X-Git-Url: https://git.rapsys.eu/youtubedl/blobdiff_plain/848723ea972c09f28787db91d8c06e98a274ab89..e2cfcb2a7ff312228a25442500a885832a1829af:/youtube_dl/extractor/ivi.py diff --git a/youtube_dl/extractor/ivi.py b/youtube_dl/extractor/ivi.py index 1ba4966..472d72b 100644 --- a/youtube_dl/extractor/ivi.py +++ b/youtube_dl/extractor/ivi.py @@ -6,15 +6,16 @@ import json from .common import InfoExtractor from ..utils import ( - compat_urllib_request, ExtractorError, + int_or_none, + sanitized_Request, ) class IviIE(InfoExtractor): IE_DESC = 'ivi.ru' IE_NAME = 'ivi' - _VALID_URL = r'https?://(?:www\.)?ivi\.ru/watch(?:/(?P[^/]+))?/(?P\d+)' + _VALID_URL = r'https?://(?:www\.)?ivi\.ru/(?:watch/(?:[^/]+/)?|video/player\?.*?videoId=)(?P\d+)' _TESTS = [ # Single movie @@ -27,98 +28,110 @@ class IviIE(InfoExtractor): 'title': 'Иван Васильевич меняет профессию', 'description': 'md5:b924063ea1677c8fe343d8a72ac2195f', 'duration': 5498, - 'thumbnail': 'http://thumbs.ivi.ru/f20.vcp.digitalaccess.ru/contents/d/1/c3c885163a082c29bceeb7b5a267a6.jpg', + 'thumbnail': 're:^https?://.*\.jpg$', }, 'skip': 'Only works from Russia', }, - # Serial's serie + # Serial's series { - 'url': 'http://www.ivi.ru/watch/dezhurnyi_angel/74791', - 'md5': '3e6cc9a848c1d2ebcc6476444967baa9', + 'url': 'http://www.ivi.ru/watch/dvoe_iz_lartsa/9549', + 'md5': '221f56b35e3ed815fde2df71032f4b3e', 'info_dict': { - 'id': '74791', + 'id': '9549', 'ext': 'mp4', - 'title': 'Дежурный ангел - 1 серия', - 'duration': 2490, - 'thumbnail': 'http://thumbs.ivi.ru/f7.vcp.digitalaccess.ru/contents/8/e/bc2f6c2b6e5d291152fdd32c059141.jpg', + 'title': 'Двое из ларца - Дело Гольдберга (1 часть)', + 'series': 'Двое из ларца', + 'season': 'Сезон 1', + 'season_number': 1, + 'episode': 'Дело Гольдберга (1 часть)', + 'episode_number': 1, + 'duration': 2655, + 'thumbnail': 're:^https?://.*\.jpg$', }, 'skip': 'Only works from Russia', - } + } ] # Sorted by quality - _known_formats = ['MP4-low-mobile', 'MP4-mobile', 'FLV-lo', 'MP4-lo', 'FLV-hi', 'MP4-hi', 'MP4-SHQ'] - - # Sorted by size - _known_thumbnails = ['Thumb-120x90', 'Thumb-160', 'Thumb-640x480'] - - def _extract_description(self, html): - m = re.search(r'', html) - return m.group('description') if m is not None else None - - def _extract_comment_count(self, html): - m = re.search('(?s)\s*Комментарии:\s*(?P\d+)\s*', html) - return int(m.group('commentcount')) if m is not None else 0 + _KNOWN_FORMATS = ['MP4-low-mobile', 'MP4-mobile', 'FLV-lo', 'MP4-lo', 'FLV-hi', 'MP4-hi', 'MP4-SHQ'] def _real_extract(self, url): - mobj = re.match(self._VALID_URL, url) - video_id = mobj.group('videoid') - - api_url = 'http://api.digitalaccess.ru/api/json/' - - data = {'method': 'da.content.get', - 'params': [video_id, {'site': 's183', - 'referrer': 'http://www.ivi.ru/watch/%s' % video_id, - 'contentid': video_id - } - ] + video_id = self._match_id(url) + + data = { + 'method': 'da.content.get', + 'params': [ + video_id, { + 'site': 's183', + 'referrer': 'http://www.ivi.ru/watch/%s' % video_id, + 'contentid': video_id } + ] + } - request = compat_urllib_request.Request(api_url, json.dumps(data)) - - video_json_page = self._download_webpage(request, video_id, 'Downloading video JSON') - video_json = json.loads(video_json_page) + request = sanitized_Request( + 'http://api.digitalaccess.ru/api/json/', json.dumps(data)) + video_json = self._download_json( + request, video_id, 'Downloading video JSON') if 'error' in video_json: error = video_json['error'] if error['origin'] == 'NoRedisValidData': raise ExtractorError('Video %s does not exist' % video_id, expected=True) - raise ExtractorError('Unable to download video %s: %s' % (video_id, error['message']), expected=True) + raise ExtractorError( + 'Unable to download video %s: %s' % (video_id, error['message']), + expected=True) result = video_json['result'] formats = [{ 'url': x['url'], 'format_id': x['content_format'], - 'preference': self._known_formats.index(x['content_format']), - } for x in result['files'] if x['content_format'] in self._known_formats] + 'preference': self._KNOWN_FORMATS.index(x['content_format']), + } for x