullable=True, comment="Picture")
-    links = Column(JSON, nullable=True, comment="Links")
-    created_at = Column(Integer, nullable=False, default=lambda: int(time.time()))
-    last_seen = Column(Integer, nullable=False, default=lambda: int(time.time()))
-    updated_at = Column(Integer, nullable=False, default=lambda: int(time.time()))
-    deleted_at = Column(Integer, nullable=True, comment="Deleted at")
-
-    # search_vector = Column(
-    #    TSVectorType("name", "slug", "bio", "about", regconfig="pg_catalog.russian")
-    # )
-
-    # Определяем индексы
-    __table_args__ = (
-        # Индекс для быстрого поиска по slug
-        Index("idx_author_slug", "slug"),
-        # Индекс для быстрого поиска по идентификатору пользователя
-        Index("idx_author_user", "user"),
-        # Индекс для фильтрации неудаленных авторов
-        Index("idx_author_deleted_at", "deleted_at", postgresql_where=deleted_at.is_(None)),
-        # Индекс для сортировки по времени создания (для новых авторов)
-        Index("idx_author_created_at", "created_at"),
-        # Индекс для сортировки по времени последнего посещения
-        Index("idx_author_last_seen", "last_seen"),
-    )
--- a/orm/collection.py
+++ b/orm/collection.py
@@ -1,14 +1,12 @@
-import time
+from sqlalchemy import Column, DateTime, ForeignKey, String, func

-from sqlalchemy import Column, ForeignKey, Integer, String
-
-from services.db import Base
+from base.orm import Base


 class ShoutCollection(Base):
    __tablename__ = "shout_collection"

-    id = None  # type: ignore
+    id = None
    shout = Column(ForeignKey("shout.id"), primary_key=True)
    collection = Column(ForeignKey("collection.id"), primary_key=True)

@@ -20,6 +18,6 @@ class Collection(Base):
    title = Column(String, nullable=False, comment="Title")
    body = Column(String, nullable=True, comment="Body")
    pic = Column(String, nullable=True, comment="Picture")
-    created_at = Column(Integer, default=lambda: int(time.time()))
-    created_by = Column(ForeignKey("author.id"), comment="Created By")
-    published_at = Column(Integer, default=lambda: int(time.time()))
+    createdAt = Column(DateTime(timezone=True), server_default=func.now(), comment="Created At")
+    createdBy = Column(ForeignKey("user.id"), comment="Created By")
+    publishedAt = Column(DateTime(timezone=True), server_default=func.now(), comment="Published At")
--- a/orm/community.py
+++ b/orm/community.py
@@ -1,106 +1,38 @@
-import enum
-import time
+from sqlalchemy import Column, DateTime, ForeignKey, String, func

-from sqlalchemy import Column, ForeignKey, Integer, String, Text, distinct, func
-from sqlalchemy.ext.hybrid import hybrid_property
-
-from orm.author import Author
-from services.db import Base
-
-
-class CommunityRole(enum.Enum):
-    READER = "reader"  # can read and comment
-    AUTHOR = "author"  # + can vote and invite collaborators
-    ARTIST = "artist"  # + can be credited as featured artist
-    EXPERT = "expert"  # + can add proof or disproof to shouts, can manage topics
-    EDITOR = "editor"  # + can manage topics, comments and community settings
-
-    @classmethod
-    def as_string_array(cls, roles):
-        return [role.value for role in roles]
+from base.orm import Base, local_session


 class CommunityFollower(Base):
-    __tablename__ = "community_author"
+    __tablename__ = "community_followers"

-    author = Column(ForeignKey("author.id"), primary_key=True)
-    community = Column(ForeignKey("community.id"), primary_key=True)
-    joined_at = Column(Integer, nullable=False, default=lambda: int(time.time()))
-    roles = Column(Text, nullable=True, comment="Roles (comma-separated)")
-
-    def set_roles(self, roles):
-        self.roles = CommunityRole.as_string_array(roles)
-
-    def get_roles(self):
-        return [CommunityRole(role) for role in self.roles]
+    id = None
+    follower: Column = Column(ForeignKey("user.id"), primary_key=True)
+    community: Column = Column(ForeignKey("community.id"), primary_key=True)
+    joinedAt = Column(
+        DateTime(timezone=True), nullable=False, server_default=func.now(), comment="Created at"
+    )
+    # role = Column(ForeignKey(Role.id), nullable=False, comment="Role for member")


 class Community(Base):
    __tablename__ = "community"

-    name = Column(String, nullable=False)
-    slug = Column(String, nullable=False, unique=True)
+    name = Column(String, nullable=False, comment="Name")
+    slug = Column(String, nullable=False, unique=True, comment="Slug")
    desc = Column(String, nullable=False, default="")
    pic = Column(String, nullable=False, default="")
-    created_at = Column(Integer, nullable=False, default=lambda: int(time.time()))
-    created_by = Column(ForeignKey("author.id"), nullable=False)
+    createdAt = Column(
+        DateTime(timezone=True), nullable=False, server_default=func.now(), comment="Created at"
+    )

-    @hybrid_property
-    def stat(self):
-        return CommunityStats(self)
-
-    @property
-    def role_list(self):
-        return self.roles.split(",") if self.roles else []
-
-    @role_list.setter
-    def role_list(self, value):
-        self.roles = ",".join(value) if value else None
-
-
-class CommunityStats:
-    def __init__(self, community):
-        self.community = community
-
-    @property
-    def shouts(self):
-        from orm.shout import Shout
-
-        return self.community.session.query(func.count(Shout.id)).filter(Shout.community == self.community.id).scalar()
-
-    @property
-    def followers(self):
-        return (
-            self.community.session.query(func.count(CommunityFollower.author))
-            .filter(CommunityFollower.community == self.community.id)
-            .scalar()
-        )
-
-    @property
-    def authors(self):
-        from orm.shout import Shout
-
-        # author has a shout with community id and its featured_at is not null
-        return (
-            self.community.session.query(func.count(distinct(Author.id)))
-            .join(Shout)
-            .filter(Shout.community == self.community.id, Shout.featured_at.is_not(None), Author.id.in_(Shout.authors))
-            .scalar()
-        )
-
-
-class CommunityAuthor(Base):
-    __tablename__ = "community_author"
-
-    id = Column(Integer, primary_key=True)
-    community_id = Column(Integer, ForeignKey("community.id"))
-    author_id = Column(Integer, ForeignKey("author.id"))
-    roles = Column(Text, nullable=True, comment="Roles (comma-separated)")
-
-    @property
-    def role_list(self):
-        return self.roles.split(",") if self.roles else []
-
-    @role_list.setter
-    def role_list(self, value):
-        self.roles = ",".join(value) if value else None
+    @staticmethod
+    def init_table():
+        with local_session() as session:
+            d = session.query(Community).filter(Community.slug == "discours").first()
+            if not d:
+                d = Community.create(name="Дискурс", slug="discours")
+                session.add(d)
+                session.commit()
+            Community.default_community = d
+            print("[orm] default community id: %s" % d.id)
--- a/orm/draft.py
+++ b/orm/draft.py
@@ -1,106 +0,0 @@
-import time
-
-from sqlalchemy import JSON, Boolean, Column, ForeignKey, Integer, String
-from sqlalchemy.orm import relationship
-
-from orm.author import Author
-from orm.topic import Topic
-from services.db import Base
-from orm.shout import Shout
-
-
-class DraftTopic(Base):
-    __tablename__ = "draft_topic"
-
-    id = None  # type: ignore
-    shout = Column(ForeignKey("draft.id"), primary_key=True, index=True)
-    topic = Column(ForeignKey("topic.id"), primary_key=True, index=True)
-    main = Column(Boolean, nullable=True)
-
-
-class DraftAuthor(Base):
-    __tablename__ = "draft_author"
-
-    id = None  # type: ignore
-    shout = Column(ForeignKey("draft.id"), primary_key=True, index=True)
-    author = Column(ForeignKey("author.id"), primary_key=True, index=True)
-    caption = Column(String, nullable=True, default="")
-
-
-
-class Draft(Base):
-    __tablename__ = "draft"
-    # required
-    created_at: int = Column(Integer, nullable=False, default=lambda: int(time.time()))
-    # Колонки для связей с автором
-    created_by: int = Column("created_by", ForeignKey("author.id"), nullable=False)
-    community: int = Column("community", ForeignKey("community.id"), nullable=False, default=1)
-
-    # optional
-    layout: str = Column(String, nullable=True, default="article")
-    slug: str = Column(String, unique=True)
-    title: str = Column(String, nullable=True)
-    subtitle: str | None = Column(String, nullable=True)
-    lead: str | None = Column(String, nullable=True)
-    body: str = Column(String, nullable=False, comment="Body")
-    media: dict | None = Column(JSON, nullable=True)
-    cover: str | None = Column(String, nullable=True, comment="Cover image url")
-    cover_caption: str | None = Column(String, nullable=True, comment="Cover image alt caption")
-    lang: str = Column(String, nullable=False, default="ru", comment="Language")
-    seo: str | None = Column(String, nullable=True)  # JSON
-
-    # auto
-    updated_at: int | None = Column(Integer, nullable=True, index=True)
-    deleted_at: int | None = Column(Integer, nullable=True, index=True)
-    updated_by: int | None = Column("updated_by", ForeignKey("author.id"), nullable=True)
-    deleted_by: int | None = Column("deleted_by", ForeignKey("author.id"), nullable=True)
-    
-    # --- Relationships --- 
-    # Только many-to-many связи через вспомогательные таблицы
-    authors = relationship(Author, secondary="draft_author", lazy="select")
-    topics = relationship(Topic, secondary="draft_topic", lazy="select")
-    
-    # Связь с Community (если нужна как объект, а не ID)
-    # community = relationship("Community", foreign_keys=[community_id], lazy="joined")
-    # Пока оставляем community_id как ID
-
-    # Связь с публикацией (один-к-одному или один-к-нулю)
-    # Загружается через joinedload в резолвере
-    publication = relationship(
-        "Shout", 
-        primaryjoin="Draft.id == Shout.draft", 
-        foreign_keys="Shout.draft", 
-        uselist=False, 
-        lazy="noload", # Не грузим по умолчанию, только через options
-        viewonly=True # Указываем, что это связь только для чтения
-    )
-
-    def dict(self):
-        """
-        Сериализует объект Draft в словарь.
-        Гарантирует, что поля topics и authors всегда будут списками.
-        """
-        return {
-            "id": self.id,
-            "created_at": self.created_at,
-            "created_by": self.created_by,
-            "community": self.community,
-            "layout": self.layout,
-            "slug": self.slug,
-            "title": self.title,
-            "subtitle": self.subtitle,
-            "lead": self.lead,
-            "body": self.body,
-            "media": self.media or [],
-            "cover": self.cover,
-            "cover_caption": self.cover_caption,
-            "lang": self.lang,
-            "seo": self.seo,
-            "updated_at": self.updated_at,
-            "deleted_at": self.deleted_at,
-            "updated_by": self.updated_by,
-            "deleted_by": self.deleted_by,
-            # Гарантируем, что topics и authors всегда будут списками
-            "topics": [topic.dict() for topic in (self.topics or [])],
-            "authors": [author.dict() for author in (self.authors or [])]
-        }
--- a/orm/invite.py
+++ b/orm/invite.py
@@ -1,35 +0,0 @@
-import enum
-
-from sqlalchemy import Column, ForeignKey, String
-from sqlalchemy.orm import relationship
-
-from services.db import Base
-
-
-class InviteStatus(enum.Enum):
-    PENDING = "PENDING"
-    ACCEPTED = "ACCEPTED"
-    REJECTED = "REJECTED"
-
-    @classmethod
-    def from_string(cls, value):
-        return cls(value)
-
-
-class Invite(Base):
-    __tablename__ = "invite"
-
-    inviter_id = Column(ForeignKey("author.id"), primary_key=True)
-    author_id = Column(ForeignKey("author.id"), primary_key=True)
-    shout_id = Column(ForeignKey("shout.id"), primary_key=True)
-    status = Column(String, default=InviteStatus.PENDING.value)
-
-    inviter = relationship("Author", foreign_keys=[inviter_id])
-    author = relationship("Author", foreign_keys=[author_id])
-    shout = relationship("Shout")
-
-    def set_status(self, status: InviteStatus):
-        self.status = status.value
-
-    def get_status(self) -> InviteStatus:
-        return InviteStatus.from_string(self.status)
--- a/orm/notification.py
+++ b/orm/notification.py
@@ -1,63 +1,26 @@
-import enum
-import time
+from enum import Enum as Enumeration

-from sqlalchemy import JSON, Column, ForeignKey, Integer, String
-from sqlalchemy.orm import relationship
+from sqlalchemy import Boolean, Column, DateTime, Enum, ForeignKey, Integer, func
+from sqlalchemy.dialects.postgresql import JSONB

-from orm.author import Author
-from services.db import Base
+from base.orm import Base


-class NotificationEntity(enum.Enum):
-    REACTION = "reaction"
-    SHOUT = "shout"
-    FOLLOWER = "follower"
-    COMMUNITY = "community"
-
-    @classmethod
-    def from_string(cls, value):
-        return cls(value)
-
-
-class NotificationAction(enum.Enum):
-    CREATE = "create"
-    UPDATE = "update"
-    DELETE = "delete"
-    SEEN = "seen"
-    FOLLOW = "follow"
-    UNFOLLOW = "unfollow"
-
-    @classmethod
-    def from_string(cls, value):
-        return cls(value)
-
-
-class NotificationSeen(Base):
-    __tablename__ = "notification_seen"
-
-    viewer = Column(ForeignKey("author.id"), primary_key=True)
-    notification = Column(ForeignKey("notification.id"), primary_key=True)
+class NotificationType(Enumeration):
+    NEW_COMMENT = 1
+    NEW_REPLY = 2


 class Notification(Base):
    __tablename__ = "notification"

-    id = Column(Integer, primary_key=True, autoincrement=True)
-    created_at = Column(Integer, server_default=str(int(time.time())))
-    entity = Column(String, nullable=False)
-    action = Column(String, nullable=False)
-    payload = Column(JSON, nullable=True)
-
-    seen = relationship(Author, secondary="notification_seen")
-
-    def set_entity(self, entity: NotificationEntity):
-        self.entity = entity.value
-
-    def get_entity(self) -> NotificationEntity:
-        return NotificationEntity.from_string(self.entity)
-
-    def set_action(self, action: NotificationAction):
-        self.action = action.value
-
-    def get_action(self) -> NotificationAction:
-        return NotificationAction.from_string(self.action)
+    shout: Column = Column(ForeignKey("shout.id"), index=True)
+    reaction: Column = Column(ForeignKey("reaction.id"), index=True)
+    user: Column = Column(ForeignKey("user.id"), index=True)
+    createdAt = Column(
+        DateTime(timezone=True), nullable=False, server_default=func.now(), index=True
+    )
+    seen = Column(Boolean, nullable=False, default=False, index=True)
+    type = Column(Enum(NotificationType), nullable=False)
+    data = Column(JSONB, nullable=True)
+    occurrences = Column(Integer, default=1)
--- a/orm/rating.py
+++ b/orm/rating.py
@@ -1,30 +0,0 @@
-from orm.reaction import ReactionKind
-
-PROPOSAL_REACTIONS = [
-    ReactionKind.ACCEPT.value,
-    ReactionKind.REJECT.value,
-    ReactionKind.AGREE.value,
-    ReactionKind.DISAGREE.value,
-    ReactionKind.ASK.value,
-    ReactionKind.PROPOSE.value,
-]
-
-PROOF_REACTIONS = [ReactionKind.PROOF.value, ReactionKind.DISPROOF.value]
-
-RATING_REACTIONS = [ReactionKind.LIKE.value, ReactionKind.DISLIKE.value]
-
-
-def is_negative(x):
-    return x in [
-        ReactionKind.DISLIKE.value,
-        ReactionKind.DISPROOF.value,
-        ReactionKind.REJECT.value,
-    ]
-
-
-def is_positive(x):
-    return x in [
-        ReactionKind.ACCEPT.value,
-        ReactionKind.LIKE.value,
-        ReactionKind.PROOF.value,
-    ]
--- a/orm/rbac.py
+++ b/orm/rbac.py
@@ -0,0 +1,178 @@
+import warnings
+
+from sqlalchemy import Column, ForeignKey, String, TypeDecorator, UniqueConstraint
+from sqlalchemy.orm import relationship
+
+from base.orm import REGISTRY, Base, local_session
+
+# Role Based Access Control #
+
+
+class ClassType(TypeDecorator):
+    impl = String
+
+    @property
+    def python_type(self):
+        return NotImplemented
+
+    def process_literal_param(self, value, dialect):
+        return NotImplemented
+
+    def process_bind_param(self, value, dialect):
+        return value.__name__ if isinstance(value, type) else str(value)
+
+    def process_result_value(self, value, dialect):
+        class_ = REGISTRY.get(value)
+        if class_ is None:
+            warnings.warn(f"Can't find class <{value}>,find it yourself!", stacklevel=2)
+        return class_
+
+
+class Role(Base):
+    __tablename__ = "role"
+
+    name = Column(String, nullable=False, comment="Role Name")
+    desc = Column(String, nullable=True, comment="Role Description")
+    community = Column(
+        ForeignKey("community.id", ondelete="CASCADE"),
+        nullable=False,
+        comment="Community",
+    )
+    permissions = relationship(lambda: Permission)
+
+    @staticmethod
+    def init_table():
+        with local_session() as session:
+            r = session.query(Role).filter(Role.name == "author").first()
+            if r:
+                Role.default_role = r
+                return
+
+        r1 = Role.create(
+            name="author",
+            desc="Role for an author",
+            community=1,
+        )
+
+        session.add(r1)
+
+        Role.default_role = r1
+
+        r2 = Role.create(
+            name="reader",
+            desc="Role for a reader",
+            community=1,
+        )
+
+        session.add(r2)
+
+        r3 = Role.create(
+            name="expert",
+            desc="Role for an expert",
+            community=1,
+        )
+
+        session.add(r3)
+
+        r4 = Role.create(
+            name="editor",
+            desc="Role for an editor",
+            community=1,
+        )
+
+        session.add(r4)
+
+
+class Operation(Base):
+    __tablename__ = "operation"
+    name = Column(String, nullable=False, unique=True, comment="Operation Name")
+
+    @staticmethod
+    def init_table():
+        with local_session() as session:
+            for name in ["create", "update", "delete", "load"]:
+                """
+                * everyone can:
+                    - load shouts
+                    - load topics
+                    - load reactions
+                    - create an account to become a READER
+                * readers can:
+                    - update and delete their account
+                    - load chats
+                    - load messages
+                    - create reaction of some shout's author allowed kinds
+                    - create shout to become an AUTHOR
+                * authors can:
+                    - update and delete their shout
+                    - invite other authors to edit shout and chat
+                    - manage allowed reactions for their shout
+                * pros can:
+                    - create/update/delete their community
+                    - create/update/delete topics for their community
+
+                """
+                op = session.query(Operation).filter(Operation.name == name).first()
+                if not op:
+                    op = Operation.create(name=name)
+                    session.add(op)
+            session.commit()
+
+
+class Resource(Base):
+    __tablename__ = "resource"
+    resourceClass = Column(String, nullable=False, unique=True, comment="Resource class")
+    name = Column(String, nullable=False, unique=True, comment="Resource name")
+    # TODO: community = Column(ForeignKey())
+
+    @staticmethod
+    def init_table():
+        with local_session() as session:
+            for res in [
+                "shout",
+                "topic",
+                "reaction",
+                "chat",
+                "message",
+                "invite",
+                "community",
+                "user",
+            ]:
+                r = session.query(Resource).filter(Resource.name == res).first()
+                if not r:
+                    r = Resource.create(name=res, resourceClass=res)
+                    session.add(r)
+            session.commit()
+
+
+class Permission(Base):
+    __tablename__ = "permission"
+    __table_args__ = (
+        UniqueConstraint("role", "operation", "resource"),
+        {"extend_existing": True},
+    )
+
+    role: Column = Column(ForeignKey("role.id", ondelete="CASCADE"), nullable=False, comment="Role")
+    operation: Column = Column(
+        ForeignKey("operation.id", ondelete="CASCADE"),
+        nullable=False,
+        comment="Operation",
+    )
+    resource: Column = Column(
+        ForeignKey("resource.id", ondelete="CASCADE"),
+        nullable=False,
+        comment="Resource",
+    )
+
+
+# if __name__ == "__main__":
+#     Base.metadata.create_all(engine)
+#     ops = [
+#         Permission(role=1, operation=1, resource=1),
+#         Permission(role=1, operation=2, resource=1),
+#         Permission(role=1, operation=3, resource=1),
+#         Permission(role=1, operation=4, resource=1),
+#         Permission(role=2, operation=4, resource=1),
+#     ]
+#     global_session.add_all(ops)
+#     global_session.commit()
--- a/orm/reaction.py
+++ b/orm/reaction.py
@@ -1,45 +1,47 @@
-import time
 from enum import Enum as Enumeration

-from sqlalchemy import Column, ForeignKey, Integer, String
+from sqlalchemy import Column, DateTime, Enum, ForeignKey, String, func

-from services.db import Base
+from base.orm import Base


 class ReactionKind(Enumeration):
+    AGREE = 1  # +1
+    DISAGREE = 2  # -1
+    PROOF = 3  # +1
+    DISPROOF = 4  # -1
+    ASK = 5  # +0
+    PROPOSE = 6  # +0
+    QUOTE = 7  # +0 bookmark
+    COMMENT = 8  # +0
+    ACCEPT = 9  # +1
+    REJECT = 0  # -1
+    LIKE = 11  # +1
+    DISLIKE = 12  # -1
+    REMARK = 13  # 0
+    FOOTNOTE = 14  # 0
    # TYPE = <reaction index> # rating diff

