GitLab

They were being caught by YoutubeUserIE, now it only extracts a url if the rest of extractors aren't suitable.
Now the url tests check that the urls can only be extracted with an specific extractor.

Urls like https://www.youtube.com/feed/watch_later were being as users (before the last changes to YoutubeUserIE, as videos)

There could be duplicate videos or other videos if the description have links.

Vevo sig changed again, please update for us! Thanks very much! (fixes #...

Video urls like http://youtube.com/BaW_jenozKc are not valid, but http://youtu.be/BaW_jenozKc is correct.

'media$player' is not defined for private videos.

[generic] Recognize html5 video in the format '<video src=".+?"' and only unquote the url when extracting the id (fixes #1372)

[daum] Get the video page from a canonical url to extract the full id (fixes #1373) and extract description.

add support for "-f mp4" for YouTube

defense.gouv.fr

In some pages the html tags are not closed, python 2.6 cannot handle it.

They are listed in the "supportedsites.html" page.