Now, we use Vimeo's built-in (sort of) support for getting download URLs. It also adds a test case for Vimeo scraping.
more working around Vimeo flakiness
add a test for Vimeo scraping
fix failing test case
When I got to the HTML like, I see several download links that the regex will match. Should we add a bit of logic to select the best one. Maybe make an attempt to pick the HD link, then fallback on picking any one.
I'll go and document that better, but that's what we're doing now. The HD is the first one, so we pick that by default. Otherwise, we'll pick the first other one that's left.
document output from Vimeo to our callback
Are you sure this is true for all videos? I went to http://vimeo.com/41751540 and got this output from that page:
Mobile .MP4 file (480x270 / 4MB)
HD .MP4 file (1280x720 / 36MB)
SD .MP4 file (640x360 / 11MB)
Original .MP4 file (1280x720 / 62MB)
Hrm; I did not expect that. I'll add a real check.
check all the Vimeo URLs; return the largest
I think this should be fine. I had one small nitpick/question. If you want to change something based on that, go for it. If I'm not reading the regex right, ignore it.
Either way, I think it's safe to merge and you should go for it once you either make the change or decide to ignore.