Support Web Video Text Tracks Format subtitles

[schplurtz]

Hi,
video already have alternative media, and poster. This adds support for Web Video Text Tracks Format subtitles, based on the same principle. When displaying media foo.*, if there are media files named foo.kind.lang.vtt, then they are vtt tracks and added to the video. kind is a 3letter abbreviation of one of the 5 vtt kinds defined by the w3c, lang is a language abbreviation. for example foo.cap.fr.vtt or foo.sub.de.vtt. The idea comes from a similar request in the DW French forum. A user needs subtitle for the hard of hearing visiting his site.

See the patch in action here https://schplurtz.darktech.org/doku-test/start (for a few weeks only)

Hoping you'll like the idea too.
/Schplurtz

[mention-bot]

@schplurtz, thanks for your PR! By analyzing the history of the files in this pull request, we identified @splitbrain, @akate and @Klap-in to be potential reviewers.

[splitbrain]

I don't like the global language name array. In fact I would like to skip that array completely. Instead I would simply use the capture group from the regular expression and thus show the language code to the end user.

[schplurtz]

Just in case you did not notice, current code will show lang code if there is no match in the
global lang array.

Anyway, I suppress this part as you request, and use only the language code.

[splitbrain]

two more minor things before I merge this:

• the attributes should be properly enclosed in quotes, eg. kind="sub" instead of kind=sub
• even though it's probably not exploitable, I'd feel safer if $kind and $srclang would be escaped via hsc() before outputting them

[schplurtz]

Doing as you request, of course. I'll add a commit tonight to the PR.

But note that using hsc($kind) seems useless. $kind can only be one of subtitles, captions, descriptions, chapters and metadata : Initial value is validated by the regexp $re='/\\.(sub|cap|des|cha|met)\\.([^.]+)\\.vtt$/'; and then $kinds array is used to map the 3 letters to the word that is used in html.

$kinds=array(
        'sub' => 'subtitles',
        'cap' => 'captions',
        'des' => 'descriptions',
        'cha' => 'chapters',
        'met' => 'metadata'
);

[splitbrain]

You're totally right of course, but that happens in a different method... one of the two methods may change in the future and someone might not be aware of the implications and accidentally introduce a problem. Since it doesn't hurt to escape here, I prefer to be safe rather than sorry later ;-)

[schplurtz]

yes, that makes sense.

[splitbrain]

@schplurtz can you add infos about this to the wiki? I think it would make sense to add a new page at https://www.dokuwiki.org/video and link it from https://www.dokuwiki.org/images

[schplurtz]

@splitbrain, I already had some text for this. It's now on https://www.dokuwiki.org/video . I did my best, but please, find someone to review and check for correctness. I'm better at translating English to French than writing English. No link from /images yet.

Also I copied the whole https://www.dokuwiki.org/wiki:syntax#fallback_formats section to https://www.dokuwiki.org/video so that everything that deals with video is on a single page. Just seemed the right thing to do. Eventually, this section will have to be removed from the syntax.txt file.

Unfortunately, I am most probably going to be offline for some weeks now, so to finish this up, you'll have to ask someone else or remind me in, say, 4 or 5 weeks as I'm pretty sure I won't remember.