Comic Strip Browser Files

Changelog on GitHub — full README

v4.0.3

• This version is mainly meant to address the cases when curl_cffi fails to work on some Windows systems!
• Automatically fall back to using requests (Legacy mode) if curl_cffi fails due to unexpected errors or after 7 date fallback failures.
• Saved the legacy_mode setting in cache/preferences.json to persist across sessions.
• Added a “Legacy mode!” status indicator to the left of the “Prefer Firefox“ button.

v4.0.2

• This version is mainly for debugging and revealing the errors that break this app on some Windows systems!
• The next version will automatically fall back to using requests instead of curl_cffi. The current version doesn’t do that!
• Enabled conditional debug mode triggered by a debug.txt file located in the application root or cache/ folder.
• Redirected debug log output to cache/comic_browser.log so it can be viewed directly via the “Open Cache” button.
• Added a “Debug mode!“ status indicator to the left of the “Prefer Firefox“ button when debug mode is active.
• Unmasked generic errors in the GUI viewer to show the exact underlying exception details instead of a generic message.

v4.0.1

• As it’s often the case with dot-zero releases, 4.0.0 had severe bugs. Upgrading to this version is a must! HOWEVER, I have reports that version 4.0.1 doesn’t work on some Windows systems, and version 4.0.0 freezes when GoComics acts up. For now, the issue is under investigation, as I couldn’t reproduce it.
• Fixed a bug that made curl_cffi send worse browser headers than requests and triggered more blocks from GoComics.
• Fixed several browser-specific bugs that prevented the correct collection and the use of GoComics clearance cookies.
• Fixed several bugs that made curl_cffi use different headers than those in the clearance cookies.
• Fixed a browser solver HTML discard bug that created redundant network calls.
• Fixed browsers installed as per-user Flatpaks not being detected (e.g., com.brave.Browser).
• Added continuous Referer tracking to mimic the behavior of a real browser.
• Added buttons to override the OS-specific priorities and prefer Firefox or Chromium/Edge to solve Bunny Shield challenges.
• Preferences (this one and the hiding of GoComics titles) are persistent, but are deleted when the cache is manually deleted.

v4.0.0

• GoComics: implemented mechanisms to defeat the Bunny Shield challenge using a headless browser!
• Windows: browsers are queried in this order: Edge, Chrome, Brave, Firefox.
• Linux: browsers are queried in this order: firefox-esr, firefox, chromium, google-chrome (plus Flatpak).
• GoComics: in some cases, accessing GoComics via a VPN should now be possible!
• GoComics: the removal of the 403 block is impossible; change the server/IP.
• GoComics: the removal of the 429 limitation is impossible; just wait.
• CK: Fixed bugs related to Python and browsers receiving different contents.
• Overhauled the web-related error detection.
• Improved the feedback about what the app is trying to do.
• New, unknown bugs added by Gemini :)

v3.1.2

• GoComics: refined the message for the case of VPN blocking.
• GoComics: added a message for the “403 Forbidden” response.

v3.1.1

• The Up/Down shortcuts are now mentioned in the UI alongside the others.

v3.1.0

• Fixed Dick Tracy from ComicsKingdom, which displayed comics from GoComics.
• New titles added from GoComics: Get Fuzzy, Herman.
• GoComics: handling of the “429 Too Many Requests” response.
• GoComics: handling of VPN blocking.
• “Hide GoComics” option added (temporary, not saved).
• Functional update: Up/Down arrow keys now directly navigate comic titles, mirroring the PgUp/PgDn keys.
• AppImage changes: added comic-strip-browser.appdata.xml; rebased on Ubuntu 22.04 for a broader compatibility.
• Minor code cleanup and consistency fixes.

v3.0.0

• Major release with substantial changes under the hood. Despite being a 0.0 version, I hope the new bugs aren't a deal-breaker.
• 96 comic titles in total, or 91 if we only count the unique ones (a few are retrieved from 2 or 3 sources).
• Comics Kingdom removed Animal Crackers. GoComics still has it.
• Mother Goose and Grimm is now also retrieved from grimmy.com. This extends its first day of availability from 2002-11-25 back to 1994-01-01.
• Shoe now has 3 sources: GoComics, Comics Kingdom, and ShoeComics. The latter extends availability slightly back to 2001-01-01.
• 16 new comic titles: Arctic Circle, The Barn, Break of Day, Edge City, Goomer, Loose Parts, Mutt & Jeff, Never Been Deader, Pooch Café, Safe Havens, Sam and Silo, Savage Chickens, Scary Gary, Take it from the Tinkersons, Tiger, Zack Hill.
• The GUI has been slightly modernized.
• Keyboard navigation has been extended, and the shortcuts are printed on the screen (PgUp/PgDn: Previous/Next title; Left/Right: Previous/Next date; Home: Earliest date; End: Today). This should make mouse-free navigation possible.
• Upon public request, CLI-only operation support has been added. See README.md (github.com) for details. The main request was to be able to download to the cache folder all the comics for today (or for yesterday). Now that PgUp/PgDn navigates in the list of comics, this makes even more sense than when it was requested.
• Image auto-scaling logic has been slightly adjusted to accommodate comics of various sizes better.
• python3 comic_titles.py generates COMIC_TITLES.md (github.com) with more detail than the list in README.md (github.com).
• Basic information for developers and agents can now be found in ARCHITECTURE.md (github.com).
• Added helpers for AI-assisted development: an AGENTS.md (github.com) suggestion, a PLAN.md (github.com) template, and LLM_HOWTO.md (github.com).