- updated to 0.5.0

[packages/clive.git] / clive-delfi.patch

--- clive-0.4.10/src/clive/parse.py	2008-04-21 10:54:55.000000000 +0300
+++ clive-0.4.10-delfi/src/clive/parse.py	2008-04-26 01:09:31.000000000 +0300
@@ -50,6 +50,7 @@
             ('sevenload.com', 'sevenl', self._parse_sevenload),
             ('myvideo.', 'myvideo', self._parse_myvideo),
             ('break.com', 'break', self._parse_break),
+            ('delfi.', 'delfi', self._parse_delfi),
         ]
 
     ## Parses a video page data (HTML)
@@ -221,6 +222,72 @@
         url = self._parse_from_to(data, 'mediaURL=', '&', skip_from=1)
         return (url, vid, low_quality)
 
+    def _graburl(self, url, read=1):
+        from urlgrabber.grabber import URLGrabber, URLGrabError
+        data = ''
+        try:
+            g = URLGrabber(
+                http_headers = (('accept-encoding', 'gzip'),),
+                user_agent = self._opts.http_agent,
+                proxies = self._opts.http_proxy
+            )
+            o = g.urlopen(url)
+            if read:
+                data = o.read()
+                if o.hdr.get('content-encoding') == 'gzip':
+                    data = gzip.GzipFile(fileobj=StringIO(data)).read()
+            return data
+        except URLGrabError, err:
+            self._say('%s [%s]' % (e.strerror,url), is_error=1)
+            raise SystemExit
+        return data
+
+    def _parse_delfi(self, url, data, low_quality):
+        url = None
+
+        s = self._parse_from_to(data, "flv_url:", ",")
+        if s:
+            # videobox
+            # http://www.delfi.ee/news/paevauudised/paevavideo/article.php?id=15218215
+            try:
+                url = s.replace("'", "").split('flv_url: ',2)[1]
+            except IndexError:
+                pass
+
+        if not url:
+            # videoproject embed
+            # http://www.delfi.ee/news/paevauudised/paevavideo/article.php?id=18759038
+            s = self._parse_from_to(data, "_delfiVideoSalt", ";")
+            salt = None
+            try:
+                salt = s.split('"',2)[1]
+            except IndexError:
+                pass
+
+            if salt != None:
+                # find url to videoproject
+                s = self._parse_from_to(data, "src=", "/js/embed.js")
+                s = s[5 + s.rfind("src"):] + "/video/" + salt + "/"
+                # fetch video page
+                s = self._graburl(s)
+                s = self._parse_from_to(s, "s1.addVariable('file'", ";");
+                try:
+                    url = s.split("'")[3]
+                except IndexError:
+                    pass
+
+        if not url:
+            # videoproject
+            # http://video.delfi.ee/video/Uu4gF58g/
+            s = self._parse_from_to(data, "s1.addVariable('file'", ";");
+            try:
+                url = s.split("'")[3]
+            except IndexError:
+                pass
+
+        vid = md5.new(str(time.time())).hexdigest()[:8]
+        return (url, vid, low_quality)
+
     def _random_vid(self):
         return md5.new(str(time.time())).hexdigest()[:8]