Настройка гитигнор

.gitignore

@@ -1 +1,4 @@
 src/rag_agent/.env
+.env
+docker/ssh
+docker/postgres_test_data

Dockerfile

@@ -14,8 +14,9 @@ COPY README.md ./
 
 RUN pip install --no-cache-dir -e .
 
-# Default: run CLI (override in compose or when running)
+# Default: run webhook server (override in compose or when running)
 ENV RAG_DB_DSN=""
 ENV RAG_REPO_PATH="/data"
+EXPOSE 8000
 ENTRYPOINT ["rag-agent"]
-CMD ["ask", "--help"]
+CMD ["serve", "--host", "0.0.0.0", "--port", "8000"]

README.md

@@ -1,8 +1,10 @@
+
+
+
 # RAG Agent (Postgres)
 
 Custom RAG agent that indexes text files from a git repository into Postgres
-and answers queries using retrieval + LLM generation. Commits are tied to
-**stories**; indexing and retrieval can be scoped by story.
+and answers queries using retrieval + LLM generation. **Changes are always in the context of a Story**: the unit of work is the story, not individual commits. The agent indexes **all changes from all commits** in the story range (base_ref..head_ref); per-commit indexing is not used.
 
 ## Quick start
 
@@ -15,15 +17,61 @@ and answers queries using retrieval + LLM generation. Commits are tied to
 2. Configure environment variables:
    - `RAG_REPO_PATH` — path to git repo with text files
    - `RAG_DB_DSN` — Postgres DSN (e.g. `postgresql://rag:rag_secret@localhost:5432/rag`)
-   - `RAG_EMBEDDINGS_DIM` — embedding vector dimension (e.g. `1536`)
+   - `RAG_EMBEDDINGS_DIM` — embedding vector dimension: **1024** for GigaChat Embeddings (default), 1536 for OpenAI
 3. Create DB schema (only if not using Docker, or if init was disabled):
    - `python scripts/create_db.py` (or `psql "$RAG_DB_DSN" -f scripts/schema.sql`)
-4. Index files for a story (e.g. branch name as story slug):
-   - `rag-agent index --story my-branch --changed --base-ref HEAD~1 --head-ref HEAD`
+4. Index files for a story (e.g. branch name as story slug). Use the **full story range** so all commits in the story are included:
+   - `rag-agent index --story my-branch --changed --base-ref main --head-ref HEAD`
+   - Or `--base-ref auto` to use merge-base(default-branch, head-ref) as the start of the story.
 5. Ask a question (optionally scoped to a story):
    - `rag-agent ask "What is covered?"`
    - `rag-agent ask "What is covered?" --story my-branch`
 
+## Webhook: index on push to remote
+
+When the app runs as a service in Docker, it can start a **webhook server** so that each push to the remote repository triggers a pull and incremental indexing.
+
+1. Start the stack with the webhook server (default in Docker):
+   - `docker compose up -d` — app runs `rag-agent serve` and listens on port 8000.
+   - Repo is mounted at `RAG_REPO_PATH` (e.g. `/data`) **writable**, so the container can run `git fetch` + `git merge --ff-only` to pull changes.
+2. Clone the knowledge-base repo into the mounted directory (once), e.g. on the host: `git clone <url> ./data` so that `./data` is the worktree (or set `RAG_REPO_PATH` to that path and mount it).
+3. In GitHub (or GitLab) add a **Webhook**:
+   - URL: `http://<your-server>:8000/webhook` (use HTTPS in production and put a reverse proxy in front).
+   - Content type: `application/json`.
+   - Secret: set a shared secret and export `WEBHOOK_SECRET` in the app environment (Docker: in `docker-compose.yml` or `.env`). If `WEBHOOK_SECRET` is empty, signature is not checked.
+4. On each push to a branch, the server receives the webhook, pulls that branch into the worktree, and runs `rag-agent index --story <branch> --changed --base-ref <old_head> --head-ref <new_head>` so only changed files are re-indexed.
+
+Health check: `GET http://<host>:8000/health` → `ok`. Port is configurable via `WEBHOOK_PORT` (default 8000) in docker-compose.
+
+### Webhook diagnostics (202 Accepted but no new rows in DB)
+
+1. **Logs** — After a push, check app logs. Each webhook logs `pull_and_index started branch=… repo_path=…`; then one of:
+   - `not a git repo or missing` — `/data` in the container is not a git clone; clone the repo into the mounted dir.
+   - `git fetch failed` — SSH/network (see `docker/ssh/README.md`) or wrong remote.
+   - `git checkout … failed` — branch missing in the clone.
+   - `git merge --ff-only failed` — non–fast-forward (e.g. force-push); index is skipped. Use a normal push or re-clone.
+   - `no new commits for branch=…` — merge was a no-op (already up to date); nothing to index.
+   - `running index story=…` then `index completed` — index ran; check tables for that story.
+   - `index failed` — stderr shows the `rag-agent index` error (e.g. DB, embeddings, repo path).
+
+   ```bash
+   docker compose logs -f app
+   # or: docker logs -f rag-agent
+   ```
+   Trigger a push and watch for the lines above.
+
+2. **Story and tables** — Rows are per **story** (branch name). Query by story, e.g. `SELECT * FROM stories;` then `SELECT * FROM chunks WHERE story_id = (SELECT id FROM stories WHERE slug = 'main');`.
+
+3. **Manual index** — Run index inside the container to confirm DB and repo work:
+   ```bash
+   docker compose exec app rag-agent index --story main --changed --base-ref main --head-ref HEAD
+   ```
+   If this inserts rows, the issue is in the webhook path (fetch/merge/refs).
+
+4. **Allowed extensions** — Only `.md`, `.txt`, `.rst` (or `RAG_ALLOWED_EXTENSIONS`) are indexed; other files are skipped.
+
+5. **"expected 1536 dimensions, not 1024"** — GigaChat Embeddings returns 1024-dim vectors; the default is now 1024. If the DB was created earlier with vector(1536), drop and recreate the tables so the app can create them with 1024: `psql "$RAG_DB_DSN" -c "DROP TABLE IF EXISTS chunks; DROP TABLE IF EXISTS documents;"` then restart the app (ensure_schema will recreate the tables).
+
 ## Git hook (index on commit)
 