-    # editor mode
-    AGREE = "AGREE"  # +1
-    DISAGREE = "DISAGREE"  # -1
-    ASK = "ASK"  # +0
-    PROPOSE = "PROPOSE"  # +0
-    ACCEPT = "ACCEPT"  # +1
-    REJECT = "REJECT"  # -1
-
-    # expert mode
-    PROOF = "PROOF"  # +1
-    DISPROOF = "DISPROOF"  # -1
-
-    # public feed
-    QUOTE = "QUOTE"  # +0 TODO: use to bookmark in collection
-    COMMENT = "COMMENT"  # +0
-    LIKE = "LIKE"  # +1
-    DISLIKE = "DISLIKE"  # -1
-

 class Reaction(Base):
    __tablename__ = "reaction"
-
-    body = Column(String, default="", comment="Reaction Body")
-    created_at = Column(Integer, nullable=False, default=lambda: int(time.time()), index=True)
-    updated_at = Column(Integer, nullable=True, comment="Updated at", index=True)
-    deleted_at = Column(Integer, nullable=True, comment="Deleted at", index=True)
-    deleted_by = Column(ForeignKey("author.id"), nullable=True)
-    reply_to = Column(ForeignKey("reaction.id"), nullable=True)
-    quote = Column(String, nullable=True, comment="Original quoted text")
-    shout = Column(ForeignKey("shout.id"), nullable=False, index=True)
-    created_by = Column(ForeignKey("author.id"), nullable=False)
-    kind = Column(String, nullable=False, index=True)
-
-    oid = Column(String)
+    body = Column(String, nullable=True, comment="Reaction Body")
+    createdAt = Column(
+        DateTime(timezone=True), nullable=False, server_default=func.now(), comment="Created at"
+    )
+    createdBy: Column = Column(ForeignKey("user.id"), nullable=False, index=True, comment="Sender")
+    updatedAt = Column(DateTime(timezone=True), nullable=True, comment="Updated at")
+    updatedBy: Column = Column(
+        ForeignKey("user.id"), nullable=True, index=True, comment="Last Editor"
+    )
+    deletedAt = Column(DateTime(timezone=True), nullable=True, comment="Deleted at")
+    deletedBy: Column = Column(
+        ForeignKey("user.id"), nullable=True, index=True, comment="Deleted by"
+    )
+    shout: Column = Column(ForeignKey("shout.id"), nullable=False, index=True)
+    replyTo: Column = Column(
+        ForeignKey("reaction.id"), nullable=True, comment="Reply to reaction ID"
+    )
+    range = Column(String, nullable=True, comment="Range in format <start index>:<end>")
+    kind = Column(Enum(ReactionKind), nullable=False, comment="Reaction kind")
+    oid = Column(String, nullable=True, comment="Old ID")
--- a/orm/shout.py
+++ b/orm/shout.py
@@ -1,126 +1,98 @@
-import time
+from sqlalchemy import (
+    JSON,
+    Boolean,
+    Column,
+    DateTime,
+    ForeignKey,
+    Integer,
+    String,
+    func,
+)
+from sqlalchemy.orm import column_property, relationship

-from sqlalchemy import JSON, Boolean, Column, ForeignKey, Index, Integer, String
-from sqlalchemy.orm import relationship
-
-from orm.author import Author
+from base.orm import Base, local_session
 from orm.reaction import Reaction
 from orm.topic import Topic
-from services.db import Base
+from orm.user import User


 class ShoutTopic(Base):
-    """
-    Связь между публикацией и темой.
-
-    Attributes:
-        shout (int): ID публикации
-        topic (int): ID темы
-        main (bool): Признак основной темы
-    """
-
    __tablename__ = "shout_topic"

-    id = None  # type: ignore
-    shout = Column(ForeignKey("shout.id"), primary_key=True, index=True)
-    topic = Column(ForeignKey("topic.id"), primary_key=True, index=True)
-    main = Column(Boolean, nullable=True)
-
-    # Определяем дополнительные индексы
-    __table_args__ = (
-        # Оптимизированный составной индекс для запросов, которые ищут публикации по теме
-        Index("idx_shout_topic_topic_shout", "topic", "shout"),
-    )
+    id = None
+    shout: Column = Column(ForeignKey("shout.id"), primary_key=True, index=True)
+    topic: Column = Column(ForeignKey("topic.id"), primary_key=True, index=True)


 class ShoutReactionsFollower(Base):
    __tablename__ = "shout_reactions_followers"

-    id = None  # type: ignore
-    follower = Column(ForeignKey("author.id"), primary_key=True, index=True)
-    shout = Column(ForeignKey("shout.id"), primary_key=True, index=True)
+    id = None
+    follower: Column = Column(ForeignKey("user.id"), primary_key=True, index=True)
+    shout: Column = Column(ForeignKey("shout.id"), primary_key=True, index=True)
    auto = Column(Boolean, nullable=False, default=False)
-    created_at = Column(Integer, nullable=False, default=lambda: int(time.time()))
-    deleted_at = Column(Integer, nullable=True)
+    createdAt = Column(
+        DateTime(timezone=True), nullable=False, server_default=func.now(), comment="Created at"
+    )
+    deletedAt = Column(DateTime(timezone=True), nullable=True)


 class ShoutAuthor(Base):
-    """
-    Связь между публикацией и автором.
-
-    Attributes:
-        shout (int): ID публикации
-        author (int): ID автора
-        caption (str): Подпись автора
-    """
-
    __tablename__ = "shout_author"

-    id = None  # type: ignore
-    shout = Column(ForeignKey("shout.id"), primary_key=True, index=True)
-    author = Column(ForeignKey("author.id"), primary_key=True, index=True)
-    caption = Column(String, nullable=True, default="")
-
-    # Определяем дополнительные индексы
-    __table_args__ = (
-        # Оптимизированный индекс для запросов, которые ищут публикации по автору
-        Index("idx_shout_author_author_shout", "author", "shout"),
-    )
+    id = None
+    shout: Column = Column(ForeignKey("shout.id"), primary_key=True, index=True)
+    user: Column = Column(ForeignKey("user.id"), primary_key=True, index=True)
+    caption: Column = Column(String, nullable=True, default="")


 class Shout(Base):
-    """
-    Публикация в системе.
-    """
-
    __tablename__ = "shout"

-    created_at: int = Column(Integer, nullable=False, default=lambda: int(time.time()))
-    updated_at: int | None = Column(Integer, nullable=True, index=True)
-    published_at: int | None = Column(Integer, nullable=True, index=True)
-    featured_at: int | None = Column(Integer, nullable=True, index=True)
-    deleted_at: int | None = Column(Integer, nullable=True, index=True)
-
-    created_by: int = Column(ForeignKey("author.id"), nullable=False)
-    updated_by: int | None = Column(ForeignKey("author.id"), nullable=True)
-    deleted_by: int | None = Column(ForeignKey("author.id"), nullable=True)
-    community: int = Column(ForeignKey("community.id"), nullable=False)
-
-    body: str = Column(String, nullable=False, comment="Body")
-    slug: str = Column(String, unique=True)
-    cover: str | None = Column(String, nullable=True, comment="Cover image url")
-    cover_caption: str | None = Column(String, nullable=True, comment="Cover image alt caption")
-    lead: str | None = Column(String, nullable=True)
-    title: str = Column(String, nullable=False)
-    subtitle: str | None = Column(String, nullable=True)
-    layout: str = Column(String, nullable=False, default="article")
-    media: dict | None = Column(JSON, nullable=True)
-
-    authors = relationship(Author, secondary="shout_author")
-    topics = relationship(Topic, secondary="shout_topic")
-    reactions = relationship(Reaction)
-
-    lang: str = Column(String, nullable=False, default="ru", comment="Language")
-    version_of: int | None = Column(ForeignKey("shout.id"), nullable=True)
-    oid: str | None = Column(String, nullable=True)
-
-    seo: str | None = Column(String, nullable=True)  # JSON
-
-    draft: int | None = Column(ForeignKey("draft.id"), nullable=True)
-
-    # Определяем индексы
-    __table_args__ = (
-        # Индекс для быстрого поиска неудаленных публикаций
-        Index("idx_shout_deleted_at", "deleted_at", postgresql_where=deleted_at.is_(None)),
-        # Индекс для быстрой фильтрации по community
-        Index("idx_shout_community", "community"),
-        # Индекс для быстрого поиска по slug
-        Index("idx_shout_slug", "slug"),
-        # Составной индекс для фильтрации опубликованных неудаленных публикаций
-        Index(
-            "idx_shout_published_deleted",
-            "published_at",
-            "deleted_at",
-            postgresql_where=published_at.is_not(None) & deleted_at.is_(None),
-        ),
+    # timestamps
+    createdAt = Column(
+        DateTime(timezone=True), nullable=False, server_default=func.now(), comment="Created at"
    )
+    updatedAt = Column(DateTime(timezone=True), nullable=True, comment="Updated at")
+    publishedAt = Column(DateTime(timezone=True), nullable=True)
+    deletedAt = Column(DateTime(timezone=True), nullable=True)
+
+    createdBy: Column = Column(ForeignKey("user.id"), comment="Created By")
+    deletedBy: Column = Column(ForeignKey("user.id"), nullable=True)
+
+    slug = Column(String, unique=True)
+    cover = Column(String, nullable=True, comment="Cover image url")
+    lead = Column(String, nullable=True)
+    description = Column(String, nullable=True)
+    body = Column(String, nullable=False, comment="Body")
+    title = Column(String, nullable=True)
+    subtitle = Column(String, nullable=True)
+    layout = Column(String, nullable=True)
+    media = Column(JSON, nullable=True)
+    authors = relationship(lambda: User, secondary=ShoutAuthor.__tablename__)
+    topics = relationship(lambda: Topic, secondary=ShoutTopic.__tablename__)
+
+    # views from the old Discours website
+    viewsOld = Column(Integer, default=0)
+    # views from Ackee tracker on the new Discours website
+    viewsAckee = Column(Integer, default=0)
+    views = column_property(viewsOld + viewsAckee)
+    reactions = relationship(lambda: Reaction)
+
+    # TODO: these field should be used or modified
+    community: Column = Column(ForeignKey("community.id"), default=1)
+    lang = Column(String, nullable=False, default="ru", comment="Language")
+    mainTopic: Column = Column(ForeignKey("topic.slug"), nullable=True)
+    visibility = Column(String, nullable=True)  # owner authors community public
+    versionOf: Column = Column(ForeignKey("shout.id"), nullable=True)
+    oid = Column(String, nullable=True)
+
+    @staticmethod
+    def init_table():
+        with local_session() as session:
+            s = session.query(Shout).first()
+            if not s:
+                entry = {"slug": "genesis-block", "body": "", "title": "Ничего", "lang": "ru"}
+                s = Shout.create(**entry)
+                session.add(s)
+                session.commit()
--- a/orm/topic.py
+++ b/orm/topic.py
@@ -1,66 +1,26 @@
-import time
+from sqlalchemy import Boolean, Column, DateTime, ForeignKey, String, func

-from sqlalchemy import JSON, Boolean, Column, ForeignKey, Index, Integer, String
-
-from services.db import Base
+from base.orm import Base


 class TopicFollower(Base):
-    """
-    Связь между топиком и его подписчиком.
-
-    Attributes:
-        follower (int): ID подписчика
-        topic (int): ID топика
-        created_at (int): Время создания связи
-        auto (bool): Автоматическая подписка
-    """
-
    __tablename__ = "topic_followers"

-    id = None  # type: ignore
-    follower = Column(Integer, ForeignKey("author.id"), primary_key=True)
-    topic = Column(Integer, ForeignKey("topic.id"), primary_key=True)
-    created_at = Column(Integer, nullable=False, default=int(time.time()))
-    auto = Column(Boolean, nullable=False, default=False)
-
-    # Определяем индексы
-    __table_args__ = (
-        # Индекс для быстрого поиска всех подписчиков топика
-        Index("idx_topic_followers_topic", "topic"),
-        # Индекс для быстрого поиска всех топиков, на которые подписан автор
-        Index("idx_topic_followers_follower", "follower"),
+    id = None
+    follower: Column = Column(ForeignKey("user.id"), primary_key=True, index=True)
+    topic: Column = Column(ForeignKey("topic.id"), primary_key=True, index=True)
+    createdAt = Column(
+        DateTime(timezone=True), nullable=False, server_default=func.now(), comment="Created at"
    )
+    auto = Column(Boolean, nullable=False, default=False)


 class Topic(Base):
-    """
-    Модель топика (темы) публикаций.
-
-    Attributes:
-        slug (str): Уникальный строковый идентификатор темы
-        title (str): Название темы
-        body (str): Описание темы
-        pic (str): URL изображения темы
-        community (int): ID сообщества
-        oid (str): Старый ID
-        parent_ids (list): IDs родительских тем
-    """
-
    __tablename__ = "topic"

    slug = Column(String, unique=True)
    title = Column(String, nullable=False, comment="Title")
    body = Column(String, nullable=True, comment="Body")
    pic = Column(String, nullable=True, comment="Picture")
-    community = Column(ForeignKey("community.id"), default=1)
+    community: Column = Column(ForeignKey("community.id"), default=1, comment="Community")
    oid = Column(String, nullable=True, comment="Old ID")
-    parent_ids = Column(JSON, nullable=True, comment="Parent Topic IDs")
-
-    # Определяем индексы
-    __table_args__ = (
-        # Индекс для быстрого поиска по slug
-        Index("idx_topic_slug", "slug"),
-        # Индекс для быстрого поиска по сообществу
-        Index("idx_topic_community", "community"),
-    )
--- a/orm/user.py
+++ b/orm/user.py
@@ -0,0 +1,105 @@
+from sqlalchemy import JSON as JSONType
+from sqlalchemy import Boolean, Column, DateTime, ForeignKey, Integer, String, func
+from sqlalchemy.orm import relationship
+
+from base.orm import Base, local_session
+from orm.rbac import Role
+
+
+class UserRating(Base):
+    __tablename__ = "user_rating"
+
+    id = None
+    rater: Column = Column(ForeignKey("user.id"), primary_key=True, index=True)
+    user: Column = Column(ForeignKey("user.id"), primary_key=True, index=True)
+    value: Column = Column(Integer)
+
+    @staticmethod
+    def init_table():
+        pass
+
+
+class UserRole(Base):
+    __tablename__ = "user_role"
+
+    id = None
+    user = Column(ForeignKey("user.id"), primary_key=True, index=True)
+    role = Column(ForeignKey("role.id"), primary_key=True, index=True)
+
+
+class AuthorFollower(Base):
+    __tablename__ = "author_follower"
+
+    id = None
+    follower: Column = Column(ForeignKey("user.id"), primary_key=True, index=True)
+    author: Column = Column(ForeignKey("user.id"), primary_key=True, index=True)
+    createdAt = Column(
+        DateTime(timezone=True), nullable=False, server_default=func.now(), comment="Created at"
+    )
+    auto = Column(Boolean, nullable=False, default=False)
+
+
+class User(Base):
+    __tablename__ = "user"
+    default_user = None
+
+    email = Column(String, unique=True, nullable=False, comment="Email")
+    username = Column(String, nullable=False, comment="Login")
+    password = Column(String, nullable=True, comment="Password")
+    bio = Column(String, nullable=True, comment="Bio")  # status description
+    about = Column(String, nullable=True, comment="About")  # long and formatted
+    userpic = Column(String, nullable=True, comment="Userpic")
+    name = Column(String, nullable=True, comment="Display name")
+    slug = Column(String, unique=True, comment="User's slug")
+    muted = Column(Boolean, default=False)
+    emailConfirmed = Column(Boolean, default=False)
+    createdAt = Column(
+        DateTime(timezone=True), nullable=False, server_default=func.now(), comment="Created at"
+    )
+    lastSeen = Column(
+        DateTime(timezone=True), nullable=False, server_default=func.now(), comment="Was online at"
+    )
+    deletedAt = Column(DateTime(timezone=True), nullable=True, comment="Deleted at")
+    links = Column(JSONType, nullable=True, comment="Links")
+    oauth = Column(String, nullable=True)
+    ratings = relationship(UserRating, foreign_keys=UserRating.user)
+    roles = relationship(lambda: Role, secondary=UserRole.__tablename__)
+    oid = Column(String, nullable=True)
+
+    @staticmethod
+    def init_table():
+        with local_session() as session:
+            default = session.query(User).filter(User.slug == "anonymous").first()
+            if not default:
+                default_dict = {
+                    "email": "noreply@discours.io",
+                    "username": "noreply@discours.io",
+                    "name": "Аноним",
+                    "slug": "anonymous",
+                }
+                default = User.create(**default_dict)
+                session.add(default)
+                discours_dict = {
+                    "email": "welcome@discours.io",
+                    "username": "welcome@discours.io",
+                    "name": "Дискурс",
+                    "slug": "discours",
+                }
+                discours = User.create(**discours_dict)
+                session.add(discours)
+                session.commit()
+            User.default_user = default
+
+    def get_permission(self):
+        scope = {}
+        for role in self.roles:
+            for p in role.permissions:
+                if p.resource not in scope:
+                    scope[p.resource] = set()
+                scope[p.resource].add(p.operation)
+        print(scope)
+        return scope
+
+
+# if __name__ == "__main__":
+#   print(User.get_permission(user_id=1))
--- a/poetry.lock
+++ b/poetry.lock
--- a/requirements-dev.txt
+++ b/requirements-dev.txt
@@ -0,0 +1,8 @@
+black==23.10.1
+flake8==6.1.0
+gql_schema_codegen==1.0.1
+isort==5.12.0
+mypy==1.6.1
+pre-commit==3.5.0
+pymongo-stubs==0.2.0
+sqlalchemy-stubs==0.4
--- a/requirements.dev.txt
+++ b/requirements.dev.txt
@@ -1,6 +0,0 @@
-fakeredis
-pytest
-pytest-asyncio
-pytest-cov
-mypy
-ruff
--- a/requirements.txt
+++ b/requirements.txt
@@ -1,18 +1,37 @@
-# own auth
-bcrypt
-authlib
-passlib
-opensearch-py
-google-analytics-data
-colorlog
+aiohttp==3.8.6
+alembic==1.11.3
+ariadne>=0.17.0
+asyncio~=3.4.3
+authlib==1.2.1
+bcrypt>=4.0.0
+beautifulsoup4~=4.11.1
+boto3~=1.28.2
+botocore~=1.31.2
+bson~=0.5.10
+DateTime~=4.7
+gql~=3.4.0
+graphql-core>=3.0.3
+httpx>=0.23.0
+itsdangerous
+lxml
+Mako==1.2.4
+MarkupSafe==2.1.3
+nltk~=3.8.1
+passlib~=1.7.4
 psycopg2-binary
-httpx
-redis[hiredis]
-sentry-sdk[starlette,sqlalchemy]
-starlette
-gql
-ariadne
-granian
-orjson
-pydantic
-trafilatura
+pydantic>=1.10.2
+pyjwt>=2.6.0
+pymystem3~=0.2.0
+python-dateutil~=2.8.2
+python-frontmatter~=1.0.0
+python-multipart~=0.0.6
+PyYAML>=5.4
+requests~=2.28.1
+sentry-sdk>=1.14.0
+sqlalchemy>=1.4.41
+sse-starlette==1.6.5
+starlette~=0.23.1
+transliterate~=1.10.2
+uvicorn>=0.18.3
+
+redis
--- a/resetdb.sh
+++ b/resetdb.sh
@@ -0,0 +1,55 @@
+database_name="discoursio"
+remote_backup_dir="/var/backups/mongodb"
+user="root"
+host="v2.discours.io"
+server="$user@$host"
+dump_dir="./dump"
+local_backup_filename="discours-backup.bson.gz.tar"
+
+echo "DATABASE RESET STARTED"
+echo "server: $server"
+echo "remote backup directory: $remote_backup_dir"
+
+echo "Searching for last backup file..."
+last_backup_filename=$(ssh $server "ls -t $remote_backup_dir | head -1")
+if [ $? -ne 0 ]; then { echo "Failed to get last backup filename, aborting." ; exit 1; } fi
+echo "Last backup file found: $last_backup_filename"
+
+echo "Downloading..."
+scp $server:$remote_backup_dir/"$last_backup_filename" "$local_backup_filename"
+if [ $? -ne 0 ]; then { echo "Failed to download backup file, aborting." ; exit 1; } fi
+echo "Backup file $local_backup_filename downloaded successfully"
+
+echo "Creating dump directory: $dump_dir"
+mkdir -p "$dump_dir"
+if [ $? -ne 0 ]; then { echo "Failed to create dump directory, aborting." ; exit 1; } fi
+echo "$dump_dir directory created"
+
+echo "Unpacking backup file $local_backup_filename to $dump_dir"
+tar -xzf "$local_backup_filename" --directory "$dump_dir" --strip-components 1
+if [ $? -ne 0 ]; then { echo "Failed to unpack backup, aborting." ; exit 1; } fi
+echo "Backup file $local_backup_filename successfully unpacked to $dump_dir"
+
+echo "Removing backup file $local_backup_filename"
+rm "$local_backup_filename"
+if [ $? -ne 0 ]; then { echo "Failed to remove backup file, aborting." ; exit 1; } fi
+echo "Backup file removed"
+
+echo "Dropping database $database_name"
+dropdb $database_name --force
+if [ $? -ne 0 ]; then { echo "Failed to drop database, aborting." ; exit 1; } fi
+echo "Database $database_name dropped"
+
+echo "Creating database $database_name"
+createdb $database_name
+if [ $? -ne 0 ]; then { echo "Failed to create database, aborting." ; exit 1; } fi
+echo "Database $database_name successfully created"
+
+echo "BSON -> JSON"
+python3 server.py bson
+if [ $? -ne 0 ]; then { echo "BSON -> JSON failed, aborting." ; exit 1; } fi
+
+echo "Start migration"
+python3 server.py migrate
+if [ $? -ne 0 ]; then { echo "Migration failed, aborting." ; exit 1; } fi
+echo 'Done!'
--- a/resolvers/init.py
+++ b/resolvers/init.py
@@ -1,132 +1,46 @@
-from cache.triggers import events_register
-from resolvers.author import (  # search_authors,
-    get_author,
-    get_author_followers,
-    get_author_follows,
-    get_author_follows_authors,
-    get_author_follows_topics,
-    get_author_id,
+# flake8: noqa
+
+from resolvers.auth import (
+    auth_send_link,
+    confirm_email,
+    get_current_user,
+    is_email_used,
+    login,
+    register_by_email,
+    sign_out,
+)
+from resolvers.create.editor import create_shout, delete_shout, update_shout
+from resolvers.inbox.chats import create_chat, delete_chat, update_chat
+from resolvers.inbox.load import load_chats, load_messages_by, load_recipients
+from resolvers.inbox.messages import (
+    create_message,
+    delete_message,
+    mark_as_read,
+    update_message,
+)
+from resolvers.inbox.search import search_recipients
+from resolvers.notifications import load_notifications
+from resolvers.zine.following import follow, unfollow
+from resolvers.zine.load import load_shout, load_shouts_by
+from resolvers.zine.profile import (
    get_authors_all,
    load_authors_by,
-    update_author,
+    rate_user,
+    update_profile,
 )
-from resolvers.community import get_communities_all, get_community
-from resolvers.draft import (
-    create_draft,
-    delete_draft,
-    load_drafts,
-    publish_draft,
-    update_draft,
-)
-from resolvers.editor import (
-    unpublish_shout,
-)
-from resolvers.feed import (
-    load_shouts_coauthored,
-    load_shouts_discussed,
-    load_shouts_feed,
-    load_shouts_followed_by,
-)
-from resolvers.follower import follow, get_shout_followers, unfollow
-from resolvers.notifier import (
-    load_notifications,
-    notification_mark_seen,
-    notifications_seen_after,
-    notifications_seen_thread,
-)
-from resolvers.rating import get_my_rates_comments, get_my_rates_shouts, rate_author
-from resolvers.reaction import (
+from resolvers.zine.reactions import (
    create_reaction,
    delete_reaction,
-    load_comment_ratings,
-    load_comments_branch,
    load_reactions_by,
-    load_shout_comments,
-    load_shout_ratings,
+    reactions_follow,
+    reactions_unfollow,
    update_reaction,
 )
-from resolvers.reader import (
-    get_shout,
-    load_shouts_by,
-    load_shouts_random_top,
-    load_shouts_search,
-    load_shouts_unrated,
-)
-from resolvers.topic import (
+from resolvers.zine.topics import (
    get_topic,
-    get_topic_authors,
-    get_topic_followers,
-    get_topics_all,
-    get_topics_by_author,
-    get_topics_by_community,
+    topic_follow,
+    topic_unfollow,
+    topics_all,
+    topics_by_author,
+    topics_by_community,
 )
-
-events_register()
-
-__all__ = [
-    # author
-    "get_author",
-    "get_author_id",
-    "get_author_followers",
-    "get_author_follows",
-    "get_author_follows_topics",
-    "get_author_follows_authors",
-    "get_authors_all",
-    "load_authors_by",
-    "update_author",
-    ## "search_authors",
-    # community
-    "get_community",
-    "get_communities_all",
-    # topic
-    "get_topic",
-    "get_topics_all",
-    "get_topics_by_community",
-    "get_topics_by_author",
-    "get_topic_followers",
-    "get_topic_authors",
-    # reader
-    "get_shout",
-    "load_shouts_by",
-    "load_shouts_random_top",
-    "load_shouts_search",
-    "load_shouts_unrated",
-    # feed
-    "load_shouts_feed",
-    "load_shouts_coauthored",
-    "load_shouts_discussed",
-    "load_shouts_with_topic",
-    "load_shouts_followed_by",
-    "load_shouts_authored_by",
-    # follower
-    "follow",
-    "unfollow",
-    "get_shout_followers",
-    # reaction
-    "create_reaction",
-    "update_reaction",
-    "delete_reaction",
-    "load_reactions_by",
-    "load_shout_comments",
-    "load_shout_ratings",
-    "load_comment_ratings",
-    "load_comments_branch",
-    # notifier
-    "load_notifications",
-    "notifications_seen_thread",
-    "notifications_seen_after",
-    "notification_mark_seen",
-    # rating
-    "rate_author",
-    "get_my_rates_comments",
-    "get_my_rates_shouts",
-    # draft
-    "load_drafts",
-    "create_draft",
-    "update_draft",
-    "delete_draft",
-    "publish_draft",
-    "publish_shout",
-    "unpublish_shout",
-    "unpublish_draft",
-]
--- a/resolvers/auth.py
+++ b/resolvers/auth.py
@@ -5,17 +5,18 @@ from datetime import datetime, timezone
 from urllib.parse import quote_plus

