X-Git-Url: http://git.bitcoin.ninja/index.cgi?a=blobdiff_plain;f=youtube_dl%2Fextractor%2Fgoogledrive.py;h=97ff28219f1062a0b345c7af28c7ee006e64766e;hb=37d9af306a928ce2184dcb60883e98ec0dd570ae;hp=8c611fa47fb5ceed066c5b3760996a583d6040c6;hpb=984e4d487520bd2a860b31b3165416c879b28096;p=youtube-dl diff --git a/youtube_dl/extractor/googledrive.py b/youtube_dl/extractor/googledrive.py index 8c611fa47..97ff28219 100644 --- a/youtube_dl/extractor/googledrive.py +++ b/youtube_dl/extractor/googledrive.py @@ -1,106 +1,213 @@ +from __future__ import unicode_literals + +import re + from .common import InfoExtractor -from ..utils import RegexNotFoundError +from ..utils import ( + ExtractorError, + int_or_none, + lowercase_escape, + update_url_query, +) + class GoogleDriveIE(InfoExtractor): - _VALID_URL = r'(?:https?://)?(?:video\.google\.com/get_player\?.*?docid=|(?:docs|drive)\.google\.com/(?:uc\?.*?id=|file/d/))(?P.+?)(?:&|/|$)' - _TEST = { - 'url': 'https://drive.google.com/file/d/0BzpExh0WzJF0NlR5WUlxdEVsY0U/edit?pli=1', + _VALID_URL = r'https?://(?:(?:docs|drive)\.google\.com/(?:uc\?.*?id=|file/d/)|video\.google\.com/get_player\?.*?docid=)(?P[a-zA-Z0-9_-]{28,})' + _TESTS = [{ + 'url': 'https://drive.google.com/file/d/0ByeS4oOUV-49Zzh4R1J6R09zazQ/edit?pli=1', + 'md5': 'd109872761f7e7ecf353fa108c0dbe1e', 'info_dict': { - 'id': '0BzpExh0WzJF0NlR5WUlxdEVsY0U', + 'id': '0ByeS4oOUV-49Zzh4R1J6R09zazQ', 'ext': 'mp4', - 'title': '[AHSH] Fairy Tail S2 - 01 [720p].mp4', + 'title': 'Big Buck Bunny.mp4', + 'duration': 45, } + }, { + # video id is longer than 28 characters + 'url': 'https://drive.google.com/file/d/1ENcQ_jeCuj7y19s66_Ou9dRP4GKGsodiDQ/edit', + 'md5': 'c230c67252874fddd8170e3fd1a45886', + 'info_dict': { + 'id': '1ENcQ_jeCuj7y19s66_Ou9dRP4GKGsodiDQ', + 'ext': 'mp4', + 'title': 'Andreea Banica feat Smiley - Hooky Song (Official Video).mp4', + 'duration': 189, + }, + 'only_matching': True + }] + _FORMATS_EXT = { + '5': 'flv', + '6': 'flv', + '13': '3gp', + '17': '3gp', + '18': 'mp4', + '22': 'mp4', + '34': 'flv', + '35': 'flv', + '36': '3gp', + '37': 'mp4', + '38': 'mp4', + '43': 'webm', + '44': 'webm', + '45': 'webm', + '46': 'webm', + '59': 'mp4', } - _formats = { - '5': {'ext': 'flv'}, - '6': {'ext': 'flv'}, - '13': {'ext': '3gp'}, - '17': {'ext': '3gp'}, - '18': {'ext': 'mp4'}, - '22': {'ext': 'mp4'}, - '34': {'ext': 'flv'}, - '35': {'ext': 'flv'}, - '36': {'ext': '3gp'}, - '37': {'ext': 'mp4'}, - '38': {'ext': 'mp4'}, - '43': {'ext': 'webm'}, - '44': {'ext': 'webm'}, - '45': {'ext': 'webm'}, - '46': {'ext': 'webm'}, - '59': {'ext': 'mp4'} + _BASE_URL_CAPTIONS = 'https://drive.google.com/timedtext' + _CAPTIONS_ENTRY_TAG = { + 'subtitles': 'track', + 'automatic_captions': 'target', } + _caption_formats_ext = [] + _captions_xml = None + + @staticmethod + def _extract_url(webpage): + mobj = re.search( + r']+src="https?://(?:video\.google\.com/get_player\?.*?docid=|(?:docs|drive)\.google\.com/file/d/)(?P[a-zA-Z0-9_-]{28,})', + webpage) + if mobj: + return 'https://drive.google.com/file/d/%s' % mobj.group('id') + + def _download_subtitles_xml(self, video_id, subtitles_id, hl): + if self._captions_xml: + return + self._captions_xml = self._download_xml( + self._BASE_URL_CAPTIONS, video_id, query={ + 'id': video_id, + 'vid': subtitles_id, + 'hl': hl, + 'v': video_id, + 'type': 'list', + 'tlangs': '1', + 'fmts': '1', + 'vssids': '1', + }, note='Downloading subtitles XML', + errnote='Unable to download subtitles XML', fatal=False) + if self._captions_xml: + for f in self._captions_xml.findall('format'): + if f.attrib.get('fmt_code') and not f.attrib.get('default'): + self._caption_formats_ext.append(f.attrib['fmt_code']) + + def _get_captions_by_type(self, video_id, subtitles_id, caption_type, + origin_lang_code=None): + if not subtitles_id or not caption_type: + return + captions = {} + for caption_entry in self._captions_xml.findall( + self._CAPTIONS_ENTRY_TAG[caption_type]): + caption_lang_code = caption_entry.attrib.get('lang_code') + if not caption_lang_code: + continue + caption_format_data = [] + for caption_format in self._caption_formats_ext: + query = { + 'vid': subtitles_id, + 'v': video_id, + 'fmt': caption_format, + 'lang': (caption_lang_code if origin_lang_code is None + else origin_lang_code), + 'type': 'track', + 'name': '', + 'kind': '', + } + if origin_lang_code is not None: + query.update({'tlang': caption_lang_code}) + caption_format_data.append({ + 'url': update_url_query(self._BASE_URL_CAPTIONS, query), + 