diff --git a/youtube-dl b/youtube-dl index 43566b6153..64f62e8a13 100755 --- a/youtube-dl +++ b/youtube-dl @@ -398,6 +398,10 @@ class FileDownloader(object): print info_dict['title'].encode(preferredencoding(), 'xmlcharrefreplace') if self.params.get('forceurl', False): print info_dict['url'].encode(preferredencoding(), 'xmlcharrefreplace') + if self.params.get('forcethumbnail', False) and 'thumbnail' in info_dict: + print info_dict['thumbnail'].encode(preferredencoding(), 'xmlcharrefreplace') + if self.params.get('forcedescription', False) and 'description' in info_dict: + print info_dict['description'].encode(preferredencoding(), 'xmlcharrefreplace') return @@ -599,6 +603,14 @@ class InfoExtractor(object): ext: Video filename extension. format: Video format. + The following fields are optional. Their primary purpose is to allow + youtube-dl to serve as the backend for a video search function, such + as the one in youtube2mp3. They are only used when their respective + forced printing functions are called: + + thumbnail: Full URL to a video thumbnail image. + description: One-line video description. + Subclasses of this one should re-define the _real_initialize() and _real_extract() methods, as well as the suitable() static method. Probably, they should also be instantiated and added to the main @@ -842,6 +854,28 @@ class YoutubeIE(InfoExtractor): simple_title = re.sub(ur'(?u)([^%s]+)' % simple_title_chars, ur'_', video_title) simple_title = simple_title.strip(ur'_') + # thumbnail image + if 'thumbnail_url' not in video_info: + self._downloader.trouble(u'WARNING: unable to extract video thumbnail') + video_thumbnail = '' + else: # don't panic if we can't find it + video_thumbnail = urllib.unquote_plus(video_info['thumbnail_url'][0]) + + # get video description + video_description = 'No description available.' # we need something to pass to self._downloader + # this requires an additional HTTP request and a little + # more time, so don't do it unless absolutely necessary + if self._downloader.params.get('forcedescription', False): + video_page_url = 'http://www.youtube.com/watch?v=' + video_id + request = urllib2.Request(video_page_url, None, std_headers) + try: + video_page_webpage = urllib2.urlopen(request).read() + mobj = re.search(r'', video_page_webpage) + if mobj is not None: + video_description = mobj.group(1) + except (urllib2.URLError, httplib.HTTPException, socket.error), err: + pass # don't panic if we can't find it + try: # Process video information self._downloader.process_info({ @@ -852,6 +886,8 @@ class YoutubeIE(InfoExtractor): 'stitle': simple_title, 'ext': video_extension.decode('utf-8'), 'format': (format_param is None and u'NA' or format_param.decode('utf-8')), + 'thumbnail': video_thumbnail.decode('utf-8'), + 'description': video_description.decode('utf-8'), }) if all_formats: @@ -1080,6 +1116,32 @@ class GoogleIE(InfoExtractor): video_title = sanitize_title(video_title) simple_title = re.sub(ur'(?u)([^%s]+)' % simple_title_chars, ur'_', video_title) + # Extract video description + mobj = re.search(r'([^<]*)', webpage) + if mobj is None: + self._downloader.trouble(u'ERROR: unable to extract video description') + return + video_description = mobj.group(1).decode('utf-8') + if not video_description: + video_description = 'No description available.' + + # Extract video thumbnail + if self._downloader.params.get('forcethumbnail', False): + request = urllib2.Request('http://video.google.com/videosearch?q=%s+site:video.google.com&hl=en' % abs(int(video_id))) + try: + webpage = urllib2.urlopen(request).read() + except (urllib2.URLError, httplib.HTTPException, socket.error), err: + self._downloader.trouble(u'ERROR: Unable to retrieve video webpage: %s' % str(err)) + return + mobj = re.search(r'', webpage) + if mobj is None: + self._downloader.trouble(u'ERROR: unable to extract video thumbnail') + return + video_thumbnail = mobj.group(1) + else: # we need something to pass to process_info + video_thumbnail = '' + + try: # Process video information self._downloader.process_info({ @@ -1258,6 +1320,21 @@ class YahooIE(InfoExtractor): return video_uploader = mobj.group(1).decode('utf-8') + # Extract video thumbnail + mobj = re.search(r'', webpage) + if mobj is None: + self._downloader.trouble(u'ERROR: unable to extract video thumbnail') + return + video_thumbnail = mobj.group(1).decode('utf-8') + + # Extract video description + mobj = re.search(r'', webpage) + if mobj is None: + self._downloader.trouble(u'ERROR: