X-Git-Url: https://git.jankratochvil.net/?a=blobdiff_plain;f=youtube-dl;h=5224611d27b097d851c228d30edcb0f8c1d3a260;hb=ceba827e9aab563ae7c7190fc236ec1aa358ee59;hp=1b381d7b75307008b3686e036417edf155d896e7;hpb=597e7b18054b7632db6f8ba316e2410ccf748023;p=youtube-dl.git diff --git a/youtube-dl b/youtube-dl index 1b381d7..5224611 100755 --- a/youtube-dl +++ b/youtube-dl @@ -490,6 +490,8 @@ class FileDownloader(object): updatetime: Use the Last-modified header to set output file timestamps. writedescription: Write the video description to a .description file writeinfojson: Write the video description to a .info.json file + writesubtitles: Write the video subtitles to a .srt file + subtitleslang: Language of the subtitles to download """ params = None @@ -681,6 +683,10 @@ class FileDownloader(object): """ Report that the description file is being written """ self.to_screen(u'[info] Writing video description to: ' + descfn) + def report_writesubtitles(self, srtfn): + """ Report that the subtitles file is being written """ + self.to_screen(u'[info] Writing video subtitles to: ' + srtfn) + def report_writeinfojson(self, infofn): """ Report that the metadata file has been written """ self.to_screen(u'[info] Video description metadata as JSON to: ' + infofn) @@ -808,6 +814,21 @@ class FileDownloader(object): except (OSError, IOError): self.trouble(u'ERROR: Cannot write description file ' + descfn) return + + if self.params.get('writesubtitles', False) and 'subtitles' in info_dict and info_dict['subtitles']: + # subtitles download errors are already managed as troubles in relevant IE + # that way it will silently go on when used with unsupporting IE + try: + srtfn = filename.rsplit('.', 1)[0] + u'.srt' + self.report_writesubtitles(srtfn) + srtfile = open(_encodeFilename(srtfn), 'wb') + try: + srtfile.write(info_dict['subtitles'].encode('utf-8')) + finally: + srtfile.close() + except (OSError, IOError): + self.trouble(u'ERROR: Cannot write subtitles file ' + descfn) + return if self.params.get('writeinfojson', False): infofn = filename + u'.info.json' @@ -1206,6 +1227,10 @@ class YoutubeIE(InfoExtractor): """Report attempt to download video info webpage.""" self._downloader.to_screen(u'[youtube] %s: Downloading video info webpage' % video_id) + def report_video_subtitles_download(self, video_id): + """Report attempt to download video info webpage.""" + self._downloader.to_screen(u'[youtube] %s: Downloading video subtitles' % video_id) + def report_information_extraction(self, video_id): """Report attempt to extract video information.""" self._downloader.to_screen(u'[youtube] %s: Extracting video information' % video_id) @@ -1218,6 +1243,23 @@ class YoutubeIE(InfoExtractor): """Indicate the download will use the RTMP protocol.""" self._downloader.to_screen(u'[youtube] RTMP download detected') + def _closed_captions_xml_to_srt(self, xml_string): + srt = '' + texts = re.findall(r'([^<]+)', xml_string, re.MULTILINE) + # TODO parse xml instead of regex + for n, (start, dur_tag, dur, caption) in enumerate(texts): + if not dur: dur = '4' + start = float(start) + end = start + float(dur) + start = "%02i:%02i:%02i,%03i" %(start/(60*60), start/60%60, start%60, start%1*1000) + end = "%02i:%02i:%02i,%03i" %(end/(60*60), end/60%60, end%60, end%1*1000) + caption = re.sub(ur'(?u)&(.+?);', htmlentity_transform, caption) + caption = re.sub(ur'(?u)&(.