 from graphql.type import GraphQLResolveInfo
+from transliterate import translit

 from auth.authenticate import login_required
 from auth.credentials import AuthCredentials
 from auth.email import send_auth_email
-from auth.exceptions import InvalidPassword, InvalidToken, ObjectNotExist, Unauthorized
 from auth.identity import Identity, Password
 from auth.jwtcodec import JWTCodec
 from auth.tokenstorage import TokenStorage
+from base.exceptions import InvalidPassword, InvalidToken, ObjectNotExist, Unauthorized
+from base.orm import local_session
+from base.resolvers import mutation, query
 from orm import Role, User
-from services.db import local_session
-from services.schema import mutation, query
 from settings import SESSION_TOKEN_HEADER


@@ -65,50 +66,9 @@ def create_user(user_dict):
    return user


-def replace_translit(src):
-    ruchars = "абвгдеёжзийклмнопрстуфхцчшщъыьэюя."
-    enchars = [
-        "a",
-        "b",
-        "v",
-        "g",
-        "d",
-        "e",
-        "yo",
-        "zh",
-        "z",
-        "i",
-        "y",
-        "k",
-        "l",
-        "m",
-        "n",
-        "o",
-        "p",
-        "r",
-        "s",
-        "t",
-        "u",
-        "f",
-        "h",
-        "c",
-        "ch",
-        "sh",
-        "sch",
-        "",
-        "y",
-        "'",
-        "e",
-        "yu",
-        "ya",
-        "-",
-    ]
-    return src.translate(str.maketrans(ruchars, enchars))
-
-
 def generate_unique_slug(src):
    print("[resolvers.auth] generating slug from: " + src)
-    slug = replace_translit(src.lower())
+    slug = translit(src, "ru", reversed=True).replace(".", "-").lower()
    slug = re.sub("[^0-9a-zA-Z]+", "-", slug)
    if slug != src:
        print("[resolvers.auth] translited name: " + slug)
--- a/resolvers/author.py
+++ b/resolvers/author.py
@@ -1,460 +0,0 @@
-import asyncio
-import time
-from typing import Optional
-
-from sqlalchemy import select, text
-
-from cache.cache import (
-    cache_author,
-    cached_query,
-    get_cached_author,
-    get_cached_author_by_user_id,
-    get_cached_author_followers,
-    get_cached_follower_authors,
-    get_cached_follower_topics,
-    invalidate_cache_by_prefix,
-)
-from orm.author import Author
-from resolvers.stat import get_with_stat
-from services.auth import login_required
-from services.db import local_session
-from services.redis import redis
-from services.schema import mutation, query
-from utils.logger import root_logger as logger
-
-DEFAULT_COMMUNITIES = [1]
-
-
-# Вспомогательная функция для получения всех авторов без статистики
-async def get_all_authors():
-    """
-    Получает всех авторов без статистики.
-    Используется для случаев, когда нужен полный список авторов без дополнительной информации.
-
-    Returns:
-        list: Список всех авторов без статистики
-    """
-    cache_key = "authors:all:basic"
-
-    # Функция для получения всех авторов из БД
-    async def fetch_all_authors():
-        logger.debug("Получаем список всех авторов из БД и кешируем результат")
-
-        with local_session() as session:
-            # Запрос на получение базовой информации об авторах
-            authors_query = select(Author).where(Author.deleted_at.is_(None))
-            authors = session.execute(authors_query).scalars().all()
-
-            # Преобразуем авторов в словари
-            return [author.dict() for author in authors]
-
-    # Используем универсальную функцию для кеширования запросов
-    return await cached_query(cache_key, fetch_all_authors)
-
-
-# Вспомогательная функция для получения авторов со статистикой с пагинацией
-async def get_authors_with_stats(limit=50, offset=0, by: Optional[str] = None):
-    """
-    Получает авторов со статистикой с пагинацией.
-
-    Args:
-        limit: Максимальное количество возвращаемых авторов
-        offset: Смещение для пагинации
-        by: Опциональный параметр сортировки (new/active)
-
-    Returns:
-        list: Список авторов с их статистикой
-    """
-    # Формируем ключ кеша с помощью универсальной функции
-    cache_key = f"authors:stats:limit={limit}:offset={offset}"
-
-    # Функция для получения авторов из БД
-    async def fetch_authors_with_stats():
-        logger.debug(f"Выполняем запрос на получение авторов со статистикой: limit={limit}, offset={offset}, by={by}")
-
-        with local_session() as session:
-            # Базовый запрос для получения авторов
-            base_query = select(Author).where(Author.deleted_at.is_(None))
-
-            # Применяем сортировку
-            
-            # vars for statistics sorting
-            stats_sort_field = None
-            stats_sort_direction = "desc"
-            
-            if by:
-                if isinstance(by, dict):
-                    logger.debug(f"Processing dict-based sorting: {by}")
-                    # Обработка словаря параметров сортировки
-                    from sqlalchemy import asc, desc, func
-                    from orm.shout import ShoutAuthor
-                    from orm.author import AuthorFollower
-
-                    # Checking for order field in the dictionary
-                    if "order" in by:
-                        order_value = by["order"]
-                        logger.debug(f"Found order field with value: {order_value}")
-                        if order_value in ["shouts", "followers", "rating", "comments"]:
-                            stats_sort_field = order_value
-                            stats_sort_direction = "desc"  # По умолчанию убывающая сортировка для статистики
-                            logger.debug(f"Applying statistics-based sorting by: {stats_sort_field}")
-                        elif order_value == "name":
-                            # Sorting by name in ascending order
-                            base_query = base_query.order_by(asc(Author.name))
-                            logger.debug("Applying alphabetical sorting by name")
-                        else:
-                            # If order is not a stats field, treat it as a regular field
-                            column = getattr(Author, order_value, None)
-                            if column:
-                                base_query = base_query.order_by(desc(column))
-                    else:
-                        # Regular sorting by fields
-                        for field, direction in by.items():
-                            column = getattr(Author, field, None)
-                            if column:
-                                if direction.lower() == "desc":
-                                    base_query = base_query.order_by(desc(column))
-                                else:
-                                    base_query = base_query.order_by(column)
-                elif by == "new":
-                    base_query = base_query.order_by(desc(Author.created_at))
-                elif by == "active":
-                    base_query = base_query.order_by(desc(Author.last_seen))
-                else:
-                    # По умолчанию сортируем по времени создания
-                    base_query = base_query.order_by(desc(Author.created_at))
-            else:
-                base_query = base_query.order_by(desc(Author.created_at))
-
-            # If sorting by statistics, modify the query
-            if stats_sort_field == "shouts":
-                # Sorting by the number of shouts
-                from sqlalchemy import func, and_
-                from orm.shout import Shout, ShoutAuthor
-                
-                subquery = (
-                    select(
-                        ShoutAuthor.author,
-                        func.count(func.distinct(Shout.id)).label("shouts_count")
-                    )
-                    .select_from(ShoutAuthor)
-                    .join(Shout, ShoutAuthor.shout == Shout.id)
-                    .where(
-                        and_(
-                            Shout.deleted_at.is_(None),
-                            Shout.published_at.is_not(None)
-                        )
-                    )
-                    .group_by(ShoutAuthor.author)
-                    .subquery()
-                )
-                
-                base_query = (
-                    base_query
-                    .outerjoin(subquery, Author.id == subquery.c.author)
-                    .order_by(desc(func.coalesce(subquery.c.shouts_count, 0)))
-                )
-            elif stats_sort_field == "followers":
-                # Sorting by the number of followers
-                from sqlalchemy import func
-                from orm.author import AuthorFollower
-                
-                subquery = (
-                    select(
-                        AuthorFollower.author,
-                        func.count(func.distinct(AuthorFollower.follower)).label("followers_count")
-                    )
-                    .select_from(AuthorFollower)
-                    .group_by(AuthorFollower.author)
-                    .subquery()
-                )
-                
-                base_query = (
-                    base_query
-                    .outerjoin(subquery, Author.id == subquery.c.author)
-                    .order_by(desc(func.coalesce(subquery.c.followers_count, 0)))
-                )
-
-            # Применяем лимит и смещение
-            base_query = base_query.limit(limit).offset(offset)
-
-            # Получаем авторов
-            authors = session.execute(base_query).scalars().all()
-            author_ids = [author.id for author in authors]
-
-            if not author_ids:
-                return []
-
-            # Оптимизированный запрос для получения статистики по публикациям для авторов
-            shouts_stats_query = f"""
-            SELECT sa.author, COUNT(DISTINCT s.id) as shouts_count
-            FROM shout_author sa
-            JOIN shout s ON sa.shout = s.id AND s.deleted_at IS NULL AND s.published_at IS NOT NULL
-            WHERE sa.author IN ({",".join(map(str, author_ids))})
-            GROUP BY sa.author
-            """
-            shouts_stats = {row[0]: row[1] for row in session.execute(text(shouts_stats_query))}
-
-            # Запрос на получение статистики по подписчикам для авторов
-            followers_stats_query = f"""
-            SELECT author, COUNT(DISTINCT follower) as followers_count
-            FROM author_follower
-            WHERE author IN ({",".join(map(str, author_ids))})
-            GROUP BY author
-            """
-            followers_stats = {row[0]: row[1] for row in session.execute(text(followers_stats_query))}
-
-            # Формируем результат с добавлением статистики
-            result = []
-            for author in authors:
-                author_dict = author.dict()
-                author_dict["stat"] = {
-                    "shouts": shouts_stats.get(author.id, 0),
-                    "followers": followers_stats.get(author.id, 0),
-                }
-                result.append(author_dict)
-
-                # Кешируем каждого автора отдельно для использования в других функциях
-                await cache_author(author_dict)
-
-            return result
-
-    # Используем универсальную функцию для кеширования запросов
-    return await cached_query(cache_key, fetch_authors_with_stats)
-
-
-# Функция для инвалидации кеша авторов
-async def invalidate_authors_cache(author_id=None):
-    """
-    Инвалидирует кеши авторов при изменении данных.
-
-    Args:
-        author_id: Опциональный ID автора для точечной инвалидации.
-               Если не указан, инвалидируются все кеши авторов.
-    """
-    if author_id:
-        # Точечная инвалидация конкретного автора
-        logger.debug(f"Инвалидация кеша для автора #{author_id}")
-        specific_keys = [
-            f"author:id:{author_id}",
-            f"author:followers:{author_id}",
-            f"author:follows-authors:{author_id}",
-            f"author:follows-topics:{author_id}",
-            f"author:follows-shouts:{author_id}",
-        ]
-
-        # Получаем user_id автора, если есть
-        with local_session() as session:
-            author = session.query(Author).filter(Author.id == author_id).first()
-            if author and author.user:
-                specific_keys.append(f"author:user:{author.user.strip()}")
-
-        # Удаляем конкретные ключи
-        for key in specific_keys:
-            try:
-                await redis.execute("DEL", key)
-                logger.debug(f"Удален ключ кеша {key}")
-            except Exception as e:
-                logger.error(f"Ошибка при удалении ключа {key}: {e}")
-
-        # Также ищем и удаляем ключи коллекций, содержащих данные об этом авторе
-        collection_keys = await redis.execute("KEYS", "authors:stats:*")
-        if collection_keys:
-            await redis.execute("DEL", *collection_keys)
-            logger.debug(f"Удалено {len(collection_keys)} коллекционных ключей авторов")
-    else:
-        # Общая инвалидация всех кешей авторов
-        logger.debug("Полная инвалидация кеша авторов")
-        await invalidate_cache_by_prefix("authors")
-
-
-@mutation.field("update_author")
-@login_required
-async def update_author(_, info, profile):
-    user_id = info.context.get("user_id")
-    if not user_id:
-        return {"error": "unauthorized", "author": None}
-    try:
-        with local_session() as session:
-            author = session.query(Author).where(Author.user == user_id).first()
-            if author:
-                Author.update(author, profile)
-                session.add(author)
-                session.commit()
-                author_query = select(Author).where(Author.user == user_id)
-                result = get_with_stat(author_query)
-                if result:
-                    author_with_stat = result[0]
-                    if isinstance(author_with_stat, Author):
-                        author_dict = author_with_stat.dict()
-                        # await cache_author(author_dict)
-                        asyncio.create_task(cache_author(author_dict))
-                return {"error": None, "author": author}
-    except Exception as exc:
-        import traceback
-
-        logger.error(traceback.format_exc())
-        return {"error": exc, "author": None}
-
-
-@query.field("get_authors_all")
-async def get_authors_all(_, _info):
-    """
-    Получает список всех авторов без статистики.
-
-    Returns:
-        list: Список всех авторов
-    """
-    return await get_all_authors()
-
-
-@query.field("get_author")
-async def get_author(_, _info, slug="", author_id=0):
-    author_dict = None
-    try:
-        author_id = get_author_id_from(slug=slug, user="", author_id=author_id)
-        if not author_id:
-            raise ValueError("cant find")
-        author_dict = await get_cached_author(int(author_id), get_with_stat)
-
-        if not author_dict or not author_dict.get("stat"):
-            # update stat from db
-            author_query = select(Author).filter(Author.id == author_id)
-            result = get_with_stat(author_query)
-            if result:
-                author_with_stat = result[0]
-                if isinstance(author_with_stat, Author):
-                    author_dict = author_with_stat.dict()
-                    # await cache_author(author_dict)
-                    asyncio.create_task(cache_author(author_dict))
-    except ValueError:
-        pass
-    except Exception as exc:
-        import traceback
-
-        logger.error(f"{exc}:\n{traceback.format_exc()}")
-    return author_dict
-
-
-@query.field("get_author_id")
-async def get_author_id(_, _info, user: str):
-    user_id = user.strip()
-    logger.info(f"getting author id for {user_id}")
-    author = None
-    try:
-        author = await get_cached_author_by_user_id(user_id, get_with_stat)
-        if author:
-            return author
-
-        author_query = select(Author).filter(Author.user == user_id)
-        result = get_with_stat(author_query)
-        if result:
-            author_with_stat = result[0]
-            if isinstance(author_with_stat, Author):
-                author_dict = author_with_stat.dict()
-                # await cache_author(author_dict)
-                asyncio.create_task(cache_author(author_dict))
-                return author_with_stat
-    except Exception as exc:
-        logger.error(f"Error getting author: {exc}")
-    return None
-
-
-@query.field("load_authors_by")
-async def load_authors_by(_, _info, by, limit, offset):
-    """
-    Загружает авторов по заданному критерию с пагинацией.
-
-    Args:
-        by: Критерий сортировки авторов (new/active)
-        limit: Максимальное количество возвращаемых авторов
-        offset: Смещение для пагинации
-
-    Returns:
-        list: Список авторов с учетом критерия
-    """
-    # Используем оптимизированную функцию для получения авторов
-    return await get_authors_with_stats(limit, offset, by)
-
-
-def get_author_id_from(slug="", user=None, author_id=None):
-    try:
-        author_id = None
-        if author_id:
-            return author_id
-        with local_session() as session:
-            author = None
-            if slug:
-                author = session.query(Author).filter(Author.slug == slug).first()
-                if author:
-                    author_id = author.id
-                    return author_id
-            if user:
-                author = session.query(Author).filter(Author.user == user).first()
-                if author:
-                    author_id = author.id
-    except Exception as exc:
-        logger.error(exc)
-    return author_id
-
-
-@query.field("get_author_follows")
-async def get_author_follows(_, _info, slug="", user=None, author_id=0):
-    logger.debug(f"getting follows for @{slug}")
-    author_id = get_author_id_from(slug=slug, user=user, author_id=author_id)
-    if not author_id:
-        return {}
-
-    followed_authors = await get_cached_follower_authors(author_id)
-    followed_topics = await get_cached_follower_topics(author_id)
-
-    # TODO: Get followed communities too
-    return {
-        "authors": followed_authors,
-        "topics": followed_topics,
-        "communities": DEFAULT_COMMUNITIES,
-        "shouts": [],
-    }
-
-
-@query.field("get_author_follows_topics")
-async def get_author_follows_topics(_, _info, slug="", user=None, author_id=None):
-    logger.debug(f"getting followed topics for @{slug}")
-    author_id = get_author_id_from(slug=slug, user=user, author_id=author_id)
-    if not author_id:
-        return []
-    followed_topics = await get_cached_follower_topics(author_id)
-    return followed_topics
-
-
-@query.field("get_author_follows_authors")
-async def get_author_follows_authors(_, _info, slug="", user=None, author_id=None):
-    logger.debug(f"getting followed authors for @{slug}")
-    author_id = get_author_id_from(slug=slug, user=user, author_id=author_id)
-    if not author_id:
-        return []
-    followed_authors = await get_cached_follower_authors(author_id)
-    return followed_authors
-
-
-def create_author(user_id: str, slug: str, name: str = ""):
-    author = Author()
-    author.user = user_id  # Связь с user_id из системы авторизации
-    author.slug = slug  # Идентификатор из системы авторизации
-    author.created_at = author.updated_at = int(time.time())
-    author.name = name or slug  # если не указано
-
-    with local_session() as session:
-        session.add(author)
-        session.commit()
-        return author
-
-
-@query.field("get_author_followers")
-async def get_author_followers(_, _info, slug: str = "", user: str = "", author_id: int = 0):
-    logger.debug(f"getting followers for author @{slug} or ID:{author_id}")
-    author_id = get_author_id_from(slug=slug, user=user, author_id=author_id)
-    if not author_id:
-        return []
-    followers = await get_cached_author_followers(author_id)
-    return followers
--- a/resolvers/bookmark.py
+++ b/resolvers/bookmark.py
@@ -1,83 +0,0 @@
-from operator import and_
-
-from graphql import GraphQLError
-from sqlalchemy import delete, insert
-
-from orm.author import AuthorBookmark
-from orm.shout import Shout
-from resolvers.feed import apply_options
-from resolvers.reader import get_shouts_with_links, query_with_stat
-from services.auth import login_required
-from services.common_result import CommonResult
-from services.db import local_session
-from services.schema import mutation, query
-
-
-@query.field("load_shouts_bookmarked")
-@login_required
-def load_shouts_bookmarked(_, info, options):
-    """
-    Load bookmarked shouts for the authenticated user.
-
-    Args:
-        limit (int): Maximum number of shouts to return.
-        offset (int): Number of shouts to skip.
-
-    Returns:
-        list: List of bookmarked shouts.
-    """
-    author_dict = info.context.get("author", {})
-    author_id = author_dict.get("id")
-    if not author_id:
-        raise GraphQLError("User not authenticated")
-
-    q = query_with_stat(info)
-    q = q.join(AuthorBookmark)
-    q = q.filter(
-        and_(
-            Shout.id == AuthorBookmark.shout,
-            AuthorBookmark.author == author_id,
-        )
-    )
-    q, limit, offset = apply_options(q, options, author_id)
-    return get_shouts_with_links(info, q, limit, offset)
-
-
-@mutation.field("toggle_bookmark_shout")
-def toggle_bookmark_shout(_, info, slug: str) -> CommonResult:
-    """
-    Toggle bookmark status for a specific shout.
-
-    Args:
-        slug (str): Unique identifier of the shout.
-
-    Returns:
-        CommonResult: Result of the operation with bookmark status.
-    """
-    author_dict = info.context.get("author", {})
-    author_id = author_dict.get("id")
-    if not author_id:
-        raise GraphQLError("User not authenticated")
-
-    with local_session() as db:
-        shout = db.query(Shout).filter(Shout.slug == slug).first()
-        if not shout:
-            raise GraphQLError("Shout not found")
-
-        existing_bookmark = (
-            db.query(AuthorBookmark)
-            .filter(AuthorBookmark.author == author_id, AuthorBookmark.shout == shout.id)
-            .first()
-        )
-
-        if existing_bookmark:
-            db.execute(
-                delete(AuthorBookmark).where(AuthorBookmark.author == author_id, AuthorBookmark.shout == shout.id)
-            )
-            result = False
-        else:
-            db.execute(insert(AuthorBookmark).values(author=author_id, shout=shout.id))
-            result = True
-
-        db.commit()
-        return result
--- a/resolvers/collab.py
+++ b/resolvers/collab.py
@@ -1,147 +0,0 @@
-from orm.author import Author
-from orm.invite import Invite, InviteStatus
-from orm.shout import Shout
-from services.auth import login_required
-from services.db import local_session
-from services.schema import mutation
-
-
-@mutation.field("accept_invite")
-@login_required
-async def accept_invite(_, info, invite_id: int):
-    info.context["user_id"]
-    author_dict = info.context["author"]
-    author_id = author_dict.get("id")
-    if author_id:
-        author_id = int(author_id)
-        # Check if the user exists
-        with local_session() as session:
-            # Check if the invite exists
-            invite = session.query(Invite).filter(Invite.id == invite_id).first()
-            if invite and invite.author_id is author_id and invite.status is InviteStatus.PENDING.value:
-                # Add the user to the shout authors
-                shout = session.query(Shout).filter(Shout.id == invite.shout_id).first()
-                if shout:
-                    if author_id not in shout.authors:
-                        author = session.query(Author).filter(Author.id == author_id).first()
-                        if author:
-                            shout.authors.append(author)
-                            session.add(shout)
-                            session.delete(invite)
-                            session.commit()
-                    return {"success": True, "message": "Invite accepted"}
-                else:
-                    return {"error": "Shout not found"}
-            else:
-                return {"error": "Invalid invite or already accepted/rejected"}
-    else:
-        return {"error": "Unauthorized"}
-
-
-@mutation.field("reject_invite")
-@login_required
-async def reject_invite(_, info, invite_id: int):
-    info.context["user_id"]
-    author_dict = info.context["author"]
-    author_id = author_dict.get("id")
