Scrapes Facebook profile pages and searches profiles using a normal log in (without an API key). Inspired by facebook-scraper. Built with MechanicalSoup.
NOTE: Python 3.7 or higher is required.
To install facebook_snooper from source, follow these steps:
# clone the repository $ git clone https://github.com/gsscoder/facebook-snooper.git # change the working directory $ cd facebook-snooper # install the package $ pip install .
You can also install with PyPI:
$ pip install facebook_snooper
>>> import facebook_snooper >>> from facebook_snooper.dump import dump_search, dump_info >>> fb = facebook_snooper.init_session() >>> fb.log_in('user@account.com', 'user_password') <facebook_snooper.core.session.Session object at 0x10ffc58d0> >>> results = fb.search('frank new york') >>> dump_search(results, pretty=True) profile frankdecaro https://m.facebook.com/frankdecaro?refid=46 Frank DeCaro Author at Rizzoli New York Freelance Writer at The New York Times profile frankbruninyt https://m.facebook.com/frankbruninyt?refid=46 Frank Bruni The New York Times New York, New York >>> info = fb.profile_info('frankbruninyt') >>> dump_info(info, pretty=True) Frank Bruni Image: https://scontent-fco1-1.xx.fbcdn.net/v/t1.0-1/cp0/e15/q65/p120x120/49... work The New York Times education Columbia University, education UNC Chapel Hill education Loomis Chaffee education Loomis Chaffee, Windsor, Conn. living New York living White Plains
Create test directory and data:
$ cd facebook-snooper
$ mkdir tests/pages>>> import facebook_snooper >>> from tests.persistence import save_login, save_current_page >>> fb = facebook_snooper.init_session() >>> save_login(fb) >>> fb.log_in('user@email.com', 'user_password') ... >>> save_current_page(fb, 'logged') >>> fb.search('frank new york') ... >>> save_current_page(fb, 'search') >>> fb.profile_info('frankbruninyt') ... >>> save_current_page(fb, 'profile')
Execute tests:
$ python tests/session.py
- This package cannot gather nothing more that is publicly visible.
- Please respect people's privacy and use it ethically.
- There is no guarantee that will work if scraped pages change too much.
- Andrew Polukhin (pandrey2003)