 Install the post-commit hook so changed files are indexed after each commit:
@@ -34,11 +82,40 @@ cp scripts/post-commit .git/hooks/post-commit && chmod +x .git/hooks/post-commit
 
 Story for the commit is taken from (in order): env `RAG_STORY`, file `.rag-story` in repo root (one line = slug), or current branch name.
 
+## Git hook (server-side)
+
+Use `scripts/post-receive` in the **bare repo** on the server so that pushes trigger indexing.
+
+1. On the server, create a **non-bare clone** (worktree) that the hook will update and use for indexing, e.g. `git clone /path/to/repo.git /var/rag-worktree/repo`.
+2. In the bare repo, install the hook: `cp /path/to/RagAgent/scripts/post-receive /path/to/repo.git/hooks/post-receive && chmod +x .../post-receive`.
+3. Set env for the hook (e.g. in the hook or via systemd/sshd): `RAG_REPO_PATH=/var/rag-worktree/repo`, `RAG_DB_DSN=...`, `RAG_EMBEDDINGS_DIM=...`. Optionally `RAG_AGENT_VENV` (path to venv with `rag-agent`) or `RAG_AGENT_SRC` + `RAG_AGENT_PYTHON` for `python -m rag_agent.cli`.
+4. On each push the hook updates the worktree to the new commit, then runs `rag-agent index --changed --base-ref main --head-ref newrev --story <branch>` so the story contains **all commits** on the branch (from main to newrev).
+
+Story is taken from the ref name (e.g. `refs/heads/main` → `main`).
+
 ## DB structure
 
-- **stories** — story slug (e.g. branch name); documents and chunks are tied to a story.
+- **stories** — story slug (e.g. branch name); documents and chunks are tied to a story. Optional: `indexed_base_ref`, `indexed_head_ref`, `indexed_at` record the git range that was indexed (all commits in that range belong to the story).
 - **documents** — path + version per story; unique `(story_id, path)`.
-- **chunks** — text chunks with embeddings (pgvector); updated when documents are re-indexed.
+- **chunks** — text chunks with embeddings (pgvector), plus:
+  - `start_line`, `end_line` — position in the source file (for requirements/use-case files).
+  - `change_type` — `added` | `modified` | `unchanged` (relative to base ref when indexing with `--changed`).
+  - `previous_content` — for `modified` chunks, the content before the change (for test-case generation).
+
+Indexing is **always per-story**: `base_ref..head_ref` defines the set of commits that belong to the story. Use `--base-ref main` (or `auto`) and `--head-ref HEAD` so the story contains all commits on the branch, not a single commit. When you run `index --changed`, the base ref is compared to head; each chunk is marked as added, modified, or unchanged.
+
+### What changed in a story (for test cases)
+
+To get only the chunks that were added or modified in a story (e.g. to generate test cases for the changed part):
+
+```python
+from rag_agent.index import fetch_changed_chunks
+
+changed = fetch_changed_chunks(conn, story_id)
+for r in changed:
+    # r.path, r.content, r.change_type, r.start_line, r.end_line, r.previous_content
+    ...
+```
 
 Scripts: `scripts/create_db.py` (Python, uses `ensure_schema` and `RAG_*` env), `scripts/schema.sql` (raw SQL).
 
