feat: extract OCR worker into separate service

- Remove Python/OCR steps from backend.Dockerfile (builder + runtime)
- Remove Python/OCR steps from backend-dev.Dockerfile
- Add OCR_BASE_URL env var to both Dockerfiles (http://ocr-worker:8100)
- Add ocr-worker service to docker-compose.yml with healthcheck
- Add ocr-worker service to docker-compose.dev.yml with healthcheck + port
- Update backend depends_on in both compose files to include ocr-worker
- Replace OCR_PROJECT_ROOT with OCR_BASE_URL in compose env vars
- Remove pip install line from start-dev-container.sh
This commit is contained in:
yml
2026-05-19 23:06:15 +08:00
parent f87238b6c0
commit b7de22d665
5 changed files with 42 additions and 30 deletions
@@ -85,7 +85,6 @@ start_browser_debug_services() {
} }
npm install --no-fund --no-audit npm install --no-fund --no-audit
python3 -m pip install --no-cache-dir --break-system-packages -e /app/subservices/ocr-worker
if ! is_true "${TENCENT_BROWSER_HEADLESS:-true}"; then if ! is_true "${TENCENT_BROWSER_HEADLESS:-true}"; then
start_browser_debug_services start_browser_debug_services
+3 -8
View File
@@ -1,15 +1,14 @@
# Dev Dockerfile — single-stage, root, VNC/X11 + Playwright + OCR # Dev Dockerfile — single-stage, root, VNC/X11 + Playwright
# For production, see backend.Dockerfile (multi-stage build) # For production, see backend.Dockerfile (multi-stage build)
FROM node:22-bookworm FROM node:22-bookworm
ARG DEBIAN_MIRROR=mirrors.cloud.tencent.com ARG DEBIAN_MIRROR=mirrors.cloud.tencent.com
ARG NPM_REGISTRY=https://registry.npmmirror.com ARG NPM_REGISTRY=https://registry.npmmirror.com
ARG UV_INDEX_URL=https://mirrors.cloud.tencent.com/pypi/simple
ENV DEBIAN_FRONTEND=noninteractive ENV DEBIAN_FRONTEND=noninteractive
ENV TZ=Asia/Shanghai ENV TZ=Asia/Shanghai
ENV PLAYWRIGHT_BROWSERS_PATH=/ms-playwright ENV PLAYWRIGHT_BROWSERS_PATH=/ms-playwright
ENV PIP_INDEX_URL=${UV_INDEX_URL} ENV OCR_BASE_URL=http://ocr-worker:8100
ENV NPM_CONFIG_REGISTRY=${NPM_REGISTRY} ENV NPM_CONFIG_REGISTRY=${NPM_REGISTRY}
WORKDIR /app WORKDIR /app
@@ -20,7 +19,7 @@ RUN --mount=type=cache,target=/var/cache/apt,sharing=locked \
&& sed -i "s|http://security.debian.org/debian-security|http://${DEBIAN_MIRROR}/debian-security|g" /etc/apt/sources.list.d/debian.sources \ && sed -i "s|http://security.debian.org/debian-security|http://${DEBIAN_MIRROR}/debian-security|g" /etc/apt/sources.list.d/debian.sources \
&& apt-get update \ && apt-get update \
&& apt-get install -y --no-install-recommends \ && apt-get install -y --no-install-recommends \
curl ca-certificates python3 python3-pip tzdata \ curl ca-certificates tzdata \
xauth x11vnc novnc websockify xvfb \ xauth x11vnc novnc websockify xvfb \
&& ln -snf /usr/share/zoneinfo/${TZ} /etc/localtime \ && ln -snf /usr/share/zoneinfo/${TZ} /etc/localtime \
&& echo ${TZ} > /etc/timezone && echo ${TZ} > /etc/timezone
@@ -34,10 +33,6 @@ RUN --mount=type=cache,target=/ms-playwright-cache,sharing=locked \
npx playwright install --with-deps chromium \ npx playwright install --with-deps chromium \
&& cp -r /ms-playwright-cache /ms-playwright && cp -r /ms-playwright-cache /ms-playwright
