clive-delfi.patch

   1 --- clive-0.4.10/src/clive/parse.py     2008-04-21 10:54:55.000000000 +0300
   2 +++ clive-0.4.10-delfi/src/clive/parse.py       2008-04-26 01:09:31.000000000 +0300
   3 @@ -50,6 +50,7 @@
   4              ('sevenload.com', 'sevenl', self._parse_sevenload),
   5              ('myvideo.', 'myvideo', self._parse_myvideo),
   6              ('break.com', 'break', self._parse_break),
   7 +            ('delfi.', 'delfi', self._parse_delfi),
   8          ]
   9
  10      ## Parses a video page data (HTML)
  11 @@ -221,6 +222,72 @@
  12          url = self._parse_from_to(data, 'mediaURL=', '&', skip_from=1)
  13          return (url, vid, low_quality)
  14
  15 +    def _graburl(self, url, read=1):
  16 +        from urlgrabber.grabber import URLGrabber, URLGrabError
  17 +        data = ''
  18 +        try:
  19 +            g = URLGrabber(
  20 +                http_headers = (('accept-encoding', 'gzip'),),
  21 +                user_agent = self._opts.http_agent,
  22 +                proxies = self._opts.http_proxy
  23 +            )
  24 +            o = g.urlopen(url)
  25 +            if read:
  26 +                data = o.read()
  27 +                if o.hdr.get('content-encoding') == 'gzip':
  28 +                    data = gzip.GzipFile(fileobj=StringIO(data)).read()
  29 +            return data
  30 +        except URLGrabError, err:
  31 +            self._say('%s [%s]' % (e.strerror,url), is_error=1)
  32 +            raise SystemExit
  33 +        return data
  34 +
  35 +    def _parse_delfi(self, url, data, low_quality):
  36 +        url = None
  37 +
  38 +        s = self._parse_from_to(data, "flv_url:", ",")
  39 +        if s:
  40 +            # videobox
  41 +            # http://www.delfi.ee/news/paevauudised/paevavideo/article.php?id=15218215
  42 +            try:
  43 +                url = s.replace("'", "").split('flv_url: ',2)[1]
  44 +            except IndexError:
  45 +                pass
  46 +
  47 +        if not url:
  48 +            # videoproject embed
  49 +            # http://www.delfi.ee/news/paevauudised/paevavideo/article.php?id=18759038
  50 +            s = self._parse_from_to(data, "_delfiVideoSalt", ";")
  51 +            salt = None
  52 +            try:
  53 +                salt = s.split('"',2)[1]
  54 +            except IndexError:
  55 +                pass
  56 +
  57 +            if salt != None:
  58 +                # find url to videoproject
  59 +                s = self._parse_from_to(data, "src=", "/js/embed.js")
  60 +                s = s[5 + s.rfind("src"):] + "/video/" + salt + "/"
  61 +                # fetch video page
  62 +                s = self._graburl(s)
  63 +                s = self._parse_from_to(s, "s1.addVariable('file'", ";");
  64 +                try:
  65 +                    url = s.split("'")[3]
  66 +                except IndexError:
  67 +                    pass
  68 +
  69 +        if not url:
  70 +            # videoproject
  71 +            # http://video.delfi.ee/video/Uu4gF58g/
  72 +            s = self._parse_from_to(data, "s1.addVariable('file'", ";");
  73 +            try:
  74 +                url = s.split("'")[3]
  75 +            except IndexError:
  76 +                pass
  77 +
  78 +        vid = md5.new(str(time.time())).hexdigest()[:8]
  79 +        return (url, vid, low_quality)
  80 +
  81      def _random_vid(self):
  82          return md5.new(str(time.time())).hexdigest()[:8]
  83