'ext': caption_format, + }) + captions[caption_lang_code] = caption_format_data + return captions + + def _get_subtitles(self, video_id, subtitles_id, hl): + if not subtitles_id or not hl: + return + self._download_subtitles_xml(video_id, subtitles_id, hl) + if not self._captions_xml: + return + return self._get_captions_by_type(video_id, subtitles_id, 'subtitles') + + def _get_automatic_captions(self, video_id, subtitles_id, hl): + if not subtitles_id or not hl: + return + self._download_subtitles_xml(video_id, subtitles_id, hl) + if not self._captions_xml: + return + track = self._captions_xml.find('track') + if track is None: + return + origin_lang_code = track.attrib.get('lang_code') + if not origin_lang_code: + return + return self._get_captions_by_type( + video_id, subtitles_id, 'automatic_captions', origin_lang_code) def _real_extract(self, url): video_id = self._match_id(url) webpage = self._download_webpage( - 'http://docs.google.com/file/d/'+video_id, video_id, encoding='unicode_escape' - ) - try: - title = self._html_search_regex( - r'"title","(?P.*?)"', - webpage, - 'title', - group='title' - ) - fmt_stream_map = self._html_search_regex( - r'"fmt_stream_map","(?P<fmt_stream_map>.*?)"', - webpage, - 'fmt_stream_map', - group='fmt_stream_map' - ) - fmt_list = self._html_search_regex( - r'"fmt_list","(?P<fmt_list>.*?)"', - webpage, - 'fmt_list', - group='fmt_list' - ) -# timestamp = self._html_search_regex( -# r'"timestamp","(?P<timestamp>.*?)"', -# webpage, -# 'timestamp', -# group='timestamp' -# ) - length_seconds = self._html_search_regex( - r'"length_seconds","(?P<length_seconds>.*?)"', - webpage, - 'length_seconds', - group='length_seconds' - ) - except RegexNotFoundError: - try: - reason = self._html_search_regex( - r'"reason","(?P<reason>.*?)"', - webpage, - 'reason', - group='reason' - ) - self.report_warning(reason) - return - except RegexNotFoundError: - self.report_warning('not a video') - return - - fmt_stream_map = fmt_stream_map.split(',') - fmt_list = fmt_list.split(',') + 'http://docs.google.com/file/d/%s' % video_id, video_id) + + reason = self._search_regex( + r'"reason"\s*,\s*"([^"]+)', webpage, 'reason', default=None) + if reason: + raise ExtractorError(reason) + + title = self._search_regex(r'"title"\s*,\s*"([^"]+)', webpage, 'title') + duration = int_or_none(self._search_regex( + r'"length_seconds"\s*,\s*"([^"]+)', webpage, 'length seconds', + default=None)) + fmt_stream_map = self._search_regex( + r'"fmt_stream_map"\s*,\s*"([^"]+)', webpage, + 'fmt stream map').split(',') + fmt_list = self._search_regex( + r'"fmt_list"\s*,\s*"([^"]+)', webpage, 'fmt_list').split(',') + + resolutions = {} + for fmt in fmt_list: + mobj = re.search( + r'^(?P<format_id>\d+)/(?P<width>\d+)[xX](?P<height>\d+)', fmt) + if mobj: + resolutions[mobj.group('format_id')] = ( + int(mobj.group('width')), int(mobj.group('height'))) + formats = [] - for i in range(len(fmt_stream_map)): - fmt_id, fmt_url = fmt_stream_map[i].split('|') - resolution = fmt_list[i].split('/')[1] - width, height = resolution.split('x') - formats.append({ - 'url': fmt_url, - 'format_id': fmt_id, - 'resolution': resolution, - 'width': int(width), - 'height': int(height), - 'ext': self._formats[fmt_id]['ext'] - }) + for fmt_stream in fmt_stream_map: + fmt_stream_split = fmt_stream.split('|') + if len(fmt_stream_split) < 2: + continue + format_id, format_url = fmt_stream_split[:2] + f = { + 'url': lowercase_escape(format_url), + 'format_id': format_id, + 'ext': self._FORMATS_EXT[format_id], + } + resolution = resolutions.get(format_id) + if resolution: + f.update({ + 'width': resolution[0], + 'height': resolution[1], + }) + formats.append(f) self._sort_formats(formats) + hl = self._search_regex( + r'"hl"\s*,\s*"([^"]+)', webpage, 'hl', default=None) + subtitles_id = None + ttsurl = self._search_regex( + r'"ttsurl"\s*,\s*"([^"]+)', webpage, 'ttsurl', default=None) + if ttsurl: + # the video Id for subtitles will be the last value in the ttsurl + # query string + subtitles_id = ttsurl.encode('utf-8').decode( + 'unicode_escape').split('=')[-1] + return { 'id': video_id, 'title': title, -# 'timestamp': int(timestamp), - 'duration': int(length_seconds), - 'formats': formats + 'thumbnail': self._og_search_thumbnail(webpage, default=None), + 'duration': duration, + 'formats': formats, + 'subtitles': self.extract_subtitles(video_id, subtitles_id, hl), + 'automatic_captions': self.extract_automatic_captions( + video_id, subtitles_id, hl), }