COPY apps/backend/subservices/ocr-worker/ ./subservices/ocr-worker/
RUN --mount=type=cache,target=/root/.cache/pip,sharing=locked \
python3 -m pip install --break-system-packages ./subservices/ocr-worker
COPY apps/backend/ ./ COPY apps/backend/ ./
RUN mkdir -p /app/data /app/data/browser-sessions /app/data/redeem-screenshots RUN mkdir -p /app/data /app/data/browser-sessions /app/data/redeem-screenshots
+3 -19
View File
@@ -11,12 +11,10 @@ FROM node:22-bookworm AS builder
ARG DEBIAN_MIRROR=mirrors.cloud.tencent.com ARG DEBIAN_MIRROR=mirrors.cloud.tencent.com
ARG NPM_REGISTRY=https://registry.npmmirror.com ARG NPM_REGISTRY=https://registry.npmmirror.com
ARG UV_INDEX_URL=https://mirrors.cloud.tencent.com/pypi/simple
# NOTE: Use HTTP (not HTTPS) for apt mirrors — OrbStack BuildKit cannot verify # NOTE: Use HTTP (not HTTPS) for apt mirrors — OrbStack BuildKit cannot verify
# TLS certs during build. Apt packages are GPG-signed so HTTP is safe. # TLS certs during build. Apt packages are GPG-signed so HTTP is safe.
ENV DEBIAN_FRONTEND=noninteractive ENV DEBIAN_FRONTEND=noninteractive
ENV PIP_INDEX_URL=${UV_INDEX_URL}
ENV NPM_CONFIG_REGISTRY=${NPM_REGISTRY} ENV NPM_CONFIG_REGISTRY=${NPM_REGISTRY}
WORKDIR /app WORKDIR /app
@@ -28,7 +26,7 @@ RUN --mount=type=cache,target=/var/cache/apt,sharing=locked \
&& sed -i "s|http://security.debian.org/debian-security|http://${DEBIAN_MIRROR}/debian-security|g" /etc/apt/sources.list.d/debian.sources \ && sed -i "s|http://security.debian.org/debian-security|http://${DEBIAN_MIRROR}/debian-security|g" /etc/apt/sources.list.d/debian.sources \
&& apt-get update \ && apt-get update \
&& apt-get install -y --no-install-recommends \ && apt-get install -y --no-install-recommends \
python3 python3-pip python3-venv curl ca-certificates tzdata curl ca-certificates tzdata
# --- Node dependencies (full, including devDeps for postinstall scripts) --- # --- Node dependencies (full, including devDeps for postinstall scripts) ---
COPY apps/backend/package.json apps/backend/package-lock.json ./ COPY apps/backend/package.json apps/backend/package-lock.json ./
@@ -46,34 +44,24 @@ RUN --mount=type=cache,target=/ms-playwright-cache,sharing=locked \
npx playwright install --with-deps chromium \ npx playwright install --with-deps chromium \
&& cp -r /ms-playwright-cache /ms-playwright && cp -r /ms-playwright-cache /ms-playwright
# --- Python venv for OCR worker ---
# Use a virtual environment instead of --break-system-packages
COPY apps/backend/subservices/ocr-worker/ ./subservices/ocr-worker/
RUN python3 -m venv /app/subservices/ocr-worker/.venv \
&& /app/subservices/ocr-worker/.venv/bin/pip install --no-cache-dir -U pip \
&& /app/subservices/ocr-worker/.venv/bin/pip install --no-cache-dir ./subservices/ocr-worker
# --- Copy application source --- # --- Copy application source ---
COPY apps/backend/ ./ COPY apps/backend/ ./
# --------------------------- Runtime Stage ----------------------------------- # --------------------------- Runtime Stage -----------------------------------
FROM node:22-bookworm-slim AS runtime FROM node:22-bookworm-slim AS runtime
ARG DEBIAN_MIRROR=mirrors.cloud.tencent.com
