mirror of
https://git.mirrors.martin98.com/https://github.com/jina-ai/reader
synced 2025-04-18 11:50:00 +08:00
23a3b807c9
* fix: fine allow redefining Function.prototype.toString * wip * wip * wip * wip * wip * wip * wip * fix: contentType encoding * wip * fix: error throwing * wip * fix * wip * fix * fix * fix: jsdom * wip * wip * fix: links summary uniqueness * wip * wip * robots-txt catch no robots.txt * deps: remove puppeteer-extra-plugin-stealth * fix: dont change waring type * fix: curl * fix: replace firebase-roundtrip-check with blackhole-detector * fix: black hole detection * sercher: black hole detecting * fix: no h2c for searcher * fix: bhd * fix: search and crawl conflict * fix: bhd * fix * fix: server script * canvas: fixed avif issue * logging: move some to debug * fix * fix: pptr declare ready only when page can be created without issues * fix: bhd * cd: cloud run deploy-health-check cannot complete pptr newPage * cd: fix * fix: curl body can be null * fix * fix * fix: major fix regarding TC pdfs * fix * fix * deps: fix civkit trie router issue * fix * boom: total restructure * cd: fix docker ctx * fix * fix: switch to h2c * cd: ensure http2
39 lines
1.2 KiB
Docker
39 lines
1.2 KiB
Docker
# syntax=docker/dockerfile:1
|
|
FROM lwthiker/curl-impersonate:0.6-chrome-slim-bullseye
|
|
|
|
FROM node:20
|
|
|
|
RUN apt-get update \
|
|
&& apt-get install -y wget gnupg \
|
|
&& wget -q -O - https://dl-ssl.google.com/linux/linux_signing_key.pub | apt-key add - \
|
|
&& sh -c 'echo "deb [arch=amd64] http://dl.google.com/linux/chrome/deb/ stable main" >> /etc/apt/sources.list.d/google.list' \
|
|
&& apt-get update \
|
|
&& apt-get install -y google-chrome-stable fonts-ipafont-gothic fonts-wqy-zenhei fonts-thai-tlwg fonts-kacst fonts-freefont-ttf libxss1 zstd \
|
|
--no-install-recommends \
|
|
&& rm -rf /var/lib/apt/lists/*
|
|
|
|
COPY --from=0 /usr/local/lib/libcurl-impersonate.so /usr/local/lib/libcurl-impersonate.so
|
|
|
|
RUN groupadd -r jina
|
|
RUN useradd -g jina -G audio,video -m jina
|
|
USER jina
|
|
|
|
WORKDIR /app
|
|
|
|
COPY package.json package-lock.json ./
|
|
RUN npm ci
|
|
|
|
COPY build ./build
|
|
COPY public ./public
|
|
COPY licensed ./licensed
|
|
|
|
RUN rm -rf ~/.config/chromium && mkdir -p ~/.config/chromium
|
|
|
|
ENV OVERRIDE_CHROME_EXECUTABLE_PATH=/usr/bin/google-chrome-stable
|
|
ENV LD_PRELOAD=/usr/local/lib/libcurl-impersonate.so CURL_IMPERSONATE=chrome116 CURL_IMPERSONATE_HEADERS=no
|
|
ENV PORT=8080
|
|
|
|
EXPOSE 3000 3001 8080 8081
|
|
ENTRYPOINT ["node"]
|
|
CMD [ "build/stand-alone/crawl.js" ]
|