Vidzi is broken

[davek-cygwin]

C:\Projects\rg3\youtube-dl>python -m youtube_dl "http://vidzi.tv/jnk3om3hftz4.html" --verbose 2>&1 | tee fail-vidzi.log

[debug] System config: []
[debug] User config: []
[debug] Command-line args: [u'http://vidzi.tv/jnk3om3hftz4.html', u'--verbose']
[debug] Encodings: locale cp1252, fs mbcs, out None, pref cp1252
[debug] youtube-dl version 2015.09.09
[debug] Git HEAD: 1721fef
[debug] Python version 2.7.8 - Windows-7-6.1.7601-SP1
[debug] exe versions: ffmpeg 0.8.1, rtmpdump 2.4
[debug] Proxy map: {}
[Vidzi] jnk3om3hftz4: Downloading webpage
ERROR: Unable to extract video url; please report this issue on https://yt-dl.org/bug . Make sure you are using the latest version; see https://yt-dl.org/update on how to update. Be sure to call youtube-dl with the --verbose flag and include its complete output.
Traceback (most recent call last):
File "youtube_dl\YoutubeDL.py", line 660, in extract_info
ie_result = ie.extract(url)
File "youtube_dl\extractor\common.py", line 287, in extract
return self.real_extract(url)
File "youtube_dl\extractor\vidzi.py", line 24, in real_extract
r'{\s_file\s:\s"([^"]+)"\s_}', webpage, 'video url')
File "youtube_dl\extractor\common.py", line 584, in _html_search_regex
res = self._search_regex(pattern, string, name, default, fatal, flags, group)
File "youtube_dl\extractor\common.py", line 575, in _search_regex
raise RegexNotFoundError('Unable to extract %s' % _name)
RegexNotFoundError: Unable to extract video url; please report this issue on https://yt-dl.org/bug . Make sure you are using the latest version; see https://yt-dl.org/update on how to update. Be sure to call youtube-dl with the --verbose flag and include its complete output.

[davek-cygwin]

Body of webpage: https://gist.github.com/davek-cygwin/9ba2d360d963dbc0495f

Relevant part is:

<script type="text/javascript" src="http://static.vidzi.tv/nplayer/jwplayer.js"></script>

                      <script type='text/javascript'>eval(function(p,a,c,k,e,d){while(c--)if(k[c])p=p.replace(new RegExp('\\b'+c.toString(a)+'\\b','g'),k[c]);return p}('4("2t").2s({2r:[{2q:"2://w.u.t.s:r/i/2p/q/9.2o",2n:"2://2m.6.8/2l/6.2k",2j:[{a:"2://6.8/9-o.2i?2h=",2g:\'2f\'},{a:"2://w.u.t.s:r/2e/v.2d"}],2c:[{a:"2://6.8/p/q/2b.p",2a:"29",n:"28"},{a:"2://6.8/9-o.27",n:"26"}]}],25:24,23:"22",21:3,1z:1y,1x:1,1w:1v,1u:1t,1s:{1r:l,1q:\'1p\',1o:\'1n\',1m:20},1l:1k});h b;h g=0;h 7=0;4().1j(5(x){c(7>0)g+=x.m-7;7=x.m;c(0!=0&&g>=0){7=-1;4().1i();4().1h(l);$(\'#1g\').f();$(\'e.d\').f()}});4().1f(5(x){7=-1});4().1e(5(x){k(x)});4().1d(5(){$(\'e.d\').f()});5 k(x){$(\'e.d\').1c();c(b)1b;b=1;$.1a(\'2://6.8/j-19/18.j?17=16&15=9&14=13-12-11-10-z\',5(y){})}',36,102,'||http||jwplayer|function|vidzi|p0242296|tv|jnk3om3hftz4|file|vvplay|if|video_ad|div|show|tt242296|var||cgi|doPlay|false|position|kind|c4d4eb72532019084b9e8b42c1d68f0a|srt|00048|8777|155|98|81||77||data|7ce405bc990115957951f5a3aad0f2e7|1441844182|210|85|242296|hash|file_code|ajax|op|index_dl|bin|get|return|hide|onComplete|onPlay|onSeek|play_limit_box|setFullscreen|stop|onTime|640|width|fontsize|Verdana|fontFamily|FFFFFF|color|back|captions|900|hls_maxbackbufferlength|300|hls_maxbufferlength|hls_startfromlevel|true|androidhls||hls_minbufferlength|flash|primary|360|height|thumbnails|vtt|subtitles|Croatian|label|jnk3om3hftz4_Croatian|tracks|mp4|bfuqidr73y2qedz7nktr5ejkl4pzadqtgjicmgo6lvaoqjl3evsehut3zdva|hls|type|embed|m3u8|sources|swf|nplayer|static|provider|jpg|01|image|playlist|setup|vplayer'.split('|')))