-
-    if author_id:
-        # Check if the user exists
-        with local_session() as session:
-            author_id = int(author_id)
-            # Check if the invite exists
-            invite = session.query(Invite).filter(Invite.id == invite_id).first()
-            if invite and invite.author_id is author_id and invite.status is InviteStatus.PENDING.value:
-                # Delete the invite
-                session.delete(invite)
-                session.commit()
-                return {"success": True, "message": "Invite rejected"}
-            else:
-                return {"error": "Invalid invite or already accepted/rejected"}
-    return {"error": "User not found"}
-
-
-@mutation.field("create_invite")
-@login_required
-async def create_invite(_, info, slug: str = "", author_id: int = 0):
-    user_id = info.context["user_id"]
-    author_dict = info.context["author"]
-    author_id = author_dict.get("id")
-    if author_id:
-        # Check if the inviter is the owner of the shout
-        with local_session() as session:
-            shout = session.query(Shout).filter(Shout.slug == slug).first()
-            inviter = session.query(Author).filter(Author.user == user_id).first()
-            if inviter and shout and shout.authors and inviter.id is shout.created_by:
-                # Check if an invite already exists
-                existing_invite = (
-                    session.query(Invite)
-                    .filter(
-                        Invite.inviter_id == inviter.id,
-                        Invite.author_id == author_id,
-                        Invite.shout_id == shout.id,
-                        Invite.status == InviteStatus.PENDING.value,
-                    )
-                    .first()
-                )
-                if existing_invite:
-                    return {"error": "Invite already sent"}
-
-                # Create a new invite
-                new_invite = Invite(
-                    inviter_id=user_id,
-                    author_id=author_id,
-                    shout_id=shout.id,
-                    status=InviteStatus.PENDING.value,
-                )
-                session.add(new_invite)
-                session.commit()
-
-                return {"error": None, "invite": new_invite}
-            else:
-                return {"error": "Invalid author"}
-    else:
-        return {"error": "Access denied"}
-
-
-@mutation.field("remove_author")
-@login_required
-async def remove_author(_, info, slug: str = "", author_id: int = 0):
-    user_id = info.context["user_id"]
-    with local_session() as session:
-        author = session.query(Author).filter(Author.user == user_id).first()
-        if author:
-            shout = session.query(Shout).filter(Shout.slug == slug).first()
-            # NOTE: owner should be first in a list
-            if shout and author.id is shout.created_by:
-                shout.authors = [author for author in shout.authors if author.id != author_id]
-                session.commit()
-                return {}
-    return {"error": "Access denied"}
-
-
-@mutation.field("remove_invite")
-@login_required
-async def remove_invite(_, info, invite_id: int):
-    info.context["user_id"]
-
-    author_dict = info.context["author"]
-    author_id = author_dict.get("id")
-    if isinstance(author_id, int):
-        # Check if the user exists
-        with local_session() as session:
-            # Check if the invite exists
-            invite = session.query(Invite).filter(Invite.id == invite_id).first()
-            if isinstance(invite, Invite):
-                shout = session.query(Shout).filter(Shout.id == invite.shout_id).first()
-                if shout and shout.deleted_at is None and invite:
-                    if invite.inviter_id is author_id or author_id == shout.created_by:
-                        if invite.status is InviteStatus.PENDING.value:
-                            # Delete the invite
-                            session.delete(invite)
-                            session.commit()
-                            return {}
-            else:
-                return {"error": "Invalid invite or already accepted/rejected"}
-    else:
-        return {"error": "Author not found"}
--- a/resolvers/community.py
+++ b/resolvers/community.py
@@ -1,97 +0,0 @@
-from orm.author import Author
-from orm.community import Community, CommunityFollower
-from services.db import local_session
-from services.schema import mutation, query
-
-
-@query.field("get_communities_all")
-async def get_communities_all(_, _info):
-    return local_session().query(Community).all()
-
-
-@query.field("get_community")
-async def get_community(_, _info, slug: str):
-    q = local_session().query(Community).where(Community.slug == slug)
-    return q.first()
-
-
-@query.field("get_communities_by_author")
-async def get_communities_by_author(_, _info, slug="", user="", author_id=0):
-    with local_session() as session:
-        q = session.query(Community).join(CommunityFollower)
-        if slug:
-            author_id = session.query(Author).where(Author.slug == slug).first().id
-            q = q.where(CommunityFollower.author == author_id)
-        if user:
-            author_id = session.query(Author).where(Author.user == user).first().id
-            q = q.where(CommunityFollower.author == author_id)
-        if author_id:
-            q = q.where(CommunityFollower.author == author_id)
-        return q.all()
-    return []
-
-
-@mutation.field("join_community")
-async def join_community(_, info, slug: str):
-    author_dict = info.context.get("author", {})
-    author_id = author_dict.get("id")
-    with local_session() as session:
-        community = session.query(Community).where(Community.slug == slug).first()
-        if not community:
-            return {"ok": False, "error": "Community not found"}
-        session.add(CommunityFollower(community=community.id, author=author_id))
-        session.commit()
-        return {"ok": True}
-
-
-@mutation.field("leave_community")
-async def leave_community(_, info, slug: str):
-    author_dict = info.context.get("author", {})
-    author_id = author_dict.get("id")
-    with local_session() as session:
-        session.query(CommunityFollower).where(
-            CommunityFollower.author == author_id, CommunityFollower.community == slug
-        ).delete()
-        session.commit()
-        return {"ok": True}
-
-
-@mutation.field("create_community")
-async def create_community(_, info, community_data):
-    author_dict = info.context.get("author", {})
-    author_id = author_dict.get("id")
-    with local_session() as session:
-        session.add(Community(author=author_id, **community_data))
-        session.commit()
-        return {"ok": True}
-
-
-@mutation.field("update_community")
-async def update_community(_, info, community_data):
-    author_dict = info.context.get("author", {})
-    author_id = author_dict.get("id")
-    slug = community_data.get("slug")
-    if slug:
-        with local_session() as session:
-            try:
-                session.query(Community).where(Community.created_by == author_id, Community.slug == slug).update(
-                    community_data
-                )
-                session.commit()
-            except Exception as e:
-                return {"ok": False, "error": str(e)}
-            return {"ok": True}
-    return {"ok": False, "error": "Please, set community slug in input"}
-
-
-@mutation.field("delete_community")
-async def delete_community(_, info, slug: str):
-    author_dict = info.context.get("author", {})
-    author_id = author_dict.get("id")
-    with local_session() as session:
-        try:
-            session.query(Community).where(Community.slug == slug, Community.created_by == author_id).delete()
-            session.commit()
-            return {"ok": True}
-        except Exception as e:
-            return {"ok": False, "error": str(e)}
--- a/resolvers/create/editor.py
+++ b/resolvers/create/editor.py
@@ -0,0 +1,179 @@
+from datetime import datetime, timezone
+
+from sqlalchemy import and_
+from sqlalchemy.orm import joinedload
+
+from auth.authenticate import login_required
+from auth.credentials import AuthCredentials
+from base.orm import local_session
+from base.resolvers import mutation
+from orm.shout import Shout, ShoutAuthor, ShoutTopic
+from orm.topic import Topic
+from resolvers.zine.reactions import reactions_follow, reactions_unfollow
+
+
+@mutation.field("createShout")
+@login_required
+async def create_shout(_, info, inp):
+    auth: AuthCredentials = info.context["request"].auth
+
+    with local_session() as session:
+        topics = session.query(Topic).filter(Topic.slug.in_(inp.get("topics", []))).all()
+
+        new_shout = Shout.create(
+            **{
+                "title": inp.get("title"),
+                "subtitle": inp.get("subtitle"),
+                "lead": inp.get("lead"),
+                "description": inp.get("description"),
+                "body": inp.get("body", ""),
+                "layout": inp.get("layout"),
+                "authors": inp.get("authors", []),
+                "slug": inp.get("slug"),
+                "mainTopic": inp.get("mainTopic"),
+                "visibility": "owner",
+                "createdBy": auth.user_id,
+            }
+        )
+
+        for topic in topics:
+            t = ShoutTopic.create(topic=topic.id, shout=new_shout.id)
+            session.add(t)
+
+        # NOTE: shout made by one first author
+        sa = ShoutAuthor.create(shout=new_shout.id, user=auth.user_id)
+        session.add(sa)
+
+        session.add(new_shout)
+
+        reactions_follow(auth.user_id, new_shout.id, True)
+
+        session.commit()
+
+        # TODO
+        # GitTask(inp, user.username, user.email, "new shout %s" % new_shout.slug)
+
+        if new_shout.slug is None:
+            new_shout.slug = f"draft-{new_shout.id}"
+            session.commit()
+
+    return {"shout": new_shout}
+
+
+@mutation.field("updateShout")
+@login_required
+async def update_shout(_, info, shout_id, shout_input=None, publish=False):  # noqa: C901
+    auth: AuthCredentials = info.context["request"].auth
+
+    with local_session() as session:
+        shout = (
+            session.query(Shout)
+            .options(
+                joinedload(Shout.authors),
+                joinedload(Shout.topics),
+            )
+            .filter(Shout.id == shout_id)
+            .first()
+        )
+
+        if not shout:
+            return {"error": "shout not found"}
+
+        if shout.createdBy != auth.user_id:
+            return {"error": "access denied"}
+
+        updated = False
+
+        if shout_input is not None:
+            topics_input = shout_input["topics"]
+            del shout_input["topics"]
+
+            new_topics_to_link = []
+            new_topics = [topic_input for topic_input in topics_input if topic_input["id"] < 0]
+
+            for new_topic in new_topics:
+                del new_topic["id"]
+                created_new_topic = Topic.create(**new_topic)
+                session.add(created_new_topic)
+                new_topics_to_link.append(created_new_topic)
+
+            if len(new_topics) > 0:
+                session.commit()
+
+            for new_topic_to_link in new_topics_to_link:
+                created_unlinked_topic = ShoutTopic.create(
+                    shout=shout.id, topic=new_topic_to_link.id
+                )
+                session.add(created_unlinked_topic)
+
+            existing_topics_input = [
+                topic_input for topic_input in topics_input if topic_input.get("id", 0) > 0
+            ]
+            existing_topic_to_link_ids = [
+                existing_topic_input["id"]
+                for existing_topic_input in existing_topics_input
+                if existing_topic_input["id"] not in [topic.id for topic in shout.topics]
+            ]
+
+            for existing_topic_to_link_id in existing_topic_to_link_ids:
+                created_unlinked_topic = ShoutTopic.create(
+                    shout=shout.id, topic=existing_topic_to_link_id
+                )
+                session.add(created_unlinked_topic)
+
+            topic_to_unlink_ids = [
+                topic.id
+                for topic in shout.topics
+                if topic.id not in [topic_input["id"] for topic_input in existing_topics_input]
+            ]
+
+            shout_topics_to_remove = session.query(ShoutTopic).filter(
+                and_(ShoutTopic.shout == shout.id, ShoutTopic.topic.in_(topic_to_unlink_ids))
+            )
+
+            for shout_topic_to_remove in shout_topics_to_remove:
+                session.delete(shout_topic_to_remove)
+
+            shout_input["mainTopic"] = shout_input["mainTopic"]["slug"]
+
+            if shout_input["mainTopic"] == "":
+                del shout_input["mainTopic"]
+
+            shout.update(shout_input)
+            updated = True
+
+        if publish and shout.visibility == "owner":
+            shout.visibility = "community"
+            shout.publishedAt = datetime.now(tz=timezone.utc)
+            updated = True
+
+        if updated:
+            shout.updatedAt = datetime.now(tz=timezone.utc)
+
+        session.commit()
+    # GitTask(inp, user.username, user.email, "update shout %s" % slug)
+
+    return {"shout": shout}
+
+
+@mutation.field("deleteShout")
+@login_required
+async def delete_shout(_, info, shout_id):
+    auth: AuthCredentials = info.context["request"].auth
+
+    with local_session() as session:
+        shout = session.query(Shout).filter(Shout.id == shout_id).first()
+
+        if not shout:
+            return {"error": "invalid shout id"}
+
+        if auth.user_id != shout.createdBy:
+            return {"error": "access denied"}
+
+        for author_id in shout.authors:
+            reactions_unfollow(author_id, shout_id)
+
+        shout.deletedAt = datetime.now(tz=timezone.utc)
+        session.commit()
+
+    return {}
--- a/resolvers/draft.py
+++ b/resolvers/draft.py
@@ -1,475 +0,0 @@
-import time
-import trafilatura
-from sqlalchemy.orm import joinedload
-
-from cache.cache import (
-    cache_author,
-    cache_by_id,
-    cache_topic,
-    invalidate_shout_related_cache,
-    invalidate_shouts_cache,
-)
-from orm.author import Author
-from orm.draft import Draft, DraftAuthor, DraftTopic
-from orm.shout import Shout, ShoutAuthor, ShoutTopic
-from orm.topic import Topic
-from services.auth import login_required
-from services.db import local_session
-from services.notify import notify_shout
-from services.schema import mutation, query
-from services.search import search_service
-from utils.html_wrapper import wrap_html_fragment
-from utils.logger import root_logger as logger
-
-def create_shout_from_draft(session, draft, author_id):
-    """
-    Создаёт новый объект публикации (Shout) на основе черновика.
-
-    Args:
-        session: SQLAlchemy сессия (не используется, для совместимости)
-        draft (Draft): Объект черновика
-        author_id (int): ID автора публикации
-
-    Returns:
-        Shout: Новый объект публикации (не сохранённый в базе)
-
-    Пример:
-        >>> from orm.draft import Draft
-        >>> draft = Draft(id=1, title='Заголовок', body='Текст', slug='slug', created_by=1)
-        >>> shout = create_shout_from_draft(None, draft, 1)
-        >>> shout.title
-        'Заголовок'
-        >>> shout.body
-        'Текст'
-        >>> shout.created_by
-        1
-    """
-    # Создаем новую публикацию
-    shout = Shout(
-        body=draft.body or "",
-        slug=draft.slug,
-        cover=draft.cover,
-        cover_caption=draft.cover_caption,
-        lead=draft.lead,
-        title=draft.title or "",
-        subtitle=draft.subtitle,
-        layout=draft.layout or "article",
-        media=draft.media or [],
-        lang=draft.lang or "ru",
-        seo=draft.seo,
-        created_by=author_id,
-        community=draft.community,
-        draft=draft.id,
-        deleted_at=None,
-    )
-    
-    # Инициализируем пустые массивы для связей
-    shout.topics = []
-    shout.authors = []
-    
-    return shout
-
-
-@query.field("load_drafts")
-@login_required
-async def load_drafts(_, info):
-    """
-    Загружает все черновики, доступные текущему пользователю.
-    
-    Предварительно загружает связанные объекты (topics, authors, publication),
-    чтобы избежать ошибок с отсоединенными объектами при сериализации.
-    
-    Returns:
-        dict: Список черновиков или сообщение об ошибке
-    """
-    user_id = info.context.get("user_id")
-    author_dict = info.context.get("author", {})
-    author_id = author_dict.get("id")
-
-    if not user_id or not author_id:
-        return {"error": "User ID and author ID are required"}
-
-    try:
-        with local_session() as session:
-            # Предзагружаем authors, topics и связанную publication
-            drafts_query = (
-                session.query(Draft)
-                .options(
-                    joinedload(Draft.topics),
-                    joinedload(Draft.authors),
-                    joinedload(Draft.publication) # Загружаем связанную публикацию
-                )
-                .filter(Draft.authors.any(Author.id == author_id))
-            )
-            drafts = drafts_query.all()
-            
-            # Преобразуем объекты в словари, пока они в контексте сессии
-            drafts_data = []
-            for draft in drafts:
-                draft_dict = draft.dict()
-                # Всегда возвращаем массив для topics, даже если он пустой
-                draft_dict["topics"] = [topic.dict() for topic in (draft.topics or [])]
-                draft_dict["authors"] = [author.dict() for author in (draft.authors or [])]
-                
-                # Добавляем информацию о публикации, если она есть
-                if draft.publication:
-                    draft_dict["publication"] = {
-                        "id": draft.publication.id,
-                        "slug": draft.publication.slug,
-                        "published_at": draft.publication.published_at
-                    }
-                else:
-                     draft_dict["publication"] = None
-                     
-                drafts_data.append(draft_dict)
-                
-            return {"drafts": drafts_data}
-    except Exception as e:
-        logger.error(f"Failed to load drafts: {e}", exc_info=True)
-        return {"error": f"Failed to load drafts: {str(e)}"}
-
-
-@mutation.field("create_draft")
-@login_required
-async def create_draft(_, info, draft_input):
-    """Create a new draft.
-
-    Args:
-        info: GraphQL context
-        draft_input (dict): Draft data including optional fields:
-            - title (str, required) - заголовок черновика
-            - body (str, required) - текст черновика
-            - slug (str)
-            - etc.
-
-    Returns:
-        dict: Contains either:
-            - draft: The created draft object
-            - error: Error message if creation failed
-
-    Example:
-        >>> async def test_create():
-        ...     context = {'user_id': '123', 'author': {'id': 1}}
-        ...     info = type('Info', (), {'context': context})()
-        ...     result = await create_draft(None, info, {'title': 'Test'})
-        ...     assert result.get('error') is None
-        ...     assert result['draft'].title == 'Test'
-        ...     return result
-    """
-    user_id = info.context.get("user_id")
-    author_dict = info.context.get("author", {})
-    author_id = author_dict.get("id")
-
-    if not user_id or not author_id:
-        return {"error": "Author ID is required"}
-
-    # Проверяем обязательные поля
-    if "body" not in draft_input or not draft_input["body"]:
-        draft_input["body"] = ""  # Пустая строка вместо NULL
-
-    if "title" not in draft_input or not draft_input["title"]:
-        draft_input["title"] = ""  # Пустая строка вместо NULL
-
-    # Проверяем slug - он должен быть или не пустым, или не передаваться вообще
-    if "slug" in draft_input and (draft_input["slug"] is None or draft_input["slug"] == ""):
-        # При создании черновика удаляем пустой slug из входных данных
-        del draft_input["slug"]
-
-    try:
-        with local_session() as session:
-            # Remove id from input if present since it's auto-generated
-            if "id" in draft_input:
-                del draft_input["id"]
-                
-            # Добавляем текущее время создания и ID автора
-            draft_input["created_at"] = int(time.time())
-            draft_input["created_by"] = author_id
-            draft = Draft(**draft_input)
-            session.add(draft)
-            session.flush()
-            
-            # Добавляем создателя как автора
-            da = DraftAuthor(shout=draft.id, author=author_id)
-            session.add(da)
-            
-            session.commit()
-            return {"draft": draft}
-    except Exception as e:
-        logger.error(f"Failed to create draft: {e}", exc_info=True)
-        return {"error": f"Failed to create draft: {str(e)}"}
-
-def generate_teaser(body, limit=300):
-    body_html = wrap_html_fragment(body)
-    body_text = trafilatura.extract(body_html, include_comments=False, include_tables=False)
-    body_teaser = ". ".join(body_text[:limit].split(". ")[:-1])
-    return body_teaser
-
-
-@mutation.field("update_draft")
-@login_required
-async def update_draft(_, info, draft_id: int, draft_input):
-    """Обновляет черновик публикации.
-
-    Args:
-        draft_id: ID черновика для обновления
-        draft_input: Данные для обновления черновика согласно схеме DraftInput:
-            - layout: String
-            - author_ids: [Int!]
-            - topic_ids: [Int!]
-            - main_topic_id: Int
-            - media: [MediaItemInput]
-            - lead: String
-            - subtitle: String
-            - lang: String
-            - seo: String
-            - body: String
-            - title: String
-            - slug: String
-            - cover: String
-            - cover_caption: String
-
-    Returns:
-        dict: Обновленный черновик или сообщение об ошибке
-    """
-    user_id = info.context.get("user_id")
-    author_dict = info.context.get("author", {})
-    author_id = author_dict.get("id")
-
-    if not user_id or not author_id:
-        return {"error": "Author ID are required"}
-
-    try:
-        with local_session() as session:
-            draft = session.query(Draft).filter(Draft.id == draft_id).first()
-            if not draft:
-                return {"error": "Draft not found"}
-
-            # Фильтруем входные данные, оставляя только разрешенные поля
-            allowed_fields = {
-                "layout", "author_ids", "topic_ids", "main_topic_id", 
-                "media", "lead", "subtitle", "lang", "seo", "body", 
-                "title", "slug", "cover", "cover_caption"
-            }
-            filtered_input = {k: v for k, v in draft_input.items() if k in allowed_fields}
-
-            # Проверяем slug
-            if "slug" in filtered_input and not filtered_input["slug"]:
-                del filtered_input["slug"]
-
-            # Обновляем связи с авторами если переданы
-            if "author_ids" in filtered_input:
-                author_ids = filtered_input.pop("author_ids")
-                if author_ids:
-                    # Очищаем текущие связи
-                    session.query(DraftAuthor).filter(DraftAuthor.shout == draft_id).delete()
-                    # Добавляем новые связи
-                    for aid in author_ids:
-                        da = DraftAuthor(shout=draft_id, author=aid)
-                        session.add(da)
-
-            # Обновляем связи с темами если переданы
-            if "topic_ids" in filtered_input:
-                topic_ids = filtered_input.pop("topic_ids")
-                main_topic_id = filtered_input.pop("main_topic_id", None)
-                if topic_ids:
-                    # Очищаем текущие связи
-                    session.query(DraftTopic).filter(DraftTopic.shout == draft_id).delete()
-                    # Добавляем новые связи
-                    for tid in topic_ids:
-                        dt = DraftTopic(
-                            shout=draft_id, 
-                            topic=tid,
-                            main=(tid == main_topic_id) if main_topic_id else False
-                        )
-                        session.add(dt)
-
