[FunnyOrDie] Add support for "/articles/" URLs
Parse the xml document instead of using regexes, otherwise unicode characters are left unescaped.
When the f4m downloader gets live stream support, I expect this to work magically or with very minor changes.
The extractor works fine, but the f4m downloader cannot handle the resulting bootstrap information.
Calling add_default_extractors twice should be harmless since the first set of extractors will match.
Now they use an XML format instead of JSON.