* Adding a route of zju/cse
* Change route path to include '/zju/' prefix
* Fix the route of zju/cse
* Fix the route of zju/cse
* Response to the reviewer
* Update comment for content extraction in index.ts
Removed comment about removing images from content extraction.
* fix: update zhihu daily API call
Updated story URL to use the API endpoint and adjusted data extraction
accordingly.
* fix: remove useless cache
* fix: use data from json instead
---------
- Fix Redis cache async handling by adding await to cache.get() calls
- Refactor request_key management into separate ensureRequestKey function
- Simplify main data fetching flow by removing complex retry logic
- Add proper error type guards and logging for better debugging
- Improve code readability and maintainability
Fixes issue where Skeb works route failed with Redis cache due to
cache.get() returning Promise object instead of string value, causing
invalid request_key in API calls.
* feat(route): add Wikipedia current events
Add RSS feed for Wikipedia Portal:Current events providing the past 7 days of current events. Parses MediaWiki templates and converts wiki markup to HTML.
- Hierarchical bullet point structures with nested <ul>/<li> tags
- Proper paragraph breaks between sections
- Correct nesting depth for multi-level lists
- Clean separation of content sections
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
* refactor(route/wikipedia): break down complex wikiToHtml function
- Split 123-line wikiToHtml function into 6 smaller, focused methods
- Each function now has a single responsibility (links, formatting, lists, etc.)
- Improved code readability and maintainability
- End-to-end tested: route produces identical RSS output with proper HTML formatting
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
* fix(route/wikipedia): use config.trueUA instead of hardcoded User-Agent
Replace hardcoded User-Agent string with config.trueUA to follow
project conventions for HTTP requests.
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
* fix(route/wikipedia): use config.cache.contentExpire for cache duration
Replace hardcoded 6-hour cache duration with configurable contentExpire
setting to follow project caching conventions.
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
* feat(route/wikipedia): implement efficient current events RSS with single API request
- Switch from web scraping to MediaWiki API for better reliability
- Implement single batch API request with continuation support for multiple pages
- Add current-day inclusion parameter with time-based logic (auto/always/never/hour)
- Add proper error handling and ESLint compliance
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
* refactor(route/wikipedia): consolidate list processing and add API documentation
- Move wrapListItems and processNestedLists calls inside (renamed) processListsAndLines to eliminate non-obvious coupling
- Remove newlines from processNestedLists output for more compact HTML
- Add documentation about Wikipedia API endpoints and query options
---------
Co-authored-by: Claude <noreply@anthropic.com>
* fix: improve date parsing error handling in gdufs xwxy-news route
- Add null check before calling parseDate to prevent errors when dateText is empty
- Return undefined instead of calling toUTCString() on potentially invalid date
* feat(gdufs/xwxy): add News Center feed with categories and Radar
- Add route /gdufs/xwxy/:category? (news|notices|announcements|media)
- Parse lists with fallbacks; absolute URLs for img/a; extract content across multiple templates; derive authors from meta; recover pubDate from meta if missing
- Respect ?limit (default 10) before detail fetch; use cache.tryGet for item pages
- Relax link filtering to include internal assets and resolved relative URLs; handle external links without fetching
- Normalize route url to xwxy.gdufs.edu.cn; simplify Radar sources
* fix: add data: and vbscript: protocol checks for security
* refactor(gdufs/xwxy): simplify selector logic using precise li[id^='line_u14_'] a
- Replace complex multi-level fallback selector logic (13 lines) with single precise selector
- Remove unnecessary filtering logic for href validation and domain checking
- Improve performance by eliminating multiple DOM queries
- Maintain same functionality while reducing code from 35 lines to 1 line
- Verified working on all category pages: news, notices, media focus
* fix: use h5 instead of p for title extraction
* refactor: remove excessive try-catch wrapper
* fix: remove descriptive text for external links
* refactor: simplify content selector to use .v_news_content only
* fix: use comma separator for multiple authors
* refactor(gdufs/xwxy): dedupe date regex into handler-scope constant; remove unnecessary escape