Support for filmweb.no

[forthrin]

~/$ youtube-dl -v http://www.filmweb.no/trailere/article1264921.ece
[debug] System config: []
[debug] User config: []
[debug] Command-line args: [u'-v', u'http://www.filmweb.no/trailere/article1264921.ece']
[debug] Encodings: locale UTF-8, fs utf-8, out UTF-8, pref UTF-8
[debug] youtube-dl version 2016.03.01
[debug] Python version 2.7.11 - Darwin-15.3.0-x86_64-i386-64bit
[debug] exe versions: ffmpeg 2.8.3, ffprobe 2.8.3, rtmpdump 2.4
[debug] Proxy map: {}
[generic] article1264921: Requesting header
WARNING: Falling back on generic information extractor.
[generic] article1264921: Downloading webpage
[generic] article1264921: Extracting information
ERROR: Unsupported URL: http://www.filmweb.no/trailere/article1264921.ece

[forthrin]

Any ETA on this?

[yan12125]

No. There are lots of new site requests, and few developers.

[TRox1972]

For developers: 23video embed

[forthrin]

How do you figure out how the URL to the actual video stream for a particular site is generated? By setting breakpoints in the browser console and stepping through the JavaScript code?

If your standard approach or best practice could be explained, maybe this would make it easier for people to help out.

[TRox1972]

@forthrin I've made an extractor for the site. It's working, but needs improvement before it's ready to be merged with master.

[forthrin]

That's great news! Be sure to notify this thread about final commits to the release branch. Any comments on my question regarding discovering how the URL is generated?

[TRox1972]

@forthrin That depends on the site. A general approach for simple sites is to search the webpage or player page (search for iframes) for video-links or f4m- or m3u8-manifests. If the file hosting is done by another site, e.g. Kaltura, or 23video in this case, you often need only to extract the id of the video and delegate the rest of the extraction to the extractor for the other site. (Anonymous) APIs are also often valuable. In some cases, though, when the provider generates the URL through javascript and obfuscates it, you'll need to reverse engineer the relevant js code, but this is usually only the case of larger providers like openload.

The extractor for vidbit.com is relatively simple if you know some Python, and may be a good starting point if you want to create an extractor of your own; also read the developer instructions if you're gonna do this.

[remitamine]

will be supported in the next version.