The only platform-specific code is the data used for automatically locating cookies files for browsers.
Expand support to include well-known cookie file locations for all significant browsers on all desktop operating systems.
These platforms should be fully-supported:
- Linux
- MacOS
- Windows
Fully-supported browsers should include:
- Chrome
- Firefox
- Microsoft Edge
- Safari
Also consider supporting other significant browsers, for example:
- Arc
- Brave
- Gnome Web
- Konqueror
- Opera
- Vivaldi
In theory, additional browser support should be mostly a matter of determining possible file locations and SQL column mappings, if using SQLite.
Decryption is coded and tested as follows:
- MacOS with Chrome.
- Linux (Debian) with Chromium running in a Gnome desktop environment.
- Windows with Edge and Chrome.
Useful related links:
- https://github.com/n8henrie/pycookiecheat
- https://github.com/pyca/cryptography
- https://gist.github.com/DakuTree/428e5b737306937628f2944fbfdc4ffc#file-decryptchromecookies-py
- https://stackoverflow.com/questions/60416350/chrome-80-how-to-decode-cookies
Current sorting is hard-coded. In the future allow users to specify sort order.
Allow filter expression values to be regular expressions when the assignment
character is '~' rather than '='. E.g. 'value~myvalue-\d+'.
Support date comparison with ">" and "<" applied to standard date strings.
Support other field comparisons, like booleans, if they are useful.
This capability requires the ability to update a database or rewrite the binary cookies file format without corrupting it.
For now, cookie databases are queried for all cookies, and then filtered and sorted in Python post-processing code. But it avoids excess memory consumption by using iterators, rather than holding all cookies in memory.
Look for a simple cross-platform way to wrap the tool in a GUI. Perhaps Tk would be good enough.