</script>

This decodes to:

jwplayer("vplayer").setup({playlist:[{image:"http://77.81.98.155:8777/i/01/00048/jnk3om3hftz4.jpg",provider:"http://static.vidzi.tv/nplayer/vidzi.swf",sources:[{file:"http://vidzi.tv/jnk3om3hftz4-c4d4eb72532019084b9e8b42c1d68f0a.m3u8?embed=",type:'hls'},{file:"http://77.81.98.155:8777/bfuqidr73y2qedz7nktr5ejkl4pzadqtgjicmgo6lvaoqjl3evsehut3zdva/v.mp4"}],tracks:[{file:"http://vidzi.tv/srt/00048/jnk3om3hftz4_Croatian.srt",label:"Croatian",kind:"subtitles"},{file:"http://vidzi.tv/jnk3om3hftz4-c4d4eb72532019084b9e8b42c1d68f0a.vtt",kind:"thumbnails"}]}],height:360,primary:"flash",hls_minbufferlength:3,androidhls:true,hls_startfromlevel:1,hls_maxbufferlength:300,hls_maxbackbufferlength:900,captions:{back:false,color:'FFFFFF',fontFamily:'Verdana',fontsize:20},width:640});var vvplay;var tt242296=0;var p0242296=0;jwplayer().onTime(function(x){if(p0242296>0)tt242296+=x.position-p0242296;p0242296=x.position;if(0!=0&&tt242296>=0){p0242296=-1;jwplayer().stop();jwplayer().setFullscreen(false);$('#play_limit_box').show();$('div.video_ad').show()}});jwplayer().onSeek(function(x){p0242296=-1});jwplayer().onPlay(function(x){doPlay(x)});jwplayer().onComplete(function(){$('div.video_ad').show()});function doPlay(x){$('div.video_ad').hide();if(vvplay)return;vvplay=1;$.get('http://vidzi.tv/cgi-bin/index_dl.cgi?op=ajax&file_code=jnk3om3hftz4&hash=242296-85-210-1441844182-7ce405bc990115957951f5a3aad0f2e7',function(data){})}

[remitamine]

this can be fixed temporaly by adding this code to vidzi.py:

        def unpack_packer(p, a, c, k, s):
            k = k.split(s)
            import gmpy2
            for i in range(int(c) - 1, 1, -1):
                p = re.sub('\\b' + gmpy2.digits(i, int(a)) + '\\b', k[i], p)
            return p

        def unpack(content):
            packers = re.findall(r'function\(p,a,c,k,e,d\){.+}\(\'.*\',\d+,\d+,\'[^\']+\'\.split\(\'.\'\)', content)
            for (packer) in packers:
                p, a, c, k, s = re.search(r'function\(p,a,c,k,e,d\){.+}\(\'(.*)\',(\d+),(\d+),\'([^\']+)\'\.split\(\'(.)\'\)', packer).groups()
                content = content.replace(packer, unpack_packer(p, a, c, k, s))
            return content
        webpage = unpack(webpage)

after this line:

        webpage = self._download_webpage(url, video_id)

it requires to install gmpy2.
for linux it can be installed using this command:

pip install gmpy2

[yan12125]

@remitamine Thanks for proposing a solution. I guess gmpy2.digits(x, base) can be implemented in pure Python codes. Using pure Python reduces the effort to make the extractor work on Windows, which is used by @davek-cygwin

[remitamine]

I guess gmpy2.digits(x, base) can be implemented in pure Python codes.

there are a lot of solution but i don't know what code will be accepted to add to youtube-dl because i didn't find a code with the same licence(as i know you accept only code from projects licenced under the public domain)

for windows, gmpy2 can be downloaded from here:
https://code.google.com/p/gmpy/downloads/list

[yan12125]

Thanks for the report and all that provide ideas. Vidzi will work again in the next version.