-            # Генерируем SEO если не предоставлено
-            if "seo" not in filtered_input and not draft.seo:
-                body_src = filtered_input.get("body", draft.body)
-                lead_src = filtered_input.get("lead", draft.lead)
-                body_html = wrap_html_fragment(body_src)
-                lead_html = wrap_html_fragment(lead_src)
-                
-                try:
-                    body_text = trafilatura.extract(body_html, include_comments=False, include_tables=False) if body_src else None
-                    lead_text = trafilatura.extract(lead_html, include_comments=False, include_tables=False) if lead_src else None
-                    
-                    body_teaser = generate_teaser(body_text, 300) if body_text else ""
-                    filtered_input["seo"] = lead_text if lead_text else body_teaser
-                except Exception as e:
-                    logger.warning(f"Failed to generate SEO for draft {draft_id}: {e}")
-
-            # Обновляем основные поля черновика
-            for key, value in filtered_input.items():
-                setattr(draft, key, value)
-
-            # Обновляем метаданные
-            draft.updated_at = int(time.time())
-            draft.updated_by = author_id
-
-            session.commit()
-            
-            # Преобразуем объект в словарь для ответа
-            draft_dict = draft.dict()
-            draft_dict["topics"] = [topic.dict() for topic in draft.topics]
-            draft_dict["authors"] = [author.dict() for author in draft.authors]
-            # Добавляем объект автора в updated_by
-            draft_dict["updated_by"] = author_dict
-            
-            return {"draft": draft_dict}
-
-    except Exception as e:
-        logger.error(f"Failed to update draft: {e}", exc_info=True)
-        return {"error": f"Failed to update draft: {str(e)}"}
-
-
-@mutation.field("delete_draft")
-@login_required
-async def delete_draft(_, info, draft_id: int):
-    author_dict = info.context.get("author", {})
-    author_id = author_dict.get("id")
-
-    with local_session() as session:
-        draft = session.query(Draft).filter(Draft.id == draft_id).first()
-        if not draft:
-            return {"error": "Draft not found"}
-        if author_id != draft.created_by and draft.authors.filter(Author.id == author_id).count() == 0:
-            return {"error": "You are not allowed to delete this draft"}
-        session.delete(draft)
-        session.commit()
-        return {"draft": draft}
-
-
-def validate_html_content(html_content: str) -> tuple[bool, str]:
-    """
-    Проверяет валидность HTML контента через trafilatura.
-    
-    Args:
-        html_content: HTML строка для проверки
-        
-    Returns:
-        tuple[bool, str]: (валидность, сообщение об ошибке)
-        
-    Example:
-        >>> is_valid, error = validate_html_content("<p>Valid HTML</p>")
-        >>> is_valid
-        True
-        >>> error
-        ''
-        >>> is_valid, error = validate_html_content("Invalid < HTML")
-        >>> is_valid
-        False
-        >>> 'Invalid HTML' in error
-        True
-    """
-    if not html_content or not html_content.strip():
-        return False, "Content is empty"
-        
-    try:
-        html_content = wrap_html_fragment(html_content)
-        extracted = trafilatura.extract(html_content)
-        if not extracted:
-            return False, "Invalid HTML structure or empty content"
-        return True, ""
-    except Exception as e:
-        logger.error(f"HTML validation error: {e}", exc_info=True)
-        return False, f"Invalid HTML content: {str(e)}"
-
-
-@mutation.field("publish_draft")
-@login_required
-async def publish_draft(_, info, draft_id: int):
-    """
-    Публикует черновик, создавая новый Shout или обновляя существующий.
-    
-    Args:
-        draft_id (int): ID черновика для публикации
-        
-    Returns:
-        dict: Результат публикации с shout или сообщением об ошибке
-    """
-    user_id = info.context.get("user_id")
-    author_dict = info.context.get("author", {})
-    author_id = author_dict.get("id")
-
-    if not user_id or not author_id:
-        return {"error": "Author ID is required"}
-
-    try:
-        with local_session() as session:
-            # Загружаем черновик со всеми связями
-            draft = (
-                session.query(Draft)
-                .options(
-                    joinedload(Draft.topics),
-                    joinedload(Draft.authors),
-                    joinedload(Draft.publication)
-                )
-                .filter(Draft.id == draft_id)
-                .first()
-            )
-
-            if not draft:
-                return {"error": "Draft not found"}
-
-            # Проверка валидности HTML в body
-            is_valid, error = validate_html_content(draft.body)
-            if not is_valid:
-                return {"error": f"Cannot publish draft: {error}"}
-
-            # Проверяем, есть ли уже публикация для этого черновика
-            if draft.publication:
-                shout = draft.publication
-                # Обновляем существующую публикацию
-                for field in ["body", "title", "subtitle", "lead", "cover", "cover_caption", "media", "lang", "seo"]:
-                    if hasattr(draft, field):
-                        setattr(shout, field, getattr(draft, field))
-                shout.updated_at = int(time.time())
-                shout.updated_by = author_id
-            else:
-                # Создаем новую публикацию
-                shout = create_shout_from_draft(session, draft, author_id)
-                now = int(time.time())
-                shout.created_at = now
-                shout.published_at = now
-                session.add(shout)
-                session.flush()  # Получаем ID нового шаута
-
-            # Очищаем существующие связи
-            session.query(ShoutAuthor).filter(ShoutAuthor.shout == shout.id).delete()
-            session.query(ShoutTopic).filter(ShoutTopic.shout == shout.id).delete()
-
-            # Добавляем авторов
-            for author in (draft.authors or []):
-                sa = ShoutAuthor(shout=shout.id, author=author.id)
-                session.add(sa)
-
-            # Добавляем темы
-            for topic in (draft.topics or []):
-                st = ShoutTopic(
-                    topic=topic.id,
-                    shout=shout.id,
-                    main=topic.main if hasattr(topic, "main") else False
-                )
-                session.add(st)
-
-            session.commit()
-
-            # Инвалидируем кеш
-            invalidate_shouts_cache()
-            invalidate_shout_related_cache(shout.id)
-
-            # Уведомляем о публикации
-            await notify_shout(shout.id)
-
-            # Обновляем поисковый индекс
-            search_service.index_shout(shout)
-
-            logger.info(f"Successfully published shout #{shout.id} from draft #{draft_id}")
-            logger.debug(f"Shout data: {shout.dict()}")
-
-            return {"shout": shout}
-
-    except Exception as e:
-        logger.error(f"Failed to publish draft {draft_id}: {e}", exc_info=True)
-        return {"error": f"Failed to publish draft: {str(e)}"}
--- a/resolvers/editor.py
+++ b/resolvers/editor.py
@@ -1,826 +0,0 @@
-import time
-
-import orjson
-import trafilatura
-from sqlalchemy import and_, desc, select
-from sqlalchemy.orm import joinedload, selectinload
-from sqlalchemy.sql.functions import coalesce
-
-from cache.cache import (
-    cache_author,
-    cache_topic,
-    invalidate_shout_related_cache,
-    invalidate_shouts_cache,
-)
-from orm.author import Author
-from orm.draft import Draft
-from orm.shout import Shout, ShoutAuthor, ShoutTopic
-from orm.topic import Topic
-from resolvers.follower import follow, unfollow
-from resolvers.stat import get_with_stat
-from services.auth import login_required
-from services.db import local_session
-from services.notify import notify_shout
-from services.schema import mutation, query
-from services.search import search_service
-from utils.html_wrapper import wrap_html_fragment
-from utils.logger import root_logger as logger
-
-
-async def cache_by_id(entity, entity_id: int, cache_method):
-    """Cache an entity by its ID using the provided cache method.
-
-    Args:
-        entity: The SQLAlchemy model class to query
-        entity_id (int): The ID of the entity to cache
-        cache_method: The caching function to use
-
-    Returns:
-        dict: The cached entity data if successful, None if entity not found
-
-    Example:
-        >>> async def test_cache():
-        ...     author = await cache_by_id(Author, 1, cache_author)
-        ...     assert author['id'] == 1
-        ...     assert 'name' in author
-        ...     return author
-    """
-    caching_query = select(entity).filter(entity.id == entity_id)
-    result = get_with_stat(caching_query)
-    if not result or not result[0]:
-        logger.warning(f"{entity.__name__} with id {entity_id} not found")
-        return
-    x = result[0]
-    d = x.dict()  # convert object to dictionary
-    cache_method(d)
-    return d
-
-
-@query.field("get_my_shout")
-@login_required
-async def get_my_shout(_, info, shout_id: int):
-    """Get a shout by ID if the requesting user has permission to view it.
-
-    DEPRECATED: use `load_drafts` instead
-
-    Args:
-        info: GraphQL resolver info containing context
-        shout_id (int): ID of the shout to retrieve
-
-    Returns:
-        dict: Contains either:
-            - error (str): Error message if retrieval failed
-            - shout (Shout): The requested shout if found and accessible
-
-    Permissions:
-        User must be:
-        - The shout creator
-        - Listed as an author
-        - Have editor role
-
-    Example:
-        >>> async def test_get_my_shout():
-        ...     context = {'user_id': '123', 'author': {'id': 1}, 'roles': []}
-        ...     info = type('Info', (), {'context': context})()
-        ...     result = await get_my_shout(None, info, 1)
-        ...     assert result['error'] is None
-        ...     assert result['shout'].id == 1
-        ...     return result
-    """
-    user_id = info.context.get("user_id", "")
-    author_dict = info.context.get("author", {})
-    author_id = author_dict.get("id")
-    roles = info.context.get("roles", [])
-    shout = None
-    if not user_id or not author_id:
-        return {"error": "unauthorized", "shout": None}
-    with local_session() as session:
-        shout = (
-            session.query(Shout)
-            .filter(Shout.id == shout_id)
-            .options(joinedload(Shout.authors), joinedload(Shout.topics))
-            .filter(Shout.deleted_at.is_(None))
-            .first()
-        )
-        if not shout:
-            return {"error": "no shout found", "shout": None}
-
-        # Преобразуем media JSON в список объектов MediaItem
-        if hasattr(shout, "media") and shout.media:
-            if isinstance(shout.media, str):
-                try:
-                    shout.media = orjson.loads(shout.media)
-                except Exception as e:
-                    logger.error(f"Error parsing shout media: {e}")
-                    shout.media = []
-            if not isinstance(shout.media, list):
-                shout.media = [shout.media] if shout.media else []
-        else:
-            shout.media = []
-
-        logger.debug(f"got {len(shout.authors)} shout authors, created by {shout.created_by}")
-        is_editor = "editor" in roles
-        logger.debug(f"viewer is{'' if is_editor else ' not'} editor")
-        is_creator = author_id == shout.created_by
-        logger.debug(f"viewer is{'' if is_creator else ' not'} creator")
-        is_author = bool(list(filter(lambda x: x.id == int(author_id), [x for x in shout.authors])))
-        logger.debug(f"viewer is{'' if is_creator else ' not'} author")
-        can_edit = is_editor or is_author or is_creator
-
-        if not can_edit:
-            return {"error": "forbidden", "shout": None}
-
-        logger.debug("got shout editor with data")
-        return {"error": None, "shout": shout}
-
-
-@query.field("get_shouts_drafts")
-@login_required
-async def get_shouts_drafts(_, info):
-    # user_id = info.context.get("user_id")
-    author_dict = info.context.get("author")
-    if not author_dict:
-        return {"error": "author profile was not found"}
-    author_id = author_dict.get("id")
-    shouts = []
-    with local_session() as session:
-        if author_id:
-            q = (
-                select(Shout)
-                .options(joinedload(Shout.authors), joinedload(Shout.topics))
-                .filter(and_(Shout.deleted_at.is_(None), Shout.created_by == int(author_id)))
-                .filter(Shout.published_at.is_(None))
-                .order_by(desc(coalesce(Shout.updated_at, Shout.created_at)))
-                .group_by(Shout.id)
-            )
-            shouts = [shout for [shout] in session.execute(q).unique()]
-    return {"shouts": shouts}
-
-
-# @mutation.field("create_shout")
-# @login_required
-async def create_shout(_, info, inp):
-    logger.info(f"Starting create_shout with input: {inp}")
-    user_id = info.context.get("user_id")
-    author_dict = info.context.get("author")
-    logger.debug(f"Context user_id: {user_id}, author: {author_dict}")
-
-    if not author_dict:
-        logger.error("Author profile not found in context")
-        return {"error": "author profile was not found"}
-
-    author_id = author_dict.get("id")
-    if user_id and author_id:
-        try:
-            with local_session() as session:
-                author_id = int(author_id)
-                current_time = int(time.time())
-                slug = inp.get("slug") or f"draft-{current_time}"
-
-                logger.info(f"Creating shout with input: {inp}")
-                # Создаем публикацию без topics
-                body = inp.get("body", "")
-                lead = inp.get("lead", "")
-                body_html = wrap_html_fragment(body)
-                lead_html = wrap_html_fragment(lead)
-                body_text = trafilatura.extract(body_html)
-                lead_text = trafilatura.extract(lead_html)
-                seo = inp.get("seo", lead_text.strip() or body_text.strip()[:300].split(". ")[:-1].join(". "))
-                new_shout = Shout(
-                    slug=slug,
-                    body=body,
-                    seo=seo,
-                    lead=lead,
-                    layout=inp.get("layout", "article"),
-                    title=inp.get("title", ""),
-                    created_by=author_id,
-                    created_at=current_time,
-                    community=1,
-                )
-
-                # Проверяем уникальность slug
-                logger.debug(f"Checking for existing slug: {slug}")
-                same_slug_shout = session.query(Shout).filter(Shout.slug == new_shout.slug).first()
-                c = 1
-                while same_slug_shout is not None:
-                    logger.debug(f"Found duplicate slug, trying iteration {c}")
-                    new_shout.slug = f"{slug}-{c}"
-                    same_slug_shout = session.query(Shout).filter(Shout.slug == new_shout.slug).first()
-                    c += 1
-
-                try:
-                    logger.info("Creating new shout object")
-                    session.add(new_shout)
-                    session.commit()
-                    logger.info(f"Created shout with ID: {new_shout.id}")
-                except Exception as e:
-                    logger.error(f"Error creating shout object: {e}", exc_info=True)
-                    return {"error": f"Database error: {str(e)}"}
-
-                # Связываем с автором
-                try:
-                    logger.debug(f"Linking author {author_id} to shout {new_shout.id}")
-                    sa = ShoutAuthor(shout=new_shout.id, author=author_id)
-                    session.add(sa)
-                except Exception as e:
-                    logger.error(f"Error linking author: {e}", exc_info=True)
-                    return {"error": f"Error linking author: {str(e)}"}
-
-                # Связываем с темами
-
-                input_topics = inp.get("topics", [])
-                if input_topics:
-                    try:
-                        logger.debug(f"Linking topics: {[t.slug for t in input_topics]}")
-                        main_topic = inp.get("main_topic")
-                        for topic in input_topics:
-                            st = ShoutTopic(
-                                topic=topic.id,
-                                shout=new_shout.id,
-                                main=(topic.slug == main_topic) if main_topic else False,
-                            )
-                            session.add(st)
-                            logger.debug(f"Added topic {topic.slug} {'(main)' if st.main else ''}")
-                    except Exception as e:
-                        logger.error(f"Error linking topics: {e}", exc_info=True)
-                        return {"error": f"Error linking topics: {str(e)}"}
-
-                try:
-                    session.commit()
-                    logger.info("Final commit successful")
-                except Exception as e:
-                    logger.error(f"Error in final commit: {e}", exc_info=True)
-                    return {"error": f"Error in final commit: {str(e)}"}
-
-                # Получаем созданную публикацию
-                shout = session.query(Shout).filter(Shout.id == new_shout.id).first()
-
-                # Подписываем автора
-                try:
-                    logger.debug("Following created shout")
-                    await follow(None, info, "shout", shout.slug)
-                except Exception as e:
-                    logger.warning(f"Error following shout: {e}", exc_info=True)
-
-                logger.info(f"Successfully created shout {shout.id}")
-                return {"shout": shout}
-
-        except Exception as e:
-            logger.error(f"Unexpected error in create_shout: {e}", exc_info=True)
-            return {"error": f"Unexpected error: {str(e)}"}
-
-    error_msg = "cant create shout" if user_id else "unauthorized"
-    logger.error(f"Create shout failed: {error_msg}")
-    return {"error": error_msg}
-
-
-def patch_main_topic(session, main_topic_slug, shout):
-    """Update the main topic for a shout."""
-    logger.info(f"Starting patch_main_topic for shout#{shout.id} with slug '{main_topic_slug}'")
-    logger.debug(f"Current shout topics: {[(t.topic.slug, t.main) for t in shout.topics]}")
-
-    with session.begin():
-        # Получаем текущий главный топик
-        old_main = (
-            session.query(ShoutTopic).filter(and_(ShoutTopic.shout == shout.id, ShoutTopic.main.is_(True))).first()
-        )
-        if old_main:
-            logger.info(f"Found current main topic: {old_main.topic.slug}")
-        else:
-            logger.info("No current main topic found")
-
-        # Находим новый главный топик
-        main_topic = session.query(Topic).filter(Topic.slug == main_topic_slug).first()
-        if not main_topic:
-            logger.error(f"Main topic with slug '{main_topic_slug}' not found")
-            return
-
-        logger.info(f"Found new main topic: {main_topic.slug} (id={main_topic.id})")
-
-        # Находим связь с новым главным топиком
-        new_main = (
-            session.query(ShoutTopic)
-            .filter(and_(ShoutTopic.shout == shout.id, ShoutTopic.topic == main_topic.id))
-            .first()
-        )
-        logger.debug(f"Found new main topic relation: {new_main is not None}")
-
-        if old_main and new_main and old_main is not new_main:
-            logger.info(f"Updating main topic flags: {old_main.topic.slug} -> {new_main.topic.slug}")
-            old_main.main = False
-            session.add(old_main)
-
-            new_main.main = True
-            session.add(new_main)
-
-            session.flush()
-            logger.info(f"Main topic updated for shout#{shout.id}")
-        else:
-            logger.warning(f"No changes needed for main topic (old={old_main is not None}, new={new_main is not None})")
-
-
-def patch_topics(session, shout, topics_input):
-    """Update the topics associated with a shout.
-
-    Args:
-        session: SQLAlchemy session
-        shout (Shout): The shout to update
-        topics_input (list): List of topic dicts with fields:
-            - id (int): Topic ID (<0 for new topics)
-            - slug (str): Topic slug
-            - title (str): Topic title (for new topics)
-
-    Side Effects:
-        - Creates new topics if needed
-        - Updates shout-topic associations
-        - Refreshes shout object with new topics
-
-    Example:
-        >>> def test_patch_topics():
-        ...     topics = [
-        ...         {'id': -1, 'slug': 'new-topic', 'title': 'New Topic'},
-        ...         {'id': 1, 'slug': 'existing-topic'}
-        ...     ]
-        ...     with local_session() as session:
-        ...         shout = session.query(Shout).first()
-        ...         patch_topics(session, shout, topics)
-        ...         assert len(shout.topics) == 2
-        ...         assert any(t.slug == 'new-topic' for t in shout.topics)
-        ...         return shout.topics
-    """
-    logger.info(f"Starting patch_topics for shout#{shout.id}")
-    logger.info(f"Received topics_input: {topics_input}")
-
-    # Создаем новые топики если есть
-    new_topics_to_link = [Topic(**new_topic) for new_topic in topics_input if new_topic["id"] < 0]
-    if new_topics_to_link:
-        logger.info(f"Creating new topics: {[t.dict() for t in new_topics_to_link]}")
-        session.add_all(new_topics_to_link)
-        session.flush()
-
-    # Получаем текущие связи
-    current_links = session.query(ShoutTopic).filter(ShoutTopic.shout == shout.id).all()
-    logger.info(f"Current topic links: {[{t.topic: t.main} for t in current_links]}")
-
-    # Удаляем старые связи
-    if current_links:
-        logger.info(f"Removing old topic links for shout#{shout.id}")
-        for link in current_links:
-            session.delete(link)
-        session.flush()
-
-    # Создаем новые связи
-    for topic_input in topics_input:
-        topic_id = topic_input["id"]
-        if topic_id < 0:
-            topic = next(t for t in new_topics_to_link if t.slug == topic_input["slug"])
-            topic_id = topic.id
-
-        logger.info(f"Creating new topic link: shout#{shout.id} -> topic#{topic_id}")
-        new_link = ShoutTopic(shout=shout.id, topic=topic_id, main=False)
-        session.add(new_link)
-
-    session.flush()
-    # Обновляем связи в объекте шаута
-    session.refresh(shout)
-
-    logger.info(f"Successfully updated topics for shout#{shout.id}")
-    logger.info(f"Final shout topics: {[t.dict() for t in shout.topics]}")
-
-
-# @mutation.field("update_shout")
-# @login_required
-async def update_shout(_, info, shout_id: int, shout_input=None, publish=False):
-    logger.info(f"Starting update_shout with id={shout_id}, publish={publish}")
-    logger.debug(f"Full shout_input: {shout_input}")  # DraftInput
-
-    user_id = info.context.get("user_id")
-    roles = info.context.get("roles", [])
-    author_dict = info.context.get("author")
-    if not author_dict:
-        logger.error("Author profile not found")
-        return {"error": "author profile was not found"}
-
-    author_id = author_dict.get("id")
-    shout_input = shout_input or {}
-    current_time = int(time.time())
-    shout_id = shout_id or shout_input.get("id", shout_id)
-    slug = shout_input.get("slug")
-
-    if not user_id:
-        logger.error("Unauthorized update attempt")
-        return {"error": "unauthorized"}
-
-    try:
-        with local_session() as session:
-            if author_id:
-                logger.info(f"Processing update for shout#{shout_id} by author #{author_id}")
-                shout_by_id = (
-                    session.query(Shout)
-                    .options(joinedload(Shout.topics).joinedload(ShoutTopic.topic), joinedload(Shout.authors))
-                    .filter(Shout.id == shout_id)
-                    .first()
-                )
-
-                if not shout_by_id:
-                    logger.error(f"shout#{shout_id} not found")
-                    return {"error": "shout not found"}
-
-                logger.info(f"Found shout#{shout_id}")