unable to extract video description') + return + video_description = mobj.group(1).decode('utf-8') + if not video_description: video_description = 'No description available.' + # Extract video height and width mobj = re.search(r'', webpage) if mobj is None: @@ -1303,6 +1380,10 @@ class YahooIE(InfoExtractor): 'title': video_title, 'stitle': simple_title, 'ext': video_extension.decode('utf-8'), + 'thumbnail': video_thumbnail.decode('utf-8'), + 'description': video_description, + 'thumbnail': video_thumbnail, + 'description': video_description, }) except UnavailableFormatError: self._downloader.trouble(u'ERROR: format not available for video') @@ -1494,6 +1575,188 @@ class YoutubeSearchIE(InfoExtractor): pagenum = pagenum + 1 +class GoogleSearchIE(InfoExtractor): + """Information Extractor for Google Video search queries.""" + _VALID_QUERY = r'gvsearch(\d+|all)?:[\s\S]+' + _TEMPLATE_URL = 'http://video.google.com/videosearch?q=%s+site:video.google.com&start=%s&hl=en' + _VIDEO_INDICATOR = r'videoplay\?docid=([^\&>]+)\&' + _MORE_PAGES_INDICATOR = r'Next' + _google_ie = None + _max_google_results = 1000 + + def __init__(self, google_ie, downloader=None): + InfoExtractor.__init__(self, downloader) + self._google_ie = google_ie + + @staticmethod + def suitable(url): + return (re.match(GoogleSearchIE._VALID_QUERY, url) is not None) + + def report_download_page(self, query, pagenum): + """Report attempt to download playlist page with given number.""" + query = query.decode(preferredencoding()) + self._downloader.to_stdout(u'[video.google] query "%s": Downloading page %s' % (query, pagenum)) + + def _real_initialize(self): + self._google_ie.initialize() + + def _real_extract(self, query): + mobj = re.match(self._VALID_QUERY, query) + if mobj is None: + self._downloader.trouble(u'ERROR: invalid search query "%s"' % query) + return + + prefix, query = query.split(':') + prefix = prefix[8:] + query = query.encode('utf-8') + if prefix == '': + self._download_n_results(query, 1) + return + elif prefix == 'all': + self._download_n_results(query, self._max_google_results) + return + else: + try: + n = long(prefix) + if n <= 0: + self._downloader.trouble(u'ERROR: invalid download number %s for query "%s"' % (n, query)) + return + elif n > self._max_google_results: + self._downloader.to_stderr(u'WARNING: gvsearch returns max %i results (you requested %i)' % (self._max_google_results, n)) + n = self._max_google_results + self._download_n_results(query, n) + return + except ValueError: # parsing prefix as integer fails + self._download_n_results(query, 1) + return + + def _download_n_results(self, query, n): + """Downloads a specified number of results for a query""" + + video_ids = [] + already_seen = set() + pagenum = 1 + + while True: + self.report_download_page(query, pagenum) + result_url = self._TEMPLATE_URL % (urllib.quote_plus(query), pagenum) + request = urllib2.Request(result_url, None, std_headers) + try: + page = urllib2.urlopen(request).read() + except (urllib2.URLError, httplib.HTTPException, socket.error), err: + self._downloader.trouble(u'ERROR: unable to download webpage: %s' % str(err)) + return + + # Extract video identifiers + for mobj in re.finditer(self._VIDEO_INDICATOR, page): + video_id = mobj.group(1) + if video_id not in already_seen: + video_ids.append(video_id) + already_seen.add(video_id) + if len(video_ids) == n: + # Specified n videos reached + for id in video_ids: + self._google_ie.extract('http://video.google.com/videoplay?docid=%s' % id) + return + + if re.search(self._MORE_PAGES_INDICATOR, page) is None: + for id in video_ids: + self._google_ie.extract('http://video.google.com/videoplay?docid=%s' % id) + return + + pagenum = pagenum + 1 + +class YahooSearchIE(InfoExtractor): + """Information Extractor for Yahoo! Video search queries.""" + _VALID_QUERY = r'yvsearch(\d+|all)?:[\s\S]+' + _TEMPLATE_URL = 'http://video.yahoo.com/search/?p=%s&o=%s' + _VIDEO_INDICATOR = r'href="http://video\.yahoo\.com/watch/([0-9]+/[0-9]+)"' + _MORE_PAGES_INDICATOR = r'\s*Next' + _yahoo_ie = None + _max_yahoo_results = 1000 + + def __init__(self, yahoo_ie, downloader=None): + InfoExtractor.