A Douban API server that provides an unofficial method for information gathering, currently, supporting several providers below:
- SeleniumProvider - which will open a headless Chrome to simulate the user interaction (Optional)
- HttpRequestProvider - which has leveraged
Requests & BeautifulSoup
for information gathering with no any browser interaction (By default) - P2PCacheProvider -
ToDo
(Scrapy/SQLite/IPFS/BitTorrent/Web3Storage/Resilio File Sync/etc) - ??????
Note: Any comments and issues are welcomed!
docker run --rm -d -p 5000:5000 caryyu/douban-openapi-server:latest
docker run --rm -d -p 5000:5000 caryyu/douban-openapi-server:<commit>
Note: the
commit
can be found from git history of main branch
- the toolkits of
python3
andpipenv
are required by executing the following scripts,
pipenv install
pipenv shell
export FLASK_APP=app.py
export FLASK_ENV=development
export FLASK_DEBUG=0
flask run
Note: There're two providers to choose, by default,
HttpRequestProvider
will take place, which will be faster for information scraping, forSelenium
, You have to follow the official instruction and build the docker image by yourself
This is only for research and study, Any copyright violation should count on your own, thanks.