-
-                # Логируем текущие топики
-                current_topics = (
-                    [{"id": t.id, "slug": t.slug, "title": t.title} for t in shout_by_id.topics]
-                    if shout_by_id.topics
-                    else []
-                )
-                logger.info(f"Current topics for shout#{shout_id}: {current_topics}")
-
-                if slug != shout_by_id.slug:
-                    same_slug_shout = session.query(Shout).filter(Shout.slug == slug).first()
-                    c = 1
-                    while same_slug_shout is not None:
-                        c += 1
-                        slug = f"{slug}-{c}"
-                        same_slug_shout = session.query(Shout).filter(Shout.slug == slug).first()
-                    shout_input["slug"] = slug
-                    logger.info(f"shout#{shout_id} slug patched")
-
-                if filter(lambda x: x.id == author_id, [x for x in shout_by_id.authors]) or "editor" in roles:
-                    logger.info(f"Author #{author_id} has permission to edit shout#{shout_id}")
-
-                    # topics patch
-                    topics_input = shout_input.get("topics")
-                    if topics_input:
-                        logger.info(f"Received topics_input for shout#{shout_id}: {topics_input}")
-                        try:
-                            patch_topics(session, shout_by_id, topics_input)
-                            logger.info(f"Successfully patched topics for shout#{shout_id}")
-
-                            # Обновляем связи в сессии после patch_topics
-                            session.refresh(shout_by_id)
-
-                        except Exception as e:
-                            logger.error(f"Error patching topics: {e}", exc_info=True)
-                            return {"error": f"Failed to update topics: {str(e)}"}
-
-                        del shout_input["topics"]
-                        for tpc in topics_input:
-                            await cache_by_id(Topic, tpc["id"], cache_topic)
-                    else:
-                        logger.warning(f"No topics_input received for shout#{shout_id}")
-
-                    # main topic
-                    main_topic = shout_input.get("main_topic")
-                    if main_topic:
-                        logger.info(f"Updating main topic for shout#{shout_id} to {main_topic}")
-                        patch_main_topic(session, main_topic, shout_by_id)
-
-                    shout_input["updated_at"] = current_time
-                    if publish:
-                        logger.info(f"Publishing shout#{shout_id}")
-                        shout_input["published_at"] = current_time
-                        # Проверяем наличие связи с автором
-                        logger.info(f"Checking author link for shout#{shout_id} and author#{author_id}")
-                        author_link = (
-                            session.query(ShoutAuthor)
-                            .filter(and_(ShoutAuthor.shout == shout_id, ShoutAuthor.author == author_id))
-                            .first()
-                        )
-
-                        if not author_link:
-                            logger.info(f"Adding missing author link for shout#{shout_id}")
-                            sa = ShoutAuthor(shout=shout_id, author=author_id)
-                            session.add(sa)
-                            session.flush()
-                            logger.info("Author link added successfully")
-                        else:
-                            logger.info("Author link already exists")
-
-                    # Логируем финальное состояние перед сохранением
-                    logger.info(f"Final shout_input for update: {shout_input}")
-                    Shout.update(shout_by_id, shout_input)
-                    session.add(shout_by_id)
-
-                    try:
-                        session.commit()
-                        # Обновляем объект после коммита чтобы получить все связи
-                        session.refresh(shout_by_id)
-                        logger.info(f"Successfully committed updates for shout#{shout_id}")
-                    except Exception as e:
-                        logger.error(f"Commit failed: {e}", exc_info=True)
-                        return {"error": f"Failed to save changes: {str(e)}"}
-
-                    # После обновления проверяем топики
-                    updated_topics = (
-                        [{"id": t.id, "slug": t.slug, "title": t.title} for t in shout_by_id.topics]
-                        if shout_by_id.topics
-                        else []
-                    )
-                    logger.info(f"Updated topics for shout#{shout_id}: {updated_topics}")
-
-                    # Инвалидация кэша после обновления
-                    try:
-                        logger.info("Invalidating cache after shout update")
-
-                        cache_keys = [
-                            "feed",  # лента
-                            f"author_{author_id}",  # публикации автора
-                            "random_top",  # случайные топовые
-                            "unrated",  # неоцененные
-                        ]
-
-                        # Добавляем ключи для тем публикации
-                        for topic in shout_by_id.topics:
-                            cache_keys.append(f"topic_{topic.id}")
-                            cache_keys.append(f"topic_shouts_{topic.id}")
-
-                        await invalidate_shouts_cache(cache_keys)
-                        await invalidate_shout_related_cache(shout_by_id, author_id)
-
-                        # Обновляем кэш тем и авторов
-                        for topic in shout_by_id.topics:
-                            await cache_by_id(Topic, topic.id, cache_topic)
-                        for author in shout_by_id.authors:
-                            await cache_author(author.dict())
-
-                        logger.info("Cache invalidated successfully")
-                    except Exception as cache_error:
-                        logger.warning(f"Cache invalidation error: {cache_error}", exc_info=True)
-
-                    if not publish:
-                        await notify_shout(shout_by_id.dict(), "update")
-                    else:
-                        await notify_shout(shout_by_id.dict(), "published")
-                        # search service indexing
-                        search_service.index(shout_by_id)
-                        for a in shout_by_id.authors:
-                            await cache_by_id(Author, a.id, cache_author)
-                    logger.info(f"shout#{shout_id} updated")
-                    # Получаем полные данные шаута со связями
-                    shout_with_relations = (
-                        session.query(Shout)
-                        .options(joinedload(Shout.topics).joinedload(ShoutTopic.topic), joinedload(Shout.authors))
-                        .filter(Shout.id == shout_id)
-                        .first()
-                    )
-
-                    # Создаем словарь с базовыми полями
-                    shout_dict = shout_with_relations.dict()
-
-                    # Явно добавляем связанные данные
-                    shout_dict["topics"] = (
-                        [
-                            {"id": topic.id, "slug": topic.slug, "title": topic.title}
-                            for topic in shout_with_relations.topics
-                        ]
-                        if shout_with_relations.topics
-                        else []
-                    )
-
-                    # Add main_topic to the shout dictionary
-                    shout_dict["main_topic"] = get_main_topic(shout_with_relations.topics)
-
-                    shout_dict["authors"] = (
-                        [
-                            {"id": author.id, "name": author.name, "slug": author.slug}
-                            for author in shout_with_relations.authors
-                        ]
-                        if shout_with_relations.authors
-                        else []
-                    )
-
-                    logger.info(f"Final shout data with relations: {shout_dict}")
-                    logger.debug(
-                        f"Loaded topics details: {[(t.topic.slug if t.topic else 'no-topic', t.main) for t in shout_with_relations.topics]}"
-                    )
-                    return {"shout": shout_dict, "error": None}
-                else:
-                    logger.warning(f"Access denied: author #{author_id} cannot edit shout#{shout_id}")
-                    return {"error": "access denied", "shout": None}
-
-    except Exception as exc:
-        logger.error(f"Unexpected error in update_shout: {exc}", exc_info=True)
-        logger.error(f"Failed input data: {shout_input}")
-        return {"error": "cant update shout"}
-
-    return {"error": "cant update shout"}
-
-
-# @mutation.field("delete_shout")
-# @login_required
-async def delete_shout(_, info, shout_id: int):
-    user_id = info.context.get("user_id")
-    roles = info.context.get("roles", [])
-    author_dict = info.context.get("author")
-    if not author_dict:
-        return {"error": "author profile was not found"}
-    author_id = author_dict.get("id")
-    if user_id and author_id:
-        author_id = int(author_id)
-        with local_session() as session:
-            shout = session.query(Shout).filter(Shout.id == shout_id).first()
-            if not isinstance(shout, Shout):
-                return {"error": "invalid shout id"}
-            shout_dict = shout.dict()
-            # NOTE: only owner and editor can mark the shout as deleted
-            if shout_dict["created_by"] == author_id or "editor" in roles:
-                shout_dict["deleted_at"] = int(time.time())
-                Shout.update(shout, shout_dict)
-                session.add(shout)
-                session.commit()
-
-                for author in shout.authors:
-                    await cache_by_id(Author, author.id, cache_author)
-                    info.context["author"] = author.dict()
-                    info.context["user_id"] = author.user
-                    unfollow(None, info, "shout", shout.slug)
-
-                for topic in shout.topics:
-                    await cache_by_id(Topic, topic.id, cache_topic)
-
-                await notify_shout(shout_dict, "delete")
-                return {"error": None}
-            else:
-                return {"error": "access denied"}
-
-
-def get_main_topic(topics):
-    """Get the main topic from a list of ShoutTopic objects."""
-    logger.info(f"Starting get_main_topic with {len(topics) if topics else 0} topics")
-    logger.debug(
-        f"Topics data: {[(t.slug, getattr(t, 'main', False)) for t in topics] if topics else []}"
-    )
-
-    if not topics:
-        logger.warning("No topics provided to get_main_topic")
-        return {"id": 0, "title": "no topic", "slug": "notopic", "is_main": True}
-
-    # Проверяем, является ли topics списком объектов ShoutTopic или Topic
-    if hasattr(topics[0], 'topic') and topics[0].topic:
-        # Для ShoutTopic объектов (старый формат)
-        # Find first main topic in original order
-        main_topic_rel = next((st for st in topics if getattr(st, 'main', False)), None)
-        logger.debug(
-            f"Found main topic relation: {main_topic_rel.topic.slug if main_topic_rel and main_topic_rel.topic else None}"
-        )
-
-        if main_topic_rel and main_topic_rel.topic:
-            result = {
-                "slug": main_topic_rel.topic.slug,
-                "title": main_topic_rel.topic.title,
-                "id": main_topic_rel.topic.id,
-                "is_main": True,
-            }
-            logger.info(f"Returning main topic: {result}")
-            return result
-
-        # If no main found but topics exist, return first
-        if topics and topics[0].topic:
-            logger.info(f"No main topic found, using first topic: {topics[0].topic.slug}")
-            result = {
-                "slug": topics[0].topic.slug,
-                "title": topics[0].topic.title,
-                "id": topics[0].topic.id,
-                "is_main": True,
-            }
-            return result
-    else:
-        # Для Topic объектов (новый формат из selectinload)
-        # После смены на selectinload у нас просто список Topic объектов
-        if topics:
-            logger.info(f"Using first topic as main: {topics[0].slug}")
-            result = {
-                "slug": topics[0].slug,
-                "title": topics[0].title,
-                "id": topics[0].id,
-                "is_main": True,
-            }
-            return result
-
-    logger.warning("No valid topics found, returning default")
-    return {"slug": "notopic", "title": "no topic", "id": 0, "is_main": True}
-
-@mutation.field("unpublish_shout")
-@login_required
-async def unpublish_shout(_, info, shout_id: int):
-    """Снимает публикацию (shout) с публикации.
-
-    Предзагружает связанный черновик (draft) и его авторов/темы, чтобы избежать
-    ошибок при последующем доступе к ним в GraphQL.
-
-    Args:
-        shout_id: ID публикации для снятия с публикации
-
-    Returns:
-        dict: Снятая с публикации публикация или сообщение об ошибке
-    """
-    author_dict = info.context.get("author", {})
-    author_id = author_dict.get("id")
-    if not author_id:
-        # В идеале нужна проверка прав, имеет ли автор право снимать публикацию
-        return {"error": "Author ID is required"}
-
-    shout = None
-    with local_session() as session:
-        try:
-            # Загружаем Shout со всеми связями для правильного формирования ответа
-            shout = (
-                session.query(Shout)
-                .options(
-                    joinedload(Shout.authors),
-                    selectinload(Shout.topics)
-                )
-                .filter(Shout.id == shout_id)
-                .first()
-            )
-            
-            if not shout:
-                 logger.warning(f"Shout not found for unpublish: ID {shout_id}")
-                 return {"error": "Shout not found"} 
-            
-            # Если у публикации есть связанный черновик, загружаем его с relationships
-            if shout.draft:
-                # Отдельно загружаем черновик с его связями
-                draft = (
-                    session.query(Draft)
-                    .options(
-                        selectinload(Draft.authors),
-                        selectinload(Draft.topics)
-                    )
-                    .filter(Draft.id == shout.draft)
-                    .first()
-                )
-                
-                # Связываем черновик с публикацией вручную для доступа через API
-                if draft:
-                    shout.draft_obj = draft
-
-            # TODO: Добавить проверку прав доступа, если необходимо
-            # if author_id not in [a.id for a in shout.authors]: # Требует selectinload(Shout.authors) выше
-            #    logger.warning(f"Author {author_id} denied unpublishing shout {shout_id}")
-            #    return {"error": "Access denied"}
-
-            # Запоминаем старый slug и id для формирования поля publication
-            shout_slug = shout.slug
-            shout_id_for_publication = shout.id
-
-            # Снимаем с публикации (устанавливаем published_at в None)
-            shout.published_at = None
-            session.commit()
-            
-            # Формируем полноценный словарь для ответа
-            shout_dict = shout.dict()
-            
-            # Добавляем связанные данные
-            shout_dict["topics"] = (
-                [
-                    {"id": topic.id, "slug": topic.slug, "title": topic.title}
-                    for topic in shout.topics
-                ]
-                if shout.topics
-                else []
-            )
-            
-            # Добавляем main_topic 
-            shout_dict["main_topic"] = get_main_topic(shout.topics)
-            
-            # Добавляем авторов
-            shout_dict["authors"] = (
-                [
-                    {"id": author.id, "name": author.name, "slug": author.slug}
-                    for author in shout.authors
-                ]
-                if shout.authors
-                else []
-            )
-            
-            # Важно! Обновляем поле publication, отражая состояние "снят с публикации"
-            shout_dict["publication"] = {
-                "id": shout_id_for_publication,
-                "slug": shout_slug,
-                "published_at": None  # Ключевое изменение - устанавливаем published_at в None
-            }
-
-            # Инвалидация кэша
-            try:
-                cache_keys = [
-                    "feed",  # лента
-                    f"author_{author_id}",  # публикации автора
-                    "random_top",  # случайные топовые
-                    "unrated",  # неоцененные
-                ]
-                await invalidate_shout_related_cache(shout, author_id) 
-                await invalidate_shouts_cache(cache_keys)
-                logger.info(f"Cache invalidated after unpublishing shout {shout_id}")
-            except Exception as cache_err:
-                 logger.error(f"Failed to invalidate cache for unpublish shout {shout_id}: {cache_err}")
-
-        except Exception as e: 
-            session.rollback()
-            logger.error(f"Failed to unpublish shout {shout_id}: {e}", exc_info=True) 
-            return {"error": f"Failed to unpublish shout: {str(e)}"}
-
-    # Возвращаем сформированный словарь вместо объекта
-    logger.info(f"Shout {shout_id} unpublished successfully by author {author_id}")
-    return {"shout": shout_dict}
--- a/resolvers/feed.py
+++ b/resolvers/feed.py
@@ -1,198 +0,0 @@
-from typing import List
-
-from sqlalchemy import and_, select
-
-from orm.author import Author, AuthorFollower
-from orm.shout import Shout, ShoutAuthor, ShoutReactionsFollower, ShoutTopic
-from orm.topic import Topic, TopicFollower
-from resolvers.reader import (
-    apply_options,
-    get_shouts_with_links,
-    has_field,
-    query_with_stat,
-)
-from services.auth import login_required
-from services.db import local_session
-from services.schema import query
-from utils.logger import root_logger as logger
-
-
-@query.field("load_shouts_coauthored")
-@login_required
-async def load_shouts_coauthored(_, info, options):
-    """
-    Загрузка публикаций, написанных в соавторстве с пользователем.
-
-    :param info: Информаци о контексте GraphQL.
-    :param options: Опции фильтрации и сортировки.
-    :return: Список публикаций в соавтостве.
-    """
-    author_id = info.context.get("author", {}).get("id")
-    if not author_id:
-        return []
-    q = query_with_stat(info)
-    q = q.filter(Shout.authors.any(id=author_id))
-    q, limit, offset = apply_options(q, options)
-    return get_shouts_with_links(info, q, limit, offset=offset)
-
-
-@query.field("load_shouts_discussed")
-@login_required
-async def load_shouts_discussed(_, info, options):
-    """
-    Загрузка публикаций, которые обсуждались пользователем.
-
-    :param info: Информация о контексте GraphQL.
-    :param options: Опции фильтрации и сортировки.
-    :return: Список публикаций, обсужденых пользователем.
-    """
-    author_id = info.context.get("author", {}).get("id")
-    if not author_id:
-        return []
-    q = query_with_stat(info)
-    options["filters"]["commented"] = True
-    q, limit, offset = apply_options(q, options, author_id)
-    return get_shouts_with_links(info, q, limit, offset=offset)
-
-
-def shouts_by_follower(info, follower_id: int, options):
-    """
-    Загружает публикации, на которые подписан автор.
-
-    - по авторам
-    - по темам
-    - по реакциям
-
-    :param info: Информация о контексте GraphQL.
-    :param follower_id: Идентификатор автора.
-    :param options: Опции фильтрации и сортировки.
-    :return: Список публикаций.
-    """
-    q = query_with_stat(info)
-    reader_followed_authors = select(AuthorFollower.author).where(AuthorFollower.follower == follower_id)
-    reader_followed_topics = select(TopicFollower.topic).where(TopicFollower.follower == follower_id)
-    reader_followed_shouts = select(ShoutReactionsFollower.shout).where(ShoutReactionsFollower.follower == follower_id)
-    followed_subquery = (
-        select(Shout.id)
-        .join(ShoutAuthor, ShoutAuthor.shout == Shout.id)
-        .join(ShoutTopic, ShoutTopic.shout == Shout.id)
-        .where(
-            ShoutAuthor.author.in_(reader_followed_authors)
-            | ShoutTopic.topic.in_(reader_followed_topics)
-            | Shout.id.in_(reader_followed_shouts)
-        )
-        .scalar_subquery()
-    )
-    q = q.filter(Shout.id.in_(followed_subquery))
-    q, limit, offset = apply_options(q, options)
-    shouts = get_shouts_with_links(info, q, limit, offset=offset)
-    return shouts
-
-
-@query.field("load_shouts_followed_by")
-async def load_shouts_followed_by(_, info, slug: str, options) -> List[Shout]:
-    """
-    Загружает публикации, на которые подписан автор по slug.
-
-    :param info: Информация о контексте GraphQL.
-    :param slug: Slug автора.
-    :param options: Опции фильтрации и сортировки.
-    :return: Список публикаций.
-    """
-    with local_session() as session:
-        author = session.query(Author).filter(Author.slug == slug).first()
-        if author:
-            follower_id = author.dict()["id"]
-            shouts = shouts_by_follower(info, follower_id, options)
-            return shouts
-    return []
-
-
-@query.field("load_shouts_feed")
-@login_required
-async def load_shouts_feed(_, info, options) -> List[Shout]:
-    """
-    Загружает публикации, на которые подписан авторизованный пользователь.
-
-    :param info: Информация о контексте GraphQL.
-    :param options: Опции фильтрации и сортировки.
-    :return: Список публикаций.
-    """
-    author_id = info.context.get("author", {}).get("id")
-    return shouts_by_follower(info, author_id, options) if author_id else []
-
-
-@query.field("load_shouts_authored_by")
-async def load_shouts_authored_by(_, info, slug: str, options) -> List[Shout]:
-    """
-    Загружает публикации, написанные автором по slug.
-
-    :param info: Информация о контексте GraphQL.
-    :param slug: Slug автора.
-    :param options: Опции фильтрации и сортировки.
-    :return: Список публикаций.
-    """
-    with local_session() as session:
-        author = session.query(Author).filter(Author.slug == slug).first()
-        if author:
-            try:
-                author_id: int = author.dict()["id"]
-                q = (
-                    query_with_stat(info)
-                    if has_field(info, "stat")
-                    else select(Shout).filter(and_(Shout.published_at.is_not(None), Shout.deleted_at.is_(None)))
-                )
-                q = q.filter(Shout.authors.any(id=author_id))
-                q, limit, offset = apply_options(q, options, author_id)
-                shouts = get_shouts_with_links(info, q, limit, offset=offset)
-                return shouts
-            except Exception as error:
-                logger.debug(error)
-    return []
-
-
-@query.field("load_shouts_with_topic")
-async def load_shouts_with_topic(_, info, slug: str, options) -> List[Shout]:
-    """
-    Загружает публикации, связанные с темой по slug.
-
-    :param info: Информация о контексте GraphQL.
-    :param slug: Slug темы.
-    :param options: Опции фильтрации и сортировки.
-    :return: Список публикаций.
-    """
-    with local_session() as session:
-        topic = session.query(Topic).filter(Topic.slug == slug).first()
-        if topic:
-            try:
-                topic_id: int = topic.dict()["id"]
-                q = (
-                    query_with_stat(info)
-                    if has_field(info, "stat")
-                    else select(Shout).filter(and_(Shout.published_at.is_not(None), Shout.deleted_at.is_(None)))
-                )
-                q = q.filter(Shout.topics.any(id=topic_id))
-                q, limit, offset = apply_options(q, options)
-                shouts = get_shouts_with_links(info, q, limit, offset=offset)
-                return shouts
-            except Exception as error:
-                logger.debug(error)
-    return []
-
-
-def apply_filters(q, filters):
-    """