__init__(self, downloader) + self._yahoo_ie = yahoo_ie + + @staticmethod + def suitable(url): + return (re.match(YahooSearchIE._VALID_QUERY, url) is not None) + + def report_download_page(self, query, pagenum): + """Report attempt to download playlist page with given number.""" + query = query.decode(preferredencoding()) + self._downloader.to_stdout(u'[video.yahoo] query "%s": Downloading page %s' % (query, pagenum)) + + def _real_initialize(self): + self._yahoo_ie.initialize() + + def _real_extract(self, query): + mobj = re.match(self._VALID_QUERY, query) + if mobj is None: + self._downloader.trouble(u'ERROR: invalid search query "%s"' % query) + return + + prefix, query = query.split(':') + prefix = prefix[8:] + query = query.encode('utf-8') + if prefix == '': + self._download_n_results(query, 1) + return + elif prefix == 'all': + self._download_n_results(query, self._max_yahoo_results) + return + else: + try: + n = long(prefix) + if n <= 0: + self._downloader.trouble(u'ERROR: invalid download number %s for query "%s"' % (n, query)) + return + elif n > self._max_yahoo_results: + self._downloader.to_stderr(u'WARNING: yvsearch returns max %i results (you requested %i)' % (self._max_yahoo_results, n)) + n = self._max_yahoo_results + self._download_n_results(query, n) + return + except ValueError: # parsing prefix as integer fails + self._download_n_results(query, 1) + return + + def _download_n_results(self, query, n): + """Downloads a specified number of results for a query""" + + video_ids = [] + already_seen = set() + pagenum = 1 + + while True: + self.report_download_page(query, pagenum) + result_url = self._TEMPLATE_URL % (urllib.quote_plus(query), pagenum) + request = urllib2.Request(result_url, None, std_headers) + try: + page = urllib2.urlopen(request).read() + except (urllib2.URLError, httplib.HTTPException, socket.error), err: + self._downloader.trouble(u'ERROR: unable to download webpage: %s' % str(err)) + return + + # Extract video identifiers + for mobj in re.finditer(self._VIDEO_INDICATOR, page): + video_id = mobj.group(1) + if video_id not in already_seen: + video_ids.append(video_id) + already_seen.add(video_id) + if len(video_ids) == n: + # Specified n videos reached + for id in video_ids: + self._yahoo_ie.extract('http://video.yahoo.com/watch/%s' % id) + return + + if re.search(self._MORE_PAGES_INDICATOR, page) is None: + for id in video_ids: + self._yahoo_ie.extract('http://video.yahoo.com/watch/%s' % id) + return + + pagenum = pagenum + 1 + class YoutubePlaylistIE(InfoExtractor): """Information Extractor for YouTube playlists.""" @@ -1732,6 +1995,10 @@ if __name__ == '__main__': action='store_true', dest='geturl', help='simulate, quiet but print URL', default=False) verbosity.add_option('-e', '--get-title', action='store_true', dest='gettitle', help='simulate, quiet but print title', default=False) + verbosity.add_option('--get-thumbnail', + action='store_true', dest='getthumbnail', help='simulate, quiet but print thumbnail URL', default=False) + verbosity.add_option('--get-description', + action='store_true', dest='getdescription', help='simulate, quiet but print video description', default=False) verbosity.add_option('--no-progress', action='store_true', dest='noprogress', help='do not print progress bar', default=False) parser.add_option_group(verbosity) @@ -1788,8 +2055,10 @@ if __name__ == '__main__': youtube_user_ie = YoutubeUserIE(youtube_ie) youtube_search_ie = YoutubeSearchIE(youtube_ie) google_ie = GoogleIE() + google_search_ie = GoogleSearchIE(google_ie) photobucket_ie = PhotobucketIE() yahoo_ie = YahooIE() + yahoo_search_ie = YahooSearchIE(yahoo_ie) generic_ie = GenericIE() # File downloader @@ -1797,10 +2066,12 @@ if __name__ == '__main__': 'usenetrc': opts.usenetrc, 'username': opts.username, 'password': opts.password, - 'quiet': (opts.quiet or opts.geturl or opts.gettitle), + 'quiet': (opts.quiet or opts.geturl or opts.gettitle or opts.getthumbnail or opts.getdescription), 'forceurl': opts.geturl, 'forcetitle': opts.gettitle, - 'simulate': (opts.simulate or opts.geturl or opts.gettitle), + 'forcethumbnail': opts.getthumbnail, + 'forcedescription': opts.getdescription, + 'simulate': (opts.simulate or opts.geturl or opts.gettitle or opts.getthumbnail or opts.getdescription), 'format': opts.format, 'outtmpl': ((opts.outtmpl is not None and opts.outtmpl.decode(preferredencoding())) or (opts.format == '-1' and opts.usetitle and u'%(stitle)s-%(id)s-%(format)s.%(ext)s') @@ -1821,8 +2092,10 @@ if __name__ == '__main__': fd.add_info_extractor(metacafe_ie) fd.add_info_extractor(youtube_ie) fd.add_info_extractor(google_ie) + fd.add_info_extractor(google_search_ie) fd.add_info_extractor(photobucket_ie) fd.add_info_extractor(yahoo_ie) + fd.add_info_extractor(yahoo_search_ie) # This must come last since it's the # fallback if none of the others work