in result['files'] if x['content_format'] in self._KNOWN_FORMATS] self._sort_formats(formats) - if not formats: - raise ExtractorError('No media links available for %s' % video_id) - - duration = result['duration'] - compilation = result['compilation'] title = result['title'] - title = '%s - %s' % (compilation, title) if compilation is not None else title + duration = int_or_none(result.get('duration')) + compilation = result.get('compilation') + episode = title if compilation else None + + title = '%s - %s' % (compilation, title) if compilation is not None else title + + thumbnails = [{ + 'url': preview['url'], + 'id': preview.get('content_format'), + } for preview in result.get('preview', []) if preview.get('url')] - previews = result['preview'] - previews.sort(key=lambda fmt: self._known_thumbnails.index(fmt['content_format'])) - thumbnail = previews[-1]['url'] if len(previews) > 0 else None + webpage = self._download_webpage(url, video_id) - video_page = self._download_webpage(url, video_id, 'Downloading video page') - description = self._extract_description(video_page) - comment_count = self._extract_comment_count(video_page) + season = self._search_regex( + r']+class="season active"[^>]*>]+>([^<]+)', + webpage, 'season', default=None) + season_number = int_or_none(self._search_regex( + r']+class="season active"[^>]*>]+data-season(?:-index)?="(\d+)"', + webpage, 'season number', default=None)) + + episode_number = int_or_none(self._search_regex( + r']+itemprop="episode"[^>]*>\s*]+itemprop="episodeNumber"[^>]+content="(\d+)', + webpage, 'episode number', default=None)) + + description = self._og_search_description(webpage, default=None) or self._html_search_meta( + 'description', webpage, 'description', default=None) return { 'id': video_id, 'title': title, - 'thumbnail': thumbnail, + 'series': compilation, + 'season': season, + 'season_number': season_number, + 'episode': episode, + 'episode_number': episode_number, + 'thumbnails': thumbnails, 'description': description, 'duration': duration, - 'comment_count': comment_count, 'formats': formats, } @@ -127,27 +140,47 @@ class IviCompilationIE(InfoExtractor): IE_DESC = 'ivi.ru compilations' IE_NAME = 'ivi:compilation' _VALID_URL = r'https?://(?:www\.)?ivi\.ru/watch/(?!\d+)(?P[a-z\d_-]+)(?:/season(?P\d+))?$' + _TESTS = [{ + 'url': 'http://www.ivi.ru/watch/dvoe_iz_lartsa', + 'info_dict': { + 'id': 'dvoe_iz_lartsa', + 'title': 'Двое из ларца (2006 - 2008)', + }, + 'playlist_mincount': 24, + }, { + 'url': 'http://www.ivi.ru/watch/dvoe_iz_lartsa/season1', + 'info_dict': { + 'id': 'dvoe_iz_lartsa/season1', + 'title': 'Двое из ларца (2006 - 2008) 1 сезон', + }, + 'playlist_mincount': 12, + }] def _extract_entries(self, html, compilation_id): - return [self.url_result('http://www.ivi.ru/watch/%s/%s' % (compilation_id, serie), 'Ivi') - for serie in re.findall(r'(?:[^<]+)' % compilation_id, html)] + return [ + self.url_result( + 'http://www.ivi.ru/watch/%s/%s' % (compilation_id, serie), IviIE.ie_key()) + for serie in re.findall( + r']+data-id="\1"' % compilation_id, html)] def _real_extract(self, url): mobj = re.match(self._VALID_URL, url) compilation_id = mobj.group('compilationid') season_id = mobj.group('seasonid') - if season_id is not None: # Season link - season_page = self._download_webpage(url, compilation_id, 'Downloading season %s web page' % season_id) + if season_id is not None: # Season link + season_page = self._download_webpage( + url, compilation_id, 'Downloading season %s web page' % season_id) playlist_id = '%s/season%s' % (compilation_id, season_id) playlist_title = self._html_search_meta('title', season_page, 'title') entries = self._extract_entries(season_page, compilation_id) - else: # Compilation link + else: # Compilation link compilation_page = self._download_webpage(url, compilation_id, 'Downloading compilation web page') playlist_id = compilation_id playlist_title = self._html_search_meta('title', compilation_page, 'title') - seasons = re.findall(r'[^<]+' % compilation_id, compilation_page) - if len(seasons) == 0: # No seasons in this compilation + seasons = re.findall( + r'