-    Применяет фильтры к запросу
-    """
-    logger.info(f"Applying filters: {filters}")
-
-    if filters.get("published"):
-        q = q.filter(Shout.published_at.is_not(None))
-        logger.info("Added published filter")
-
-    if filters.get("topic"):
-        topic_slug = filters["topic"]
-        q = q.join(ShoutTopic).join(Topic).filter(Topic.slug == topic_slug)
-        logger.info(f"Added topic filter: {topic_slug}")
-
-    return q
--- a/resolvers/follower.py
+++ b/resolvers/follower.py
@@ -1,222 +0,0 @@
-from typing import List
-
-from graphql import GraphQLError
-from sqlalchemy import select
-from sqlalchemy.sql import and_
-
-from cache.cache import (
-    cache_author,
-    cache_topic,
-    get_cached_follower_authors,
-    get_cached_follower_topics,
-)
-from orm.author import Author, AuthorFollower
-from orm.community import Community, CommunityFollower
-from orm.reaction import Reaction
-from orm.shout import Shout, ShoutReactionsFollower
-from orm.topic import Topic, TopicFollower
-from resolvers.stat import get_with_stat
-from services.auth import login_required
-from services.db import local_session
-from services.notify import notify_follower
-from services.schema import mutation, query
-from utils.logger import root_logger as logger
-
-
-@mutation.field("follow")
-@login_required
-async def follow(_, info, what, slug="", entity_id=0):
-    logger.debug("Начало выполнения функции 'follow'")
-    user_id = info.context.get("user_id")
-    follower_dict = info.context.get("author")
-    logger.debug(f"follower: {follower_dict}")
-
-    if not user_id or not follower_dict:
-        return GraphQLError("unauthorized")
-
-    follower_id = follower_dict.get("id")
-    logger.debug(f"follower_id: {follower_id}")
-
-    entity_classes = {
-        "AUTHOR": (Author, AuthorFollower, get_cached_follower_authors, cache_author),
-        "TOPIC": (Topic, TopicFollower, get_cached_follower_topics, cache_topic),
-        "COMMUNITY": (Community, CommunityFollower, None, None),  # Нет методов кэша для сообщества
-        "SHOUT": (Shout, ShoutReactionsFollower, None, None),  # Нет методов кэша для shout
-    }
-
-    if what not in entity_classes:
-        logger.error(f"Неверный тип для следования: {what}")
-        return {"error": "invalid follow type"}
-
-    entity_class, follower_class, get_cached_follows_method, cache_method = entity_classes[what]
-    entity_type = what.lower()
-    entity_dict = None
-
-    try:
-        logger.debug("Попытка получить сущность из базы данных")
-        with local_session() as session:
-            entity_query = select(entity_class).filter(entity_class.slug == slug)
-            entities = get_with_stat(entity_query)
-            [entity] = entities
-            if not entity:
-                logger.warning(f"{what.lower()} не найден по slug: {slug}")
-                return {"error": f"{what.lower()} not found"}
-            if not entity_id and entity:
-                entity_id = entity.id
-            entity_dict = entity.dict()
-            logger.debug(f"entity_id: {entity_id}, entity_dict: {entity_dict}")
-
-        if entity_id:
-            logger.debug("Проверка существующей подписки")
-            with local_session() as session:
-                existing_sub = (
-                    session.query(follower_class)
-                    .filter(follower_class.follower == follower_id, getattr(follower_class, entity_type) == entity_id)
-                    .first()
-                )
-                if existing_sub:
-                    logger.info(f"Пользователь {follower_id} уже подписан на {what.lower()} с ID {entity_id}")
-                else:
-                    logger.debug("Добавление новой записи в базу данных")
-                    sub = follower_class(follower=follower_id, **{entity_type: entity_id})
-                    logger.debug(f"Создан объект подписки: {sub}")
-                    session.add(sub)
-                    session.commit()
-                    logger.info(f"Пользователь {follower_id} подписался на {what.lower()} с ID {entity_id}")
-
-            follows = None
-            if cache_method:
-                logger.debug("Обновление кэша")
-                await cache_method(entity_dict)
-            if get_cached_follows_method:
-                logger.debug("Получение подписок из кэша")
-                existing_follows = await get_cached_follows_method(follower_id)
-                follows = [*existing_follows, entity_dict] if not existing_sub else existing_follows
-                logger.debug("Обновлен список подписок")
-
-            if what == "AUTHOR" and not existing_sub:
-                logger.debug("Отправка уведомления автору о подписке")
-                await notify_follower(follower=follower_dict, author_id=entity_id, action="follow")
-
-    except Exception as exc:
-        logger.exception("Произошла ошибка в функции 'follow'")
-        return {"error": str(exc)}
-
-    return {f"{what.lower()}s": follows}
-
-
-@mutation.field("unfollow")
-@login_required
-async def unfollow(_, info, what, slug="", entity_id=0):
-    logger.debug("Начало выполнения функции 'unfollow'")
-    user_id = info.context.get("user_id")
-    follower_dict = info.context.get("author")
-    logger.debug(f"follower: {follower_dict}")
-
-    if not user_id or not follower_dict:
-        logger.warning("Неавторизованный доступ при попытке отписаться")
-        return {"error": "unauthorized"}
-
-    follower_id = follower_dict.get("id")
-    logger.debug(f"follower_id: {follower_id}")
-
-    entity_classes = {
-        "AUTHOR": (Author, AuthorFollower, get_cached_follower_authors, cache_author),
-        "TOPIC": (Topic, TopicFollower, get_cached_follower_topics, cache_topic),
-        "COMMUNITY": (Community, CommunityFollower, None, None),  # Нет методов кэша для сообщества
-        "SHOUT": (Shout, ShoutReactionsFollower, None, None),  # Нет методов кэша для shout
-    }
-
-    if what not in entity_classes:
-        logger.error(f"Неверный тип для отписки: {what}")
-        return {"error": "invalid unfollow type"}
-
-    entity_class, follower_class, get_cached_follows_method, cache_method = entity_classes[what]
-    entity_type = what.lower()
-    follows = []
-    error = None
-
-    try:
-        logger.debug("Попытка получить сущность из базы данных")
-        with local_session() as session:
-            entity = session.query(entity_class).filter(entity_class.slug == slug).first()
-            logger.debug(f"Полученная сущность: {entity}")
-            if not entity:
-                logger.warning(f"{what.lower()} не найден по slug: {slug}")
-                return {"error": f"{what.lower()} not found"}
-            if entity and not entity_id:
-                entity_id = entity.id
-                logger.debug(f"entity_id: {entity_id}")
-
-            sub = (
-                session.query(follower_class)
-                .filter(
-                    and_(
-                        getattr(follower_class, "follower") == follower_id,
-                        getattr(follower_class, entity_type) == entity_id,
-                    )
-                )
-                .first()
-            )
-            logger.debug(f"Найдена подписка для удаления: {sub}")
-            if sub:
-                session.delete(sub)
-                session.commit()
-                logger.info(f"Пользователь {follower_id} отписался от {what.lower()} с ID {entity_id}")
-
-                if cache_method:
-                    logger.debug("Обновление кэша после отписки")
-                    await cache_method(entity.dict())
-                if get_cached_follows_method:
-                    logger.debug("Получение подписок из кэша")
-                    existing_follows = await get_cached_follows_method(follower_id)
-                    follows = filter(lambda x: x["id"] != entity_id, existing_follows)
-                    logger.debug("Обновлен список подписок")
-
-                if what == "AUTHOR":
-                    logger.debug("Отправка уведомления автору об отписке")
-                    await notify_follower(follower=follower_dict, author_id=entity_id, action="unfollow")
-            else:
-                return {"error": "following was not found", f"{entity_type}s": follows}
-
-    except Exception as exc:
-        logger.exception("Произошла ошибка в функции 'unfollow'")
-        import traceback
-
-        traceback.print_exc()
-        return {"error": str(exc)}
-
-    # logger.debug(f"Функция 'unfollow' завершена успешно с результатом: {entity_type}s={follows}, error={error}")
-    return {f"{entity_type}s": follows, "error": error}
-
-
-@query.field("get_shout_followers")
-def get_shout_followers(_, _info, slug: str = "", shout_id: int | None = None) -> List[Author]:
-    logger.debug("Начало выполнения функции 'get_shout_followers'")
-    followers = []
-    try:
-        with local_session() as session:
-            shout = None
-            if slug:
-                shout = session.query(Shout).filter(Shout.slug == slug).first()
-                logger.debug(f"Найден shout по slug: {slug} -> {shout}")
-            elif shout_id:
-                shout = session.query(Shout).filter(Shout.id == shout_id).first()
-                logger.debug(f"Найден shout по ID: {shout_id} -> {shout}")
-
-            if shout:
-                reactions = session.query(Reaction).filter(Reaction.shout == shout.id).all()
-                logger.debug(f"Полученные реакции для shout ID {shout.id}: {reactions}")
-                for r in reactions:
-                    followers.append(r.created_by)
-                    logger.debug(f"Добавлен follower: {r.created_by}")
-
-    except Exception as _exc:
-        import traceback
-
-        traceback.print_exc()
-        logger.exception("Произошла ошибка в функции 'get_shout_followers'")
-        return []
-
-    # logger.debug(f"Функция 'get_shout_followers' завершена с {len(followers)} подписчиками")
-    return followers
--- a/resolvers/inbox/chats.py
+++ b/resolvers/inbox/chats.py
@@ -0,0 +1,113 @@
+import json
+import uuid
+from datetime import datetime, timezone
+
+from auth.authenticate import login_required
+from auth.credentials import AuthCredentials
+from base.redis import redis
+from base.resolvers import mutation
+from validations.inbox import Chat
+
+
+@mutation.field("updateChat")
+@login_required
+async def update_chat(_, info, chat_new: Chat):
+    """
+    updating chat
+    requires info["request"].user.slug to be in chat["admins"]
+
+    :param info: GraphQLInfo with request
+    :param chat_new: dict with chat data
+    :return: Result { error chat }
+    """
+    auth: AuthCredentials = info.context["request"].auth
+    chat_id = chat_new["id"]
+    chat = await redis.execute("GET", f"chats/{chat_id}")
+    if not chat:
+        return {"error": "chat not exist"}
+    chat = dict(json.loads(chat))
+
+    # TODO
+    if auth.user_id in chat["admins"]:
+        chat.update(
+            {
+                "title": chat_new.get("title", chat["title"]),
+                "description": chat_new.get("description", chat["description"]),
+                "updatedAt": int(datetime.now(tz=timezone.utc).timestamp()),
+                "admins": chat_new.get("admins", chat.get("admins") or []),
+                "users": chat_new.get("users", chat["users"]),
+            }
+        )
+    await redis.execute("SET", f"chats/{chat.id}", json.dumps(chat))
+    await redis.execute("COMMIT")
+
+    return {"error": None, "chat": chat}
+
+
+@mutation.field("createChat")
+@login_required
+async def create_chat(_, info, title="", members=[]):
+    auth: AuthCredentials = info.context["request"].auth
+    chat = {}
+    print("create_chat members: %r" % members)
+    if auth.user_id not in members:
+        members.append(int(auth.user_id))
+
+    # reuse chat craeted before if exists
+    if len(members) == 2 and title == "":
+        chat = None
+        print(members)
+        chatset1 = await redis.execute("SMEMBERS", f"chats_by_user/{members[0]}")
+        if not chatset1:
+            chatset1 = set([])
+        print(chatset1)
+        chatset2 = await redis.execute("SMEMBERS", f"chats_by_user/{members[1]}")
+        if not chatset2:
+            chatset2 = set([])
+        print(chatset2)
+        chatset = chatset1.intersection(chatset2)
+        print(chatset)
+        for c in chatset:
+            chat = await redis.execute("GET", f"chats/{c.decode('utf-8')}")
+            if chat:
+                chat = json.loads(chat)
+                if chat["title"] == "":
+                    print("[inbox] createChat found old chat")
+                    print(chat)
+                    break
+        if chat:
+            return {"chat": chat, "error": "existed"}
+
+    chat_id = str(uuid.uuid4())
+    chat = {
+        "id": chat_id,
+        "users": members,
+        "title": title,
+        "createdBy": auth.user_id,
+        "createdAt": int(datetime.now(tz=timezone.utc).timestamp()),
+        "updatedAt": int(datetime.now(tz=timezone.utc).timestamp()),
+        "admins": members if (len(members) == 2 and title == "") else [],
+    }
+
+    for m in members:
+        await redis.execute("SADD", f"chats_by_user/{m}", chat_id)
+    await redis.execute("SET", f"chats/{chat_id}", json.dumps(chat))
+    await redis.execute("SET", f"chats/{chat_id}/next_message_id", str(0))
+    await redis.execute("COMMIT")
+    return {"error": None, "chat": chat}
+
+
+@mutation.field("deleteChat")
+@login_required
+async def delete_chat(_, info, chat_id: str):
+    auth: AuthCredentials = info.context["request"].auth
+
+    chat = await redis.execute("GET", f"/chats/{chat_id}")
+    if chat:
+        chat = dict(json.loads(chat))
+        if auth.user_id in chat["admins"]:
+            await redis.execute("DEL", f"chats/{chat_id}")
+            await redis.execute("SREM", "chats_by_user/" + str(auth.user_id), chat_id)
+            await redis.execute("COMMIT")
+    else:
+        return {"error": "chat not exist"}
--- a/resolvers/inbox/load.py
+++ b/resolvers/inbox/load.py
@@ -0,0 +1,138 @@
+import json
+
+from auth.authenticate import login_required
+from auth.credentials import AuthCredentials
+from base.orm import local_session
+from base.redis import redis
+from base.resolvers import query
+from orm.user import User
+from resolvers.zine.profile import followed_authors
+
+from .unread import get_unread_counter
+
+# from datetime import datetime, timedelta, timezone
+
+
+async def load_messages(chat_id: str, limit: int = 5, offset: int = 0, ids=[]):
+    """load :limit messages for :chat_id with :offset"""
+    messages = []
+    message_ids = []
+    if ids:
+        message_ids += ids
+    try:
+        if limit:
+            mids = await redis.lrange(f"chats/{chat_id}/message_ids", offset, offset + limit)
+            mids = [mid.decode("utf-8") for mid in mids]
+            message_ids += mids
+    except Exception as e:
+        print(e)
+    if message_ids:
+        message_keys = [f"chats/{chat_id}/messages/{mid}" for mid in message_ids]
+        messages = await redis.mget(*message_keys)
+        messages = [json.loads(msg.decode("utf-8")) for msg in messages]
+        replies = []
+        for m in messages:
+            rt = m.get("replyTo")
+            if rt:
+                rt = int(rt)
+                if rt not in message_ids:
+                    replies.append(rt)
+        if replies:
+            messages += await load_messages(chat_id, limit=0, ids=replies)
+    return messages
+
+
+@query.field("loadChats")
+@login_required
+async def load_chats(_, info, limit: int = 50, offset: int = 0):
+    """load :limit chats of current user with :offset"""
+    auth: AuthCredentials = info.context["request"].auth
+
+    cids = await redis.execute("SMEMBERS", "chats_by_user/" + str(auth.user_id))
+    if cids:
+        cids = list(cids)[offset : offset + limit]
+    if not cids:
+        print("[inbox.load] no chats were found")
+        cids = []
+    onliners = await redis.execute("SMEMBERS", "users-online")
+    if not onliners:
+        onliners = []
+    chats = []
+    for cid in cids:
+        cid = cid.decode("utf-8")
+        c = await redis.execute("GET", "chats/" + cid)
+        if c:
+            c = dict(json.loads(c))
+            c["messages"] = await load_messages(cid, 5, 0)
+            c["unread"] = await get_unread_counter(cid, auth.user_id)
+            with local_session() as session:
+                c["members"] = []
+                for uid in c["users"]:
+                    a = session.query(User).where(User.id == uid).first()
+                    if a:
+                        c["members"].append(
+                            {
+                                "id": a.id,
+                                "slug": a.slug,
+                                "userpic": a.userpic,
+                                "name": a.name,
+                                "lastSeen": a.lastSeen,
+                                "online": a.id in onliners,
+                            }
+                        )
+                chats.append(c)
+    return {"chats": chats, "error": None}
+
+
+@query.field("loadMessagesBy")
+@login_required
+async def load_messages_by(_, info, by, limit: int = 10, offset: int = 0):
+    """load :limit messages of :chat_id with :offset"""
+
+    auth: AuthCredentials = info.context["request"].auth
+    userchats = await redis.execute("SMEMBERS", "chats_by_user/" + str(auth.user_id))
+    userchats = [c.decode("utf-8") for c in userchats]
+    # print('[inbox] userchats: %r' % userchats)
+    if userchats:
+        # print('[inbox] loading messages by...')
+        messages = []
+        by_chat = by.get("chat")
+        if by_chat in userchats:
+            chat = await redis.execute("GET", f"chats/{by_chat}")
+            # print(chat)
+            if not chat:
+                return {"messages": [], "error": "chat not exist"}
+            # everyone's messages in filtered chat
+            messages = await load_messages(by_chat, limit, offset)
+        return {"messages": sorted(list(messages), key=lambda m: m["createdAt"]), "error": None}
+    else:
+        return {"error": "Cannot access messages of this chat"}
+
+
+@query.field("loadRecipients")
+async def load_recipients(_, info, limit=50, offset=0):
+    chat_users = []
+    auth: AuthCredentials = info.context["request"].auth
+    onliners = await redis.execute("SMEMBERS", "users-online")
+    if not onliners:
+        onliners = []
+    try:
+        chat_users += await followed_authors(auth.user_id)
+        limit = limit - len(chat_users)
+    except Exception:
+        pass
+    with local_session() as session:
+        chat_users += session.query(User).where(User.emailConfirmed).limit(limit).offset(offset)
+        members = []
+        for a in chat_users:
+            members.append(
+                {
+                    "id": a.id,
+                    "slug": a.slug,
+                    "userpic": a.userpic,
+                    "name": a.name,
+                    "lastSeen": a.lastSeen,
+                    "online": a.id in onliners,
+                }
+            )
+    return {"members": members, "error": None}
--- a/resolvers/inbox/messages.py
+++ b/resolvers/inbox/messages.py
@@ -0,0 +1,129 @@
+import json
+from datetime import datetime, timezone
+
+from auth.authenticate import login_required
+from auth.credentials import AuthCredentials
+from base.redis import redis
+from base.resolvers import mutation
+from services.following import FollowingManager, FollowingResult
+
+
+@mutation.field("createMessage")
+@login_required
+async def create_message(_, info, chat: str, body: str, replyTo=None):
+    """create message with :body for :chat_id replying to :replyTo optionally"""
+    auth: AuthCredentials = info.context["request"].auth
+
+    chat = await redis.execute("GET", f"chats/{chat}")
+    if not chat:
+        return {"error": "chat is not exist"}
+    else:
+        chat_dict = dict(json.loads(chat))
+        message_id = await redis.execute("GET", f"chats/{chat_dict['id']}/next_message_id")
+        message_id = int(message_id)
+        new_message = {
+            "chatId": chat_dict["id"],
+            "id": message_id,
+            "author": auth.user_id,
+            "body": body,
+            "createdAt": int(datetime.now(tz=timezone.utc).timestamp()),
+        }
+        if replyTo:
+            new_message["replyTo"] = replyTo
+        chat_dict["updatedAt"] = new_message["createdAt"]
+        await redis.execute("SET", f"chats/{chat_dict['id']}", json.dumps(chat))
+        print(f"[inbox] creating message {new_message}")
+        await redis.execute(
+            "SET", f"chats/{chat_dict['id']}/messages/{message_id}", json.dumps(new_message)
+        )
+        await redis.execute("LPUSH", f"chats/{chat_dict['id']}/message_ids", str(message_id))
+        await redis.execute("SET", f"chats/{chat_dict['id']}/next_message_id", str(message_id + 1))
+
+        users = chat_dict["users"]
+        for user_slug in users:
+            await redis.execute(
+                "LPUSH", f"chats/{chat_dict['id']}/unread/{user_slug}", str(message_id)
+            )
+
+        result = FollowingResult("NEW", "chat", new_message)
+        await FollowingManager.push("chat", result)
+
+        return {"message": new_message, "error": None}
+
+
+@mutation.field("updateMessage")
+@login_required
+async def update_message(_, info, chat_id: str, message_id: int, body: str):
+    auth: AuthCredentials = info.context["request"].auth
+
+    chat = await redis.execute("GET", f"chats/{chat_id}")
+    if not chat:
+        return {"error": "chat not exist"}
+
+    message = await redis.execute("GET", f"chats/{chat_id}/messages/{message_id}")
+    if not message:
+        return {"error": "message  not exist"}
+
+    message = json.loads(message)
+    if message["author"] != auth.user_id:
+        return {"error": "access denied"}
+
+    message["body"] = body
+    message["updatedAt"] = int(datetime.now(tz=timezone.utc).timestamp())
+
+    await redis.execute("SET", f"chats/{chat_id}/messages/{message_id}", json.dumps(message))
+
+    result = FollowingResult("UPDATED", "chat", message)
+    await FollowingManager.push("chat", result)
+
+    return {"message": message, "error": None}
+
+
+@mutation.field("deleteMessage")
+@login_required
+async def delete_message(_, info, chat_id: str, message_id: int):
+    auth: AuthCredentials = info.context["request"].auth
+
+    chat = await redis.execute("GET", f"chats/{chat_id}")
+    if not chat:
+        return {"error": "chat not exist"}
+    chat = json.loads(chat)
+
+    message = await redis.execute("GET", f"chats/{chat_id}/messages/{str(message_id)}")
+    if not message:
+        return {"error": "message  not exist"}
+    message = json.loads(message)
+    if message["author"] != auth.user_id:
+        return {"error": "access denied"}
+
+    await redis.execute("LREM", f"chats/{chat_id}/message_ids", 0, str(message_id))
+    await redis.execute("DEL", f"chats/{chat_id}/messages/{str(message_id)}")
+
+    users = chat["users"]
+    for user_id in users:
+        await redis.execute("LREM", f"chats/{chat_id}/unread/{user_id}", 0, str(message_id))
+
+    result = FollowingResult("DELETED", "chat", message)
+    await FollowingManager.push(result)
+
+    return {}
+
+
+@mutation.field("markAsRead")
+@login_required
+async def mark_as_read(_, info, chat_id: str, messages: [int]):
+    auth: AuthCredentials = info.context["request"].auth
+
+    chat = await redis.execute("GET", f"chats/{chat_id}")
+    if not chat:
+        return {"error": "chat not exist"}
+
+    chat = json.loads(chat)
+    users = set(chat["users"])
+    if auth.user_id not in users:
+        return {"error": "access denied"}
+
+    for message_id in messages:
+        await redis.execute("LREM", f"chats/{chat_id}/unread/{auth.user_id}", 0, str(message_id))
+
+    return {"error": None}
--- a/resolvers/inbox/search.py
+++ b/resolvers/inbox/search.py
@@ -0,0 +1,96 @@
+import json
+from datetime import datetime, timedelta, timezone
+
+from auth.authenticate import login_required
+from auth.credentials import AuthCredentials
+from base.orm import local_session
+from base.redis import redis
+from base.resolvers import query
+from orm.user import AuthorFollower, User
+from resolvers.inbox.load import load_messages
+
+
+@query.field("searchRecipients")
+@login_required
+async def search_recipients(_, info, query: str, limit: int = 50, offset: int = 0):
+    result = []
+    # TODO: maybe redis scan?
