bot-backend
5da4271235
All checks were successful
CI / frontend-tests (push) Has been skipped
CI / changes (pull_request) Successful in 6s
CI / changes (push) Successful in 6s
CI / frontend-tests (pull_request) Has been skipped
CI / backend-tests (push) Successful in 6m22s
CI / backend-tests (pull_request) Successful in 6m19s
- Dockerfile: добавлены tesseract-ocr tesseract-ocr-rus в runner apt-слой - pyproject.toml: pymupdf>=1.24 (рендер PDF без poppler) + pytesseract>=0.3.13 - izyatie_client.py: list_izyatie_documents() + fetch_pdf() с екатеринбург.рф - izyatie_ocr.py: ocr_pdf_text() (PyMuPDF→PIL→Tesseract rus) + extract_izyatie_records() с нормализацией OCR-шума (пробелы в кад-номерах, кирилл. О→0, б→6) - izyatie_ocr_ingest.py: Celery task → land_reservation UPSERT (SAVEPOINT per-row) - beat_schedule.py: izyatie-ocr-ingest-weekly (пятница 07:00 МСК) - 33 теста, все зелёные; без реального Tesseract/сети в CI
110 lines
3.2 KiB
Docker
110 lines
3.2 KiB
Docker
# syntax=docker/dockerfile:1.7
|
||
#
|
||
# Multi-target build:
|
||
# target=runner — lean image для backend (uvicorn) и beat. ~1.4 ГБ.
|
||
# target=runner-with-chromium — image для worker (Playwright fetch). ~1.6 ГБ.
|
||
#
|
||
# Local: docker compose сам выбирает target по сервису (см. docker-compose.yml).
|
||
# Prod: CI пушит 2 GHCR-образа: gendesign-backend (lean) и gendesign-worker (chromium).
|
||
|
||
# ---- builder ----
|
||
FROM python:3.12-slim AS builder
|
||
|
||
ENV PYTHONUNBUFFERED=1 \
|
||
PYTHONDONTWRITEBYTECODE=1 \
|
||
UV_COMPILE_BYTECODE=1 \
|
||
UV_LINK_MODE=copy
|
||
|
||
RUN --mount=type=cache,target=/var/cache/apt,sharing=locked \
|
||
--mount=type=cache,target=/var/lib/apt,sharing=locked \
|
||
apt-get update && apt-get install -y --no-install-recommends \
|
||
build-essential \
|
||
libpq-dev \
|
||
libgdal-dev \
|
||
libproj-dev \
|
||
libgeos-dev
|
||
|
||
RUN pip install --no-cache-dir uv
|
||
|
||
WORKDIR /app
|
||
|
||
COPY pyproject.toml uv.lock* ./
|
||
# uv-кеш переживает между билдами: при неизменном lock'е sync ~5 сек вместо 1-2 мин.
|
||
RUN --mount=type=cache,target=/root/.cache/uv \
|
||
if [ -f uv.lock ]; then uv sync --frozen --no-dev; else uv sync --no-dev; fi
|
||
|
||
COPY app ./app
|
||
COPY alembic.ini ./
|
||
COPY alembic ./alembic
|
||
|
||
|
||
# ---- runner (lean — без Chromium) ----
|
||
FROM python:3.12-slim AS runner
|
||
|
||
ENV PYTHONUNBUFFERED=1 \
|
||
PYTHONDONTWRITEBYTECODE=1 \
|
||
PATH="/app/.venv/bin:$PATH"
|
||
|
||
RUN --mount=type=cache,target=/var/cache/apt,sharing=locked \
|
||
--mount=type=cache,target=/var/lib/apt,sharing=locked \
|
||
apt-get update && apt-get install -y --no-install-recommends \
|
||
libpq5 \
|
||
libgdal36 \
|
||
libproj25 \
|
||
libgeos-c1t64 \
|
||
libcairo2 \
|
||
libpango-1.0-0 \
|
||
libpangoft2-1.0-0 \
|
||
fonts-dejavu-core \
|
||
curl \
|
||
tesseract-ocr \
|
||
tesseract-ocr-rus \
|
||
&& useradd --create-home --uid 1000 app
|
||
|
||
WORKDIR /app
|
||
|
||
COPY --from=builder --chown=app:app /app/.venv /app/.venv
|
||
COPY --from=builder --chown=app:app /app/app /app/app
|
||
COPY --from=builder --chown=app:app /app/alembic.ini /app/alembic.ini
|
||
COPY --from=builder --chown=app:app /app/alembic /app/alembic
|
||
|
||
USER app
|
||
|
||
EXPOSE 8000
|
||
|
||
CMD ["uvicorn", "app.main:app", "--host", "0.0.0.0", "--port", "8000"]
|
||
|
||
|
||
# ---- runner-with-chromium (для worker) ----
|
||
# Наследуется от lean runner, добавляет Chromium-libs + Playwright-binary (~200 MB).
|
||
FROM runner AS runner-with-chromium
|
||
|
||
USER root
|
||
|
||
RUN --mount=type=cache,target=/var/cache/apt,sharing=locked \
|
||
--mount=type=cache,target=/var/lib/apt,sharing=locked \
|
||
apt-get update && apt-get install -y --no-install-recommends \
|
||
# Chromium runtime libs (раньше заходили через `playwright install --with-deps`)
|
||
libnss3 \
|
||
libnspr4 \
|
||
libdbus-1-3 \
|
||
libatk1.0-0 \
|
||
libatk-bridge2.0-0 \
|
||
libcups2 \
|
||
libdrm2 \
|
||
libxkbcommon0 \
|
||
libxcomposite1 \
|
||
libxdamage1 \
|
||
libxfixes3 \
|
||
libxrandr2 \
|
||
libgbm1 \
|
||
libasound2t64 \
|
||
libatspi2.0-0
|
||
|
||
# Playwright Chromium binary под root → shared dir, чтобы и app-user мог использовать.
|
||
ENV PLAYWRIGHT_BROWSERS_PATH=/opt/playwright
|
||
RUN mkdir -p $PLAYWRIGHT_BROWSERS_PATH \
|
||
&& /app/.venv/bin/playwright install chromium \
|
||
&& chmod -R a+rX $PLAYWRIGHT_BROWSERS_PATH
|
||
|
||
USER app
|