# HTTP mirrors — see builder note about OrbStack TLS # HTTP mirrors — see builder note about OrbStack TLS
ENV DEBIAN_FRONTEND=noninteractive ENV DEBIAN_FRONTEND=noninteractive
ENV TZ=Asia/Shanghai ENV TZ=Asia/Shanghai
ENV NODE_ENV=production ENV NODE_ENV=production
ENV PLAYWRIGHT_BROWSERS_PATH=/ms-playwright ENV PLAYWRIGHT_BROWSERS_PATH=/ms-playwright
ENV OCR_PROJECT_ROOT=/app/subservices/ocr-worker ENV OCR_BASE_URL=http://ocr-worker:8100
WORKDIR /app WORKDIR /app
# --- Minimal runtime apt packages --- # --- Minimal runtime apt packages ---
# - curl: health check (docker-compose uses `curl -fsS http://127.0.0.1:3000/health/ready`) # - curl: health check (docker-compose uses `curl -fsS http://127.0.0.1:3000/health/ready`)
# - ca-certificates: TLS for outbound HTTPS calls # - ca-certificates: TLS for outbound HTTPS calls
# - python3: needed by venv (links to system libpython)
# - tzdata: timezone data for TZ=Asia/Shanghai # - tzdata: timezone data for TZ=Asia/Shanghai
# - Chromium runtime libs: minimal set for headless Chromium on slim # - Chromium runtime libs: minimal set for headless Chromium on slim
# #
@@ -86,7 +74,7 @@ RUN --mount=type=cache,target=/var/cache/apt,sharing=locked \
--mount=type=cache,target=/var/lib/apt,sharing=locked \ --mount=type=cache,target=/var/lib/apt,sharing=locked \
apt-get update \ apt-get update \
&& apt-get install -y --no-install-recommends \ && apt-get install -y --no-install-recommends \
curl ca-certificates python3 tzdata \ curl ca-certificates tzdata \
libnss3 libnspr4 libdbus-1-3 libatk1.0-0 libatk-bridge2.0-0 \ libnss3 libnspr4 libdbus-1-3 libatk1.0-0 libatk-bridge2.0-0 \
libcups2 libdrm2 libxkbcommon0 libatspi2.0-0 libxcomposite1 \ libcups2 libdrm2 libxkbcommon0 libatspi2.0-0 libxcomposite1 \
libxdamage1 libxfixes3 libxrandr2 libgbm1 libpango-1.0-0 \ libxdamage1 libxfixes3 libxrandr2 libgbm1 libpango-1.0-0 \
@@ -105,10 +93,6 @@ COPY --from=builder /app/src ./src
COPY --from=builder /app/config ./config COPY --from=builder /app/config ./config
# Playwright browsers # Playwright browsers
COPY --from=builder /ms-playwright /ms-playwright COPY --from=builder /ms-playwright /ms-playwright
# Python venv for OCR worker (located at .venv inside subservices/ocr-worker)
COPY --from=builder /app/subservices/ocr-worker/.venv ./subservices/ocr-worker/.venv
COPY --from=builder /app/subservices/ocr-worker/src ./subservices/ocr-worker/src
COPY --from=builder /app/subservices/ocr-worker/pyproject.toml ./subservices/ocr-worker/pyproject.toml
# --- Non-root user for security --- # --- Non-root user for security ---
RUN groupadd -g 1001 appuser \ RUN groupadd -g 1001 appuser \
+19 -1
View File
@@ -21,6 +21,22 @@ services:
volumes: volumes:
- postgres_dev_data:/var/lib/postgresql/data - postgres_dev_data:/var/lib/postgresql/data
ocr-worker:
build:
context: .
dockerfile: deploy/docker/ocr-worker.Dockerfile
restart: unless-stopped
healthcheck:
test: ["CMD-SHELL", "curl -fsS http://127.0.0.1:8100/health || exit 1"]
interval: 5s
timeout: 5s
retries: 20
start_period: 15s
environment:
TZ: ${TZ:-Asia/Shanghai}
ports:
- "${OCR_PORT:-8100}:8100"
backend: backend:
build: build:
context: . context: .