+    auth: AuthCredentials = info.context["request"].auth
+    talk_before = await redis.execute("GET", f"/chats_by_user/{auth.user_id}")
+    if talk_before:
+        talk_before = list(json.loads(talk_before))[offset : offset + limit]
+        for chat_id in talk_before:
+            members = await redis.execute("GET", f"/chats/{chat_id}/users")
+            if members:
+                members = list(json.loads(members))
+                for member in members:
+                    if member.startswith(query):
+                        if member not in result:
+                            result.append(member)
+
+    more_amount = limit - len(result)
+
+    with local_session() as session:
+        # followings
+        result += (
+            session.query(AuthorFollower.author)
+            .join(User, User.id == AuthorFollower.follower)
+            .where(User.slug.startswith(query))
+            .offset(offset + len(result))
+            .limit(more_amount)
+        )
+
+        more_amount = limit
+        # followers
+        result += (
+            session.query(AuthorFollower.follower)
+            .join(User, User.id == AuthorFollower.author)
+            .where(User.slug.startswith(query))
+            .offset(offset + len(result))
+            .limit(offset + len(result) + limit)
+        )
+    return {"members": list(result), "error": None}
+
+
+@query.field("searchMessages")
+@login_required
+async def search_user_chats(by, messages, user_id: int, limit, offset):
+    cids = set([])
+    cids.union(set(await redis.execute("SMEMBERS", "chats_by_user/" + str(user_id))))
+    messages = []
+
+    by_author = by.get("author")
+    if by_author:
+        # all author's messages
+        cids.union(set(await redis.execute("SMEMBERS", f"chats_by_user/{by_author}")))
+        # author's messages in filtered chat
+        messages.union(set(filter(lambda m: m["author"] == by_author, list(messages))))
+        for c in cids:
+            c = c.decode("utf-8")
+            messages = await load_messages(c, limit, offset)
+
+    body_like = by.get("body")
+    if body_like:
+        # search in all messages in all user's chats
+        for c in cids:
+            # FIXME: use redis scan here
+            c = c.decode("utf-8")
+            mmm = await load_messages(c, limit, offset)
+            for m in mmm:
+                if body_like in m["body"]:
+                    messages.add(m)
+        else:
+            # search in chat's messages
+            messages.extend(filter(lambda m: body_like in m["body"], list(messages)))
+
+    days = by.get("days")
+    if days:
+        messages.extend(
+            filter(
+                list(messages),
+                key=lambda m: (
+                    datetime.now(tz=timezone.utc) - int(m["createdAt"]) < timedelta(days=by["days"])
+                ),
+            )
+        )
+    return {"messages": messages, "error": None}
--- a/resolvers/inbox/unread.py
+++ b/resolvers/inbox/unread.py
@@ -0,0 +1,10 @@
+from base.redis import redis
+
+
+async def get_unread_counter(chat_id: str, user_id: int):
+    try:
+        unread = await redis.execute("LLEN", f"chats/{chat_id.decode('utf-8')}/unread/{user_id}")
+        if unread:
+            return unread
+    except Exception:
+        return 0
--- a/resolvers/notifications.py
+++ b/resolvers/notifications.py
@@ -0,0 +1,89 @@
+from sqlalchemy import and_, desc, select, update
+
+from auth.authenticate import login_required
+from auth.credentials import AuthCredentials
+from base.orm import local_session
+from base.resolvers import mutation, query
+from orm import Notification
+
+
+@query.field("loadNotifications")
+@login_required
+async def load_notifications(_, info, params=None):
+    if params is None:
+        params = {}
+
+    auth: AuthCredentials = info.context["request"].auth
+    user_id = auth.user_id
+
+    limit = params.get("limit", 50)
+    offset = params.get("offset", 0)
+
+    q = (
+        select(Notification)
+        .where(Notification.user == user_id)
+        .order_by(desc(Notification.createdAt))
+        .limit(limit)
+        .offset(offset)
+    )
+
+    notifications = []
+    with local_session() as session:
+        total_count = session.query(Notification).where(Notification.user == user_id).count()
+
+        total_unread_count = (
+            session.query(Notification)
+            .where(and_(Notification.user == user_id, Notification.seen == False))  # noqa: E712
+            .count()
+        )
+
+        for [notification] in session.execute(q):
+            notification.type = notification.type.name
+            notifications.append(notification)
+
+    return {
+        "notifications": notifications,
+        "totalCount": total_count,
+        "totalUnreadCount": total_unread_count,
+    }
+
+
+@mutation.field("markNotificationAsRead")
+@login_required
+async def mark_notification_as_read(_, info, notification_id: int):
+    auth: AuthCredentials = info.context["request"].auth
+    user_id = auth.user_id
+
+    with local_session() as session:
+        notification = (
+            session.query(Notification)
+            .where(and_(Notification.id == notification_id, Notification.user == user_id))
+            .one()
+        )
+        notification.seen = True
+        session.commit()
+
+    return {}
+
+
+@mutation.field("markAllNotificationsAsRead")
+@login_required
+async def mark_all_notifications_as_read(_, info):
+    auth: AuthCredentials = info.context["request"].auth
+    user_id = auth.user_id
+
+    statement = (
+        update(Notification)
+        .where(and_(Notification.user == user_id, Notification.seen == False))  # noqa: E712
+        .values(seen=True)
+    )
+
+    with local_session() as session:
+        try:
+            session.execute(statement)
+            session.commit()
+        except Exception as e:
+            session.rollback()
+            print(f"[mark_all_notifications_as_read] error: {str(e)}")
+
+    return {}
--- a/resolvers/notifier.py
+++ b/resolvers/notifier.py
@@ -1,316 +0,0 @@
-import time
-from typing import List, Tuple
-
-import orjson
-from sqlalchemy import and_, select
-from sqlalchemy.exc import SQLAlchemyError
-from sqlalchemy.orm import aliased
-from sqlalchemy.sql import not_
-
-from orm.author import Author
-from orm.notification import (
-    Notification,
-    NotificationAction,
-    NotificationEntity,
-    NotificationSeen,
-)
-from orm.shout import Shout
-from services.auth import login_required
-from services.db import local_session
-from services.schema import mutation, query
-from utils.logger import root_logger as logger
-
-
-def query_notifications(author_id: int, after: int = 0) -> Tuple[int, int, List[Tuple[Notification, bool]]]:
-    notification_seen_alias = aliased(NotificationSeen)
-    q = select(Notification, notification_seen_alias.viewer.label("seen")).outerjoin(
-        NotificationSeen,
-        and_(
-            NotificationSeen.viewer == author_id,
-            NotificationSeen.notification == Notification.id,
-        ),
-    )
-    if after:
-        q = q.filter(Notification.created_at > after)
-    q = q.group_by(NotificationSeen.notification, Notification.created_at)
-
-    with local_session() as session:
-        total = (
-            session.query(Notification)
-            .filter(
-                and_(
-                    Notification.action == NotificationAction.CREATE.value,
-                    Notification.created_at > after,
-                )
-            )
-            .count()
-        )
-
-        unread = (
-            session.query(Notification)
-            .filter(
-                and_(
-                    Notification.action == NotificationAction.CREATE.value,
-                    Notification.created_at > after,
-                    not_(Notification.seen),
-                )
-            )
-            .count()
-        )
-
-        notifications_result = session.execute(q)
-        notifications = []
-        for n, seen in notifications_result:
-            notifications.append((n, seen))
-
-    return total, unread, notifications
-
-
-def group_notification(thread, authors=None, shout=None, reactions=None, entity="follower", action="follow"):
-    reactions = reactions or []
-    authors = authors or []
-    return {
-        "thread": thread,
-        "authors": authors,
-        "updated_at": int(time.time()),
-        "shout": shout,
-        "reactions": reactions,
-        "entity": entity,
-        "action": action,
-    }
-
-
-def get_notifications_grouped(author_id: int, after: int = 0, limit: int = 10, offset: int = 0):
-    """
-    Retrieves notifications for a given author.
-
-    Args:
-        author_id (int): The ID of the author for whom notifications are retrieved.
-        after (int, optional): If provided, selects only notifications created after this timestamp will be considered.
-        limit (int, optional): The maximum number of groupa to retrieve.
-        offset (int, optional): offset
-
-    Returns:
-        Dict[str, NotificationGroup], int, int: A dictionary where keys are thread IDs
-        and values are NotificationGroup objects, unread and total amounts.
-
-    This function queries the database to retrieve notifications for the specified author, considering optional filters.
-    The result is a dictionary where each key is a thread ID, and the corresponding value is a NotificationGroup
-    containing information about the notifications within that thread.
-
-    NotificationGroup structure:
-    {
-        entity: str,        # Type of entity (e.g., 'reaction', 'shout', 'follower').
-        updated_at: int,    # Timestamp of the latest update in the thread.
-        shout: Optional[NotificationShout]
-        reactions: List[int],  # List of reaction ids within the thread.
-        authors: List[NotificationAuthor],  # List of authors involved in the thread.
-    }
-    """
-    total, unread, notifications = query_notifications(author_id, after)
-    groups_by_thread = {}
-    groups_amount = 0
-
-    for notification, seen in notifications:
-        if (groups_amount + offset) >= limit:
-            break
-
-        payload = orjson.loads(str(notification.payload))
-
-        if str(notification.entity) == NotificationEntity.SHOUT.value:
-            shout = payload
-            shout_id = shout.get("id")
-            author_id = shout.get("created_by")
-            thread_id = f"shout-{shout_id}"
-            with local_session() as session:
-                author = session.query(Author).filter(Author.id == author_id).first()
-                shout = session.query(Shout).filter(Shout.id == shout_id).first()
-                if author and shout:
-                    author = author.dict()
-                    shout = shout.dict()
-                    group = group_notification(
-                        thread_id,
-                        shout=shout,
-                        authors=[author],
-                        action=str(notification.action),
-                        entity=str(notification.entity),
-                    )
-                    groups_by_thread[thread_id] = group
-                    groups_amount += 1
-
-        elif str(notification.entity) == NotificationEntity.REACTION.value:
-            reaction = payload
-            if not isinstance(reaction, dict):
-                raise ValueError("reaction data is not consistent")
-            shout_id = reaction.get("shout")
-            author_id = reaction.get("created_by", 0)
-            if shout_id and author_id:
-                with local_session() as session:
-                    author = session.query(Author).filter(Author.id == author_id).first()
-                    shout = session.query(Shout).filter(Shout.id == shout_id).first()
-                    if shout and author:
-                        author = author.dict()
-                        shout = shout.dict()
-                        reply_id = reaction.get("reply_to")
-                        thread_id = f"shout-{shout_id}"
-                        if reply_id and reaction.get("kind", "").lower() == "comment":
-                            thread_id += f"{reply_id}"
-                        existing_group = groups_by_thread.get(thread_id)
-                        if existing_group:
-                            existing_group["seen"] = False
-                            existing_group["authors"].append(author_id)
-                            existing_group["reactions"] = existing_group["reactions"] or []
-                            existing_group["reactions"].append(reaction)
-                            groups_by_thread[thread_id] = existing_group
-                        else:
-                            group = group_notification(
-                                thread_id,
-                                authors=[author],
-                                shout=shout,
-                                reactions=[reaction],
-                                entity=str(notification.entity),
-                                action=str(notification.action),
-                            )
-                            if group:
-                                groups_by_thread[thread_id] = group
-                                groups_amount += 1
-
-            elif str(notification.entity) == "follower":
-                thread_id = "followers"
-                follower = orjson.loads(payload)
-                group = groups_by_thread.get(thread_id)
-                if group:
-                    if str(notification.action) == "follow":
-                        group["authors"].append(follower)
-                    elif str(notification.action) == "unfollow":
-                        follower_id = follower.get("id")
-                        for author in group["authors"]:
-                            if author.get("id") == follower_id:
-                                group["authors"].remove(author)
-                                break
-                else:
-                    group = group_notification(
-                        thread_id,
-                        authors=[follower],
-                        entity=str(notification.entity),
-                        action=str(notification.action),
-                    )
-                    groups_amount += 1
-                groups_by_thread[thread_id] = group
-    return groups_by_thread, unread, total
-
-
-@query.field("load_notifications")
-@login_required
-async def load_notifications(_, info, after: int, limit: int = 50, offset=0):
-    author_dict = info.context.get("author")
-    author_id = author_dict.get("id")
-    error = None
-    total = 0
-    unread = 0
-    notifications = []
-    try:
-        if author_id:
-            groups, unread, total = get_notifications_grouped(author_id, after, limit)
-            notifications = sorted(groups.values(), key=lambda group: group.updated_at, reverse=True)
-    except Exception as e:
-        error = e
-        logger.error(e)
-    return {
-        "notifications": notifications,
-        "total": total,
-        "unread": unread,
-        "error": error,
-    }
-
-
-@mutation.field("notification_mark_seen")
-@login_required
-async def notification_mark_seen(_, info, notification_id: int):
-    author_id = info.context.get("author", {}).get("id")
-    if author_id:
-        with local_session() as session:
-            try:
-                ns = NotificationSeen(notification=notification_id, viewer=author_id)
-                session.add(ns)
-                session.commit()
-            except SQLAlchemyError as e:
-                session.rollback()
-                logger.error(f"seen mutation failed: {e}")
-                return {"error": "cant mark as read"}
-    return {"error": None}
-
-
-@mutation.field("notifications_seen_after")
-@login_required
-async def notifications_seen_after(_, info, after: int):
-    # TODO: use latest loaded notification_id as input offset parameter
-    error = None
-    try:
-        author_id = info.context.get("author", {}).get("id")
-        if author_id:
-            with local_session() as session:
-                nnn = session.query(Notification).filter(and_(Notification.created_at > after)).all()
-                for n in nnn:
-                    try:
-                        ns = NotificationSeen(notification=n.id, viewer=author_id)
-                        session.add(ns)
-                        session.commit()
-                    except SQLAlchemyError:
-                        session.rollback()
-    except Exception as e:
-        print(e)
-        error = "cant mark as read"
-    return {"error": error}
-
-
-@mutation.field("notifications_seen_thread")
-@login_required
-async def notifications_seen_thread(_, info, thread: str, after: int):
-    error = None
-    author_id = info.context.get("author", {}).get("id")
-    if author_id:
-        [shout_id, reply_to_id] = thread.split(":")
-        with local_session() as session:
-            # TODO: handle new follower and new shout notifications
-            new_reaction_notifications = (
-                session.query(Notification)
-                .filter(
-                    Notification.action == "create",
-                    Notification.entity == "reaction",
-                    Notification.created_at > after,
-                )
-                .all()
-            )
-            removed_reaction_notifications = (
-                session.query(Notification)
-                .filter(
-                    Notification.action == "delete",
-                    Notification.entity == "reaction",
-                    Notification.created_at > after,
-                )
-                .all()
-            )
-            exclude = set()
-            for nr in removed_reaction_notifications:
-                reaction = orjson.loads(str(nr.payload))
-                reaction_id = reaction.get("id")
-                exclude.add(reaction_id)
-            for n in new_reaction_notifications:
-                reaction = orjson.loads(str(n.payload))
-                reaction_id = reaction.get("id")
-                if (
-                    reaction_id not in exclude
-                    and reaction.get("shout") == shout_id
-                    and reaction.get("reply_to") == reply_to_id
-                ):
-                    try:
-                        ns = NotificationSeen(notification=n.id, viewer=author_id)
-                        session.add(ns)
-                        session.commit()
-                    except Exception as e:
-                        logger.warn(e)
-                        session.rollback()
-    else:
-        error = "You are not logged in"
-    return {"error": error}
--- a/resolvers/proposals.py
+++ b/resolvers/proposals.py
@@ -1,49 +0,0 @@
-from sqlalchemy import and_
-
-from orm.rating import is_negative, is_positive
-from orm.reaction import Reaction, ReactionKind
-from orm.shout import Shout
-from services.db import local_session
-from utils.diff import apply_diff, get_diff
-
-
-def handle_proposing(kind: ReactionKind, reply_to: int, shout_id: int):
-    with local_session() as session:
-        if is_positive(kind):
-            replied_reaction = (
-                session.query(Reaction).filter(Reaction.id == reply_to, Reaction.shout == shout_id).first()
-            )
-
-            if replied_reaction and replied_reaction.kind is ReactionKind.PROPOSE.value and replied_reaction.quote:
-                # patch all the proposals' quotes
-                proposals = (
-                    session.query(Reaction)
-                    .filter(
-                        and_(
-                            Reaction.shout == shout_id,
-                            Reaction.kind == ReactionKind.PROPOSE.value,
-                        )
-                    )
-                    .all()
-                )
-
-                # patch shout's body
-                shout = session.query(Shout).filter(Shout.id == shout_id).first()
-                body = replied_reaction.quote
-                Shout.update(shout, {body})
-                session.add(shout)
-                session.commit()
-
-                # реакция содержит цитату -> обновляются все предложения
-                # (proposals) для соответствующего Shout.
-                for proposal in proposals:
-                    if proposal.quote:
-                        proposal_diff = get_diff(shout.body, proposal.quote)
-                        proposal_dict = proposal.dict()
-                        proposal_dict["quote"] = apply_diff(replied_reaction.quote, proposal_diff)
-                        Reaction.update(proposal, proposal_dict)
-                        session.add(proposal)
-
-        if is_negative(kind):
-            # TODO: rejection logic
-            pass
--- a/Show More
+++ b/Show More
Author	SHA1	Message	Date
Stepan Vladovskiy	fb6e03c1a2	no force, deployd from gitea Some checks failed Deploy to discoursio-api / deploy (push) Failing after 46s Details	2025-04-07 11:14:58 -03:00
Stepan Vladovskiy	46c3345f45	debug: with force Some checks failed Deploy to discoursio-api / deploy (push) Failing after 1m42s Details	2025-04-07 11:11:01 -03:00
Stepan Vladovskiy	1156a32a88	feat: move map from nginx sigil to nginx global config All checks were successful Deploy to discoursio-api / deploy (push) Successful in 1m54s Details	2025-01-30 12:40:24 -03:00
Untone	d848af524f	runtime-fix	2024-12-21 23:31:19 +03:00
Untone	c9f88c36cd	trigdeploy	2024-12-21 23:00:30 +03:00
Untone	0ad44a944e	Revert ".." Some checks failed Deploy to discoursio-api / deploy (push) Failing after 1m22s Details This reverts commit `fbd0e03a33`.	2024-08-06 21:01:03 +03:00
Untone	fbd0e03a33	..	2024-02-23 17:08:08 +03:00
Stepan Vladovskii	076828f003	feat: no force any more for CI deploy from Gitea All checks were successful Deploy to discoursio-api / deploy (push) Successful in 30s Details	2024-01-28 18:38:34 -03:00
Stepan Vladovskii	4f6c459532	feat: sigil with other architecture All checks were successful Deploy to discoursio-api / deploy (push) Successful in 1m41s Details	2024-01-24 22:53:16 -03:00
Stepan Vladovskii	11524c17ea	feat: yess, it was deploy on staging All checks were successful Deploy to discoursio-api / deploy (push) Successful in 1m40s Details	2024-01-24 21:06:11 -03:00
Untone	168f845772	Merge branch 'main' of https://dev.discours.io/discours.io/core Some checks failed Deploy to discoursio-api / deploy (push) Failing after 1m31s Details	2024-01-25 02:59:43 +03:00
Untone	657146cdca	trig-redeploy	2024-01-25 02:56:30 +03:00
Stepan Vladovskii	86111bc9f5	debug: simplify main.yml for actions Some checks failed Deploy to discoursio-api / deploy (push) Failing after 58s Details	2024-01-24 20:45:34 -03:00
Stepan Vladovskii	a8018a0b2f	debug: simplify main.yml for actions Some checks failed Deploy to discoursio-api / deploy (push) Failing after 3s Details	2024-01-24 20:39:58 -03:00
Stepan Vladovskii	9d8bd629ab	feat: add CI to main for deploy on discoursio-api	2024-01-24 19:09:54 -03:00
Untone	1eddf9cc0b	topic-resolvers-fix	2024-01-24 18:21:34 +03:00
Untone	6415f86286	redis-log-fix	2024-01-24 15:32:53 +03:00
Untone	5d1c4f0084	launch-fix	2024-01-24 11:36:15 +03:00
Untone	1dce947db6	db-link-fix	2024-01-24 11:19:42 +03:00
Untone	4d9551a93c	redis-fix	2024-01-24 11:06:46 +03:00
Untone	e6471280d5	dockerfile-fix2	2024-01-24 10:59:34 +03:00
Untone	3e062b4346	untransform-dockerfile	2024-01-24 10:55:14 +03:00
Untone	5b1a93c781	Merge branch 'main' of github.com:Discours/discours-backend	2024-01-24 10:47:07 +03:00
Untone	c30001547a	Merge branch 'main' of github.com:Discours/discours-backend	2023-12-24 17:26:17 +03:00
Stepan Vladovskiy	025019b544	feat: add ACME location Some checks failed Deploy / push_to_target_repository (push) Failing after 4m25s Details	2023-11-28 14:20:19 -03:00
Untone	a862a11c91	Revert "some-fixes" Some checks failed Deploy / push_to_target_repository (push) Failing after 4m33s Details This reverts commit `f3d86daea7`.	2023-11-24 12:54:01 +03:00
Untone	f3d86daea7	some-fixes	2023-11-24 05:19:25 +03:00
Untone	296716397e	Merge branch 'main' of github.com:Discours/discours-backend	2023-11-24 00:38:20 +03:00
Tony Rewin	22c42839c1	new-sigil	2023-10-09 18:12:47 +03:00
Tony Rewin	4fd90e305f	Merge branch 'main' of https://github.com/Discours/discours-backend	2023-10-09 10:32:33 +03:00
				`@@ -1 +0,0 @@`
				`# Add directories or file patterns to ignore during indexing (e.g. foo/ or *.csv)`
				`@@ -0,0 +1 @@`
				`python -m gql_schema_codegen -p ./schema.graphql -t ./schema_types.py`