@@ -46,7 +123,21 @@ Scripts: `scripts/create_db.py` (Python, uses `ensure_schema` and `RAG_*` env),
 
 If `GIGACHAT_CREDENTIALS` is set (e.g. in `.env` for local runs), embeddings use GigaChat API; otherwise the stub client is used. Optional env: `GIGACHAT_EMBEDDINGS_MODEL` (default `Embeddings`), `GIGACHAT_VERIFY_SSL` (`true`/`false`). Ensure `RAG_EMBEDDINGS_DIM` matches the model output (see GigaChat docs).
 
+## Agent (GigaChat)
+
+Ответы на вопросы формирует агент на базе GigaChat: поиск по базе знаний (RAG) + генерация текста. Если задана переменная `GIGACHAT_CREDENTIALS`, используется `GigaChatLLMClient` в `src/rag_agent/agent/pipeline.py`; иначе — заглушка. Модель чата задаётся через `RAG_LLM_MODEL` (по умолчанию `GigaChat`).
+
+## Telegram-бот
+
+Общение с пользователем через бота в Telegram: бот отвечает на текстовые сообщения, используя знания из базы (RAG + GigaChat).
+
+1. Создайте бота через [@BotFather](https://t.me/BotFather) и получите токен.
+2. Добавьте в `.env`: `TELEGRAM_BOT_TOKEN=<токен>`.
+3. Запуск: `rag-agent bot` (или `python -m rag_agent.telegram_bot`).
+4. Через Docker: `docker compose up -d` поднимает БД, вебхук-сервер и бота в отдельных контейнерах; в `.env` должен быть задан `TELEGRAM_BOT_TOKEN`.
+
+Требуются: `RAG_DB_DSN`, `RAG_REPO_PATH`, `GIGACHAT_CREDENTIALS`, `TELEGRAM_BOT_TOKEN`. Расширенное логирование (входящие сообщения, число эмбеддингов, число чанков из БД, ответ LLM): `RAG_BOT_VERBOSE_LOGGING=true|false` (по умолчанию `true` для отладки).
+
 ## Notes
 
-- LLM client is still a stub; replace it in `src/rag_agent/agent/pipeline.py` for real answers.
 - This project requires Postgres with the `pgvector` extension.