@@ -33,6 +49,8 @@ services:
depends_on: depends_on:
postgres: postgres:
condition: service_healthy condition: service_healthy
ocr-worker:
condition: service_healthy
healthcheck: healthcheck:
test: test:
["CMD-SHELL", "curl -fsS http://127.0.0.1:3000/health/ready || exit 1"] ["CMD-SHELL", "curl -fsS http://127.0.0.1:3000/health/ready || exit 1"]
@@ -46,7 +64,7 @@ services:
DATABASE_URL: ${DATABASE_URL:-postgres://postgres:postgres@postgres:5432/order_site} DATABASE_URL: ${DATABASE_URL:-postgres://postgres:postgres@postgres:5432/order_site}
DATABASE_SSL: ${DATABASE_SSL:-false} DATABASE_SSL: ${DATABASE_SSL:-false}
DATABASE_MAX_CONNECTIONS: ${DATABASE_MAX_CONNECTIONS:-10} DATABASE_MAX_CONNECTIONS: ${DATABASE_MAX_CONNECTIONS:-10}
OCR_PROJECT_ROOT: /app/subservices/ocr-worker OCR_BASE_URL: http://ocr-worker:8100
CLAIM_BASE_URL: ${CLAIM_BASE_URL:-http://localhost/#/claim} CLAIM_BASE_URL: ${CLAIM_BASE_URL:-http://localhost/#/claim}
TENCENT_REDEEM_PROOF_MODE: ${TENCENT_REDEEM_PROOF_MODE:-full} TENCENT_REDEEM_PROOF_MODE: ${TENCENT_REDEEM_PROOF_MODE:-full}
TENCENT_BROWSER_HEADLESS: ${TENCENT_BROWSER_HEADLESS:-true} TENCENT_BROWSER_HEADLESS: ${TENCENT_BROWSER_HEADLESS:-true}
+17 -1
View File
@@ -15,6 +15,20 @@ services:
volumes: volumes:
- postgres_data:/var/lib/postgresql/data - postgres_data:/var/lib/postgresql/data
ocr-worker:
build:
context: .
dockerfile: deploy/docker/ocr-worker.Dockerfile
restart: unless-stopped
healthcheck:
test: ["CMD-SHELL", "curl -fsS http://127.0.0.1:8100/health || exit 1"]
interval: 5s
timeout: 5s
retries: 20
start_period: 15s
environment:
TZ: ${TZ:-Asia/Shanghai}
backend: backend:
build: build:
context: . context: .
@@ -25,6 +39,8 @@ services:
depends_on: depends_on:
postgres: postgres:
condition: service_healthy condition: service_healthy
ocr-worker:
condition: service_healthy
healthcheck: healthcheck:
test: ["CMD-SHELL", "curl -fsS http://127.0.0.1:3000/health/ready || exit 1"] test: ["CMD-SHELL", "curl -fsS http://127.0.0.1:3000/health/ready || exit 1"]
interval: 5s interval: 5s
@@ -37,7 +53,7 @@ services:
DATABASE_URL: ${DATABASE_URL:-postgres://postgres:postgres@postgres:5432/order_site} DATABASE_URL: ${DATABASE_URL:-postgres://postgres:postgres@postgres:5432/order_site}
DATABASE_SSL: ${DATABASE_SSL:-false} DATABASE_SSL: ${DATABASE_SSL:-false}
DATABASE_MAX_CONNECTIONS: ${DATABASE_MAX_CONNECTIONS:-20} DATABASE_MAX_CONNECTIONS: ${DATABASE_MAX_CONNECTIONS:-20}
OCR_PROJECT_ROOT: /app/subservices/ocr-worker OCR_BASE_URL: http://ocr-worker:8100
CLAIM_BASE_URL: ${CLAIM_BASE_URL:-http://localhost/#/claim} CLAIM_BASE_URL: ${CLAIM_BASE_URL:-http://localhost/#/claim}
TENCENT_REDEEM_PROOF_MODE: ${TENCENT_REDEEM_PROOF_MODE:-full} TENCENT_REDEEM_PROOF_MODE: ${TENCENT_REDEEM_PROOF_MODE:-full}
TENCENT_BROWSER_HEADLESS: ${TENCENT_BROWSER_HEADLESS:-true} TENCENT_BROWSER_HEADLESS: ${TENCENT_BROWSER_HEADLESS:-true}