Project Management: - Get viewcvs linked to the app - [DONE] Nightly builds? (requested with Craig) - 0.5 public release http://jakarta.apache.org/commons/releases/index.html - maven? - (DONE) All FeedParser exceptions should include the URL of the feed if possible. This will really help with debugging - (DONE) Implementation of RSS/Atom autodiscovery... - (DONE) Remove ALL references to newsmonster IO layer. - (DONE) Support MetaFeedParser.onCreated() for RSS 2.0 /rss/channel/item/pubdate - (DONE) Implement RSS 2.0 enclosure linkage ... this should be an onLink handler. - (DONE) Atom GUIDs and RSS 2.0 GUIDs - (DONE) Rework the factory mechanism to support multiple FeedParsers... should be an interface. - How do I want to maintain a public changelog? - How do I want to maintain a public TODO? - UTF-16 tests for unicode and checksum the content. - Fix the feedparsing bug where we'll drop chars: current-broken-drop-accents.atom The bug is in getCorrectInputStream - Networking layer should support per-request UserAgent settings. This should just be a request header I think - Won't build with the maven version with Jaxen 1.0. What version of Jaxen are we running? This is a big problem if we can't figure this out. - The older betas are NOT present so I really don't have any way to figure this out... Do I? When did I add it to NewsMonster? - What build of JDOM are we running? 1.0beta9 built January 24 2003 - Would it be possible to dump JDOM and use DOM with Jaxen until we're ready to use SAX? - Maven javadoc is broken. Why is this? - Better Atom support: - Support Base64 Atom values and the ability to enable them to be automatically decoded. - Atom's xml:base is NOT supported right now. We NEED to support this. - Do we support multiple content items in Atom? - Where do we store global options like USER_AGENT, STRICT_PARSING, and STRICT_SPECw - Unit tests don't need to be so LOUD!! - tests for: - RSS and Atom GUIDs - Atom content - Atom summary - xml:base expansion - http://www.intertwingly.net/stories/2004/04/04/title.html - http://feedparser.org/docs/html-sanitization.html#advanced.sanitization.why Hm... - FIXME: documentation on using HTTP authentication - BUG: Make SURE that all these tests work http://diveintomark.org/tests/client/autodiscovery/ - We do not support multipart/alternative in the feedparser. - Migrate away from JDOM (slower than just using DOM/SAX) - Test cases shouldn't use absolute paths. - Support textinput which we don't support now - Break down requirements into specific versions of libraries that we need.