docker-compose.yml

@@ -14,7 +14,8 @@ services:
     ports:
       - "${POSTGRES_PORT:-5432}:5432"
     volumes:
-      - rag_pgdata:/var/lib/postgresql/data
+      # PG 18+: mount at /var/lib/postgresql (data goes in versioned subdir). For pg16 use /var/lib/postgresql/data.
+      - rag_pgdata:/var/lib/postgresql
       # Init scripts run once on first start (create extension, tables). Optional: comment out to skip.
       - ./docker/postgres-init:/docker-entrypoint-initdb.d:ro
     healthcheck:
@@ -31,20 +32,54 @@ services:
       dockerfile: Dockerfile
     image: rag-agent:latest
     container_name: rag-agent
-    restart: "no"
+    restart: unless-stopped
+    depends_on:
+      postgres:
+        condition: service_healthy
+    ports:
+      - "${WEBHOOK_PORT:-8000}:8000"
+    environment:
+      RAG_DB_DSN: "postgresql://${POSTGRES_USER:-rag}:${POSTGRES_PASSWORD:-rag_secret}@postgres:5432/${POSTGRES_DB:-rag}"
+      # In container repo is always at /data (mounted below). Use RAG_REPO_HOST in .env for host path.
+      RAG_REPO_PATH: "/data"
+      # Accept host key on first connect; git fetch uses SSH from /root/.ssh (mounted below).
+      GIT_SSH_COMMAND: "ssh -o StrictHostKeyChecking=accept-new"
+      RAG_EMBEDDINGS_DIM: ${RAG_EMBEDDINGS_DIM:-1024}
+      GIGACHAT_CREDENTIALS: ${GIGACHAT_CREDENTIALS:-}
+      GIGACHAT_EMBEDDINGS_MODEL: ${GIGACHAT_EMBEDDINGS_MODEL:-Embeddings}
+      WEBHOOK_SECRET: ${WEBHOOK_SECRET:-}
+    volumes:
+      # Host path: set RAG_REPO_HOST in .env (e.g. /Users/you/repo). Falls back to RAG_REPO_PATH then ./data.
+      - ${RAG_REPO_HOST:-${RAG_REPO_PATH:-./data}}:/data
+      # SSH for git fetch (webhook): put deploy key and known_hosts in RAG_SSH_DIR. See docker/ssh/README.md.
+      - ${RAG_SSH_DIR:-./docker/ssh}:/root/.ssh:ro
+    entrypoint: ["rag-agent"]
+    command: ["serve", "--host", "0.0.0.0", "--port", "8000"]
+    networks:
+      - rag_net
+
+  bot:
+    build:
+      context: .
+      dockerfile: Dockerfile
+    image: rag-agent:latest
+    container_name: rag-bot
+    restart: unless-stopped
     depends_on:
       postgres:
         condition: service_healthy
     environment:
       RAG_DB_DSN: "postgresql://${POSTGRES_USER:-rag}:${POSTGRES_PASSWORD:-rag_secret}@postgres:5432/${POSTGRES_DB:-rag}"
-      RAG_REPO_PATH: ${RAG_REPO_PATH:-/data}
-      RAG_EMBEDDINGS_DIM: ${RAG_EMBEDDINGS_DIM:-1536}
+      RAG_REPO_PATH: "/data"
+      RAG_EMBEDDINGS_DIM: ${RAG_EMBEDDINGS_DIM:-1024}
       GIGACHAT_CREDENTIALS: ${GIGACHAT_CREDENTIALS:-}
       GIGACHAT_EMBEDDINGS_MODEL: ${GIGACHAT_EMBEDDINGS_MODEL:-Embeddings}
+      TELEGRAM_BOT_TOKEN: ${TELEGRAM_BOT_TOKEN:-}
+      RAG_BOT_VERBOSE_LOGGING: ${RAG_BOT_VERBOSE_LOGGING:-true}
     volumes:
-      - ${RAG_REPO_PATH:-./data}:/data:ro
+      - ${RAG_REPO_HOST:-${RAG_REPO_PATH:-./data}}:/data
     entrypoint: ["rag-agent"]
-    command: ["ask", "--help"]
+    command: ["bot"]
     networks:
       - rag_net
 

docker/postgres-init/01-schema.sql

@@ -1,12 +1,15 @@
 -- RAG vector DB schema (runs automatically on first Postgres init).
--- If RAG_EMBEDDINGS_DIM is not 1536, change vector(1536) below.
+-- GigaChat Embeddings = 1024; for OpenAI use vector(1536).
 
 CREATE EXTENSION IF NOT EXISTS vector;
 
 CREATE TABLE IF NOT EXISTS stories (
     id SERIAL PRIMARY KEY,
     slug TEXT UNIQUE NOT NULL,
-    created_at TIMESTAMPTZ NOT NULL DEFAULT (NOW() AT TIME ZONE 'utc')
+    created_at TIMESTAMPTZ NOT NULL DEFAULT (NOW() AT TIME ZONE 'utc'),
+    indexed_base_ref TEXT,
+    indexed_head_ref TEXT,
+    indexed_at TIMESTAMPTZ
 );
 
 CREATE TABLE IF NOT EXISTS documents (
@@ -24,9 +27,15 @@ CREATE TABLE IF NOT EXISTS chunks (
     chunk_index INTEGER NOT NULL,
     hash TEXT NOT NULL,
     content TEXT NOT NULL,
-    embedding vector(1536) NOT NULL
+    embedding vector(1024) NOT NULL,
+    start_line INTEGER,
+    end_line INTEGER,
+    change_type TEXT NOT NULL DEFAULT 'added'
+        CHECK (change_type IN ('added', 'modified', 'unchanged')),
+    previous_content TEXT
 );
 
 CREATE INDEX IF NOT EXISTS idx_documents_story_id ON documents(story_id);
 CREATE INDEX IF NOT EXISTS idx_chunks_document_id ON chunks(document_id);
 CREATE INDEX IF NOT EXISTS idx_chunks_embedding ON chunks USING ivfflat (embedding vector_cosine_ops);
+CREATE INDEX IF NOT EXISTS idx_chunks_change_type ON chunks(change_type);

docker/postgres_test_data/18/docker/PG_VERSION

@@ -0,0 +1 @@
+18