+?);', htmlentity_transform, caption) # double cycle, inentional + srt += str(n) + '\n' + srt += start + ' --> ' + end + '\n' + srt += caption + '\n\n' + return srt + def _print_formats(self, formats): print 'Available formats:' for x in formats: @@ -1389,6 +1431,37 @@ class YoutubeIE(InfoExtractor): vwebpage_doc = lxml.etree.parse(StringIO.StringIO(video_webpage), html_parser) video_description = u''.join(vwebpage_doc.xpath('id("eow-description")//text()')) # TODO use another parser + + # closed captions + video_subtitles = None + if self._downloader.params.get('writesubtitles', False): + self.report_video_subtitles_download(video_id) + request = urllib2.Request('http://video.google.com/timedtext?hl=en&type=list&v=%s' % video_id) + try: + srt_list = urllib2.urlopen(request).read() + except (urllib2.URLError, httplib.HTTPException, socket.error), err: + self._downloader.trouble(u'WARNING: unable to download video subtitles: %s' % str(err)) + else: + srt_lang_list = re.findall(r'lang_code="([\w\-]+)"', srt_list) + if srt_lang_list: + if self._downloader.params.get('subtitleslang', False): + srt_lang = self._downloader.params.get('subtitleslang') + elif 'en' in srt_lang_list: + srt_lang = 'en' + else: + srt_lang = srt_lang_list[0] + if not srt_lang in srt_lang_list: + self._downloader.trouble(u'WARNING: no closed captions found in the specified language') + else: + request = urllib2.Request('http://video.google.com/timedtext?hl=en&lang=%s&v=%s' % (srt_lang, video_id)) + try: + srt_xml = urllib2.urlopen(request).read() + except (urllib2.URLError, httplib.HTTPException, socket.error), err: + self._downloader.trouble(u'WARNING: unable to download video subtitles: %s' % str(err)) + else: + video_subtitles = self._closed_captions_xml_to_srt(srt_xml.decode('utf-8')) + else: + self._downloader.trouble(u'WARNING: video has no closed captions') # token video_token = urllib.unquote_plus(video_info['token'][0]) @@ -1461,6 +1534,7 @@ class YoutubeIE(InfoExtractor): 'thumbnail': video_thumbnail.decode('utf-8'), 'description': video_description, 'player_url': player_url, + 'subtitles': video_subtitles }) except UnavailableVideoError, err: self._downloader.trouble(u'\nERROR: unable to download video') @@ -2248,9 +2322,7 @@ class GenericIE(InfoExtractor): class YoutubeSearchIE(InfoExtractor): """Information Extractor for YouTube search queries.""" _VALID_URL = r'ytsearch(\d+|all)?:[\s\S]+' - _TEMPLATE_URL = 'http://www.youtube.com/results?search_query=%s&page=%s&gl=US&hl=en' - _VIDEO_INDICATOR = r'href="/watch\?v=.+?"' - _MORE_PAGES_INDICATOR = r'(?m)>\s*Next\s*' + _API_URL = 'https://gdata.youtube.com/feeds/api/videos?q=%s&start-index=%i&max-results=50&v=2&alt=jsonc' _youtube_ie = None _max_youtube_results = 1000 IE_NAME = u'youtube:search' @@ -2301,37 +2373,31 @@ class YoutubeSearchIE(InfoExtractor): """Downloads a specified number of results for a query""" video_ids = [] - already_seen = set() - pagenum = 1 + pagenum = 0 + limit = n - while True: - self.report_download_page(query, pagenum) - result_url = self._TEMPLATE_URL % (urllib.quote_plus(query), pagenum) + while (50 * pagenum) < limit: + self.report_download_page(query, pagenum+1) + result_url = self._API_URL % (urllib.quote_plus(query), (50*pagenum)+1) request = urllib2.Request(result_url) try: - page = urllib2.urlopen(request).read() + data = urllib2.urlopen(request).read() except (urllib2.URLError, httplib.HTTPException, socket.error), err: - self._downloader.trouble(u'ERROR: unable to download webpage: %s' % str(err)) + self._downloader.trouble(u'ERROR: unable to download API page: %s' % str(err)) return + api_response = json.loads(data)['data'] - # Extract video identifiers - for mobj in re.finditer(self._VIDEO_INDICATOR, page): - video_id = page[mobj.span()[0]:mobj.span()[1]].split('=')[2][:-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._youtube_ie.extract('http://www.youtube.com/watch?v=%s' % id) - return + new_ids = list(video['id'] for video in api_response['items']) + video_ids += new_ids - if re.search(self._MORE_PAGES_INDICATOR, page) is None: - for id in video_ids: - self._youtube_ie.extract('http://www.youtube.com/watch?v=%s' % id) - return + limit = min(n, api_response['totalItems']) + pagenum += 1 - pagenum = pagenum + 1 + if len(video_ids) > n: + video_ids = video_ids[:n] + for id in video_ids: + self._youtube_ie.extract('http://www.youtube.com/watch?v=%s' % id) + return class GoogleSearchIE(InfoExtractor): @@ -2515,7 +2581,7 @@ class YoutubePlaylistIE(InfoExtractor): _VALID_URL = r'(?:https?://)?(?:\w+\.)?youtube\.com/(?:(?:course|view_play_list|my_playlists|artist|playlist)\?.*?(p|a|list)=|user/.*?/user/|p/|user/.*?#[pg]/c/)(?:PL)?([0-9A-Za-z-_]+)(?:/.*?/([0-9A-Za-z_-]+))?.*' _TEMPLATE_URL = 'http://www.youtube.com/%s?%s=%s&page=%s&gl=US&hl=en' - _VIDEO_INDICATOR = r'/watch\?v=(.+?)&' + _VIDEO_INDICATOR_TEMPLATE = r'/watch\?v=(.+?)&list=PL%s&' _MORE_PAGES_INDICATOR = r'(?m)>\s*Next\s*' _youtube_ie = None IE_NAME = u'youtube:playlist' @@ -2567,7 +2633,7 @@ class YoutubePlaylistIE(InfoExtractor): # Extract video identifiers ids_in_page = [] - for mobj in re.finditer(self._VIDEO_INDICATOR, page): + for mobj in re.finditer(self._VIDEO_INDICATOR_TEMPLATE % playlist_id, page): if mobj.group(1) not in ids_in_page: ids_in_page.append(mobj.group(1)) video_ids.extend(ids_in_page) @@ -2578,7 +2644,10 @@ class YoutubePlaylistIE(InfoExtractor): playliststart = self._downloader.params.get('playliststart', 1) - 1 playlistend = self._downloader.params.get('playlistend', -1) - video_ids = video_ids[playliststart:playlistend] + if playlistend == -1: + video_ids = video_ids[playliststart:] + else: + video_ids = video_ids[playliststart:playlistend] for id in video_ids: self._youtube_ie.extract('http://www.youtube.com/watch?v=%s' % id) @@ -4324,6 +4393,12 @@ def parseOpts(): action='store', dest='format_limit', metavar='FORMAT', help='highest quality format to download') video_format.add_option('-F', '--list-formats', action='store_true', dest='listformats', help='list all available formats (currently youtube only)') + video_format.add_option('--write-srt', + action='store_true', dest='writesubtitles', + help='write video closed captions to a .srt file (currently youtube only)', default=False) + video_format.add_option('--srt-lang', + action='store', dest='subtitleslang', metavar='LANG', + help='language of the closed captions to download (optional) use IETF language tags like \'en\'') verbosity.add_option('-q', '--quiet', @@ -4588,6 +4663,8 @@ def _real_main(): 'updatetime': opts.updatetime, 'writedescription': opts.writedescription, 'writeinfojson': opts.writeinfojson, + 'writesubtitles': opts.writesubtitles, + 'subtitleslang': opts.subtitleslang, 'matchtitle': opts.matchtitle, 'rejecttitle': opts.rejecttitle, 'max_downloads': opts.max_downloads,