From b487ef0833678930ed58901ca67d3a13b84007ff Mon Sep 17 00:00:00 2001 From: Philipp Hagemeister Date: Wed, 14 Sep 2011 21:17:05 +0200 Subject: [PATCH] Fully implement comedycentral downloader --- youtube-dl | 70 +++++++++++++++++++++++++++++++----------------------- 1 file changed, 40 insertions(+), 30 deletions(-) diff --git a/youtube-dl b/youtube-dl index 50d83cf2d..3ecae2994 100755 --- a/youtube-dl +++ b/youtube-dl @@ -822,7 +822,7 @@ class FileDownloader(object): # Download using rtmpdump. rtmpdump returns exit code 2 when # the connection was interrumpted and resuming appears to be # possible. This is part of rtmpdump's normal usage, AFAIK. - basic_args = ['rtmpdump'] + [[], ['-W', player_url]][player_url is not None] + ['-r', url, '-o', tmpfilename] + basic_args = ['rtmpdump', '-q'] + [[], ['-W', player_url]][player_url is not None] + ['-r', url, '-o', tmpfilename] retval = subprocess.call(basic_args + [[], ['-e', '-k', '1']][self.params.get('continuedl', False)]) while retval == 2 or retval == 1: prevsize = os.path.getsize(tmpfilename) @@ -832,6 +832,11 @@ class FileDownloader(object): cursize = os.path.getsize(tmpfilename) if prevsize == cursize and retval == 1: break + # Some rtmp streams seem abort after ~ 99.8%. Don't complain for those + if prevsize == cursize and retval == 2 and cursize > 1024: + self.to_screen(u'\r[rtmpdump] Could not download the whole video. This can happen for some advertisements.') + retval = 0 + break if retval == 0: self.to_screen(u'\r[rtmpdump] %s bytes' % os.path.getsize(tmpfilename)) self.try_rename(tmpfilename, filename) @@ -3055,6 +3060,9 @@ class ComedyCentralIE(InfoExtractor): def report_config_download(self, episode_id): self._downloader.to_screen(u'[comedycentral] %s: Downloading configuration' % episode_id) + def report_index_download(self, episode_id): + self._downloader.to_screen(u'[comedycentral] %s: Downloading show index' % episode_id) + def report_player_url(self, episode_id): self._downloader.to_screen(u'[comedycentral] %s: Determining player URL' % episode_id) @@ -3102,36 +3110,38 @@ class ComedyCentralIE(InfoExtractor): return epTitle = mobj.group('episode') - mMovieParams = re.findall('', html) + mMovieParams = re.findall('', html) if len(mMovieParams) == 0: self._downloader.trouble(u'ERROR: unable to find Flash URL in webpage ' + url) return - show_id = mMovieParams[0][2] - ACT_COUNT = { # TODO: Detect this dynamically - 'thedailyshow.com': 4, - 'colbertnation.com': 3, - }.get(show_id, 4) - OFFSET = { - 'thedailyshow.com': 1, - 'colbertnation.com': 1, - }.get(show_id, 1) - first_player_url = mMovieParams[0][0] - startMediaNum = int(mMovieParams[0][3]) + OFFSET - movieId = mMovieParams[0][1] - - playerReq = urllib2.Request(first_player_url) + playerUrl_raw = mMovieParams[0][0] self.report_player_url(epTitle) try: - playerResponse = urllib2.urlopen(playerReq) + urlHandle = urllib2.urlopen(playerUrl_raw) + playerUrl = urlHandle.geturl() except (urllib2.URLError, httplib.HTTPException, socket.error), err: - self._downloader.trouble(u'ERROR: unable to download player: %s' % unicode(err)) + self._downloader.trouble(u'ERROR: unable to find out player URL: ' + unicode(err)) return - player_url = playerResponse.geturl() - for actNum in range(ACT_COUNT): - mediaNum = startMediaNum + actNum - mediaId = movieId + str(mediaNum) + uri = mMovieParams[0][1] + indexUrl = 'http://shadow.comedycentral.com/feeds/video_player/mrss/?' + urllib.urlencode({'uri': uri}) + self.report_index_download(epTitle) + try: + indexXml = urllib2.urlopen(indexUrl).read() + except (urllib2.URLError, httplib.HTTPException, socket.error), err: + self._downloader.trouble(u'ERROR: unable to download episode index: ' + unicode(err)) + return + + idoc = xml.etree.ElementTree.fromstring(indexXml) + itemEls = idoc.findall('.//item') + for itemEl in itemEls: + mediaId = itemEl.findall('./guid')[0].text + shortMediaId = mediaId.split(':')[-1] + showId = mediaId.split(':')[-2].replace('.com', '') + officialTitle = itemEl.findall('./title')[0].text + officialDate = itemEl.findall('./pubDate')[0].text + configUrl = ('http://www.comedycentral.com/global/feeds/entertainment/media/mediaGenEntertainment.jhtml?' + urllib.urlencode({'uri': mediaId})) configReq = urllib2.Request(configUrl) @@ -3149,7 +3159,7 @@ class ComedyCentralIE(InfoExtractor): turls.append(finfo) if len(turls) == 0: - self._downloader.trouble(u'\nERROR: unable to download ' + str(mediaNum) + ': No videos found') + self._downloader.trouble(u'\nERROR: unable to download ' + mediaId + ': No videos found') continue # For now, just pick the highest bitrate @@ -3157,25 +3167,25 @@ class ComedyCentralIE(InfoExtractor): self._downloader.increment_downloads() - effTitle = show_id.replace('.com', '') + '-' + epTitle + effTitle = showId + '-' + epTitle info = { - 'id': str(mediaNum), + 'id': shortMediaId, 'url': video_url, - 'uploader': show_id, - 'upload_date': 'NA', + 'uploader': showId, + 'upload_date': officialDate, 'title': effTitle, 'stitle': self._simplify_title(effTitle), 'ext': 'mp4', 'format': format, 'thumbnail': None, - 'description': 'TODO: Not yet supported', - 'player_url': player_url + 'description': officialTitle, + 'player_url': playerUrl } try: self._downloader.process_info(info) except UnavailableVideoError, err: - self._downloader.trouble(u'\nERROR: unable to download ' + str(mediaNum)) + self._downloader.trouble(u'\nERROR: unable to download ' + mediaId) continue