diff --git a/youtube_dl/extractor/ximalaya.py b/youtube_dl/extractor/ximalaya.py index d79537d86..7fc8e47cc 100644 --- a/youtube_dl/extractor/ximalaya.py +++ b/youtube_dl/extractor/ximalaya.py @@ -14,19 +14,34 @@ class XimalayaBaseIE(InfoExtractor): class XimalayaIE(XimalayaBaseIE): IE_NAME = 'ximalaya' - IE_DESC = 'ximalaya.com' - _VALID_URL = r'https?://(?:www\.|m\.)?ximalaya\.com/(?P[0-9]+)/sound/(?P[0-9]+)/?' - _USER_URL_FORMAT = 'http://www.ximalaya.com/zhubo/%i/' + IE_DESC = '喜马拉雅FM' + _VALID_URL = r'https?://(?:www\.|m\.)?ximalaya\.com/(?P[0-9]+)/sound/(?P[0-9]+)' + _USER_URL_FORMAT = '%s://www.ximalaya.com/zhubo/%i/' _TESTS = [ { 'url': 'http://www.ximalaya.com/61425525/sound/47740352/', 'info_dict': { 'id': '47740352', 'ext': 'm4a', - 'title': '261.唐诗三百首.卷八.送孟浩然之广陵.李白', - 'description': 'contains:孤帆远影碧空尽，惟见长江天际流。', 'uploader': '小彬彬爱听书', 'uploader_id': 61425525, + 'uploader_url': 'http://www.ximalaya.com/zhubo/61425525/', + 'title': '261.唐诗三百首.卷八.送孟浩然之广陵.李白', + 'description': "contains:《送孟浩然之广陵》\n作者：李白\n故人西辞黄鹤楼，烟花三月下扬州。\n孤帆远影碧空尽，惟见长江天际流。", + 'thumbnails': [ + { + 'name': 'cover_url', + 'url': 'http://fdfs.xmcdn.com/group18/M03/7F/61/wKgJJVgCaNzAIs-KAAC415nhLZs762.jpg', + }, + { + 'name': 'cover_url_142', + 'url': 'http://fdfs.xmcdn.com/group18/M03/7F/61/wKgJJVgCaNzAIs-KAAC415nhLZs762_web_large.jpg', + 'width': 180, + 'height': 180 + } + ], + 'categories': ['renwen', '人文'], + 'duration': 93, 'view_count': int, 'like_count': int, } @@ -36,10 +51,56 @@ class XimalayaIE(XimalayaBaseIE): 'info_dict': { 'id': '47740352', 'ext': 'm4a', - 'title': '261.唐诗三百首.卷八.送孟浩然之广陵.李白', - 'description': 'contains:孤帆远影碧空尽，惟见长江天际流。', 'uploader': '小彬彬爱听书', 'uploader_id': 61425525, + 'uploader_url': 'http://www.ximalaya.com/zhubo/61425525/', + 'title': '261.唐诗三百首.卷八.送孟浩然之广陵.李白', + 'description': 'contains:孤帆远影碧空尽，惟见长江天际流。', + 'thumbnails': [ + { + 'name': 'cover_url', + 'url': 'http://fdfs.xmcdn.com/group18/M03/7F/61/wKgJJVgCaNzAIs-KAAC415nhLZs762.jpg', + }, + { + 'name': 'cover_url_142', + 'url': 'http://fdfs.xmcdn.com/group18/M03/7F/61/wKgJJVgCaNzAIs-KAAC415nhLZs762_web_large.jpg', + 'width': 180, + 'height': 180 + } + ], + 'categories': ['renwen', '人文'], + 'duration': 93, + 'view_count': int, + 'like_count': int, + } + }, + { + 'url': 'https://www.ximalaya.com/11045267/sound/15705996/', + 'info_dict': { + 'id': '15705996', + 'ext': 'm4a', + 'uploader': '李延隆老师', + 'uploader_id': 11045267, + 'uploader_url': 'https://www.ximalaya.com/zhubo/11045267/', + 'title': 'Lesson 1 Excuse me!', + 'description': "contains:Listen to the tape then answer\xa0this question. Whose handbag is it?\n" + "听录音，然后回答问题，这是谁的手袋？", + 'thumbnails': [ + { + 'name': 'cover_url', + 'url': 'http://fdfs.xmcdn.com/group11/M07/5A/F9/wKgDa1c2gfyzJFVqAAFKjzSOBug106.jpg', + }, + { + 'name': 'cover_url_142', + 'url': 'http://fdfs.xmcdn.com/group11/M07/5A/F9/wKgDa1c2gfyzJFVqAAFKjzSOBug106_web_large.jpg', + 'width': 180, + 'height': 180 + } + ], + 'categories': ['train', '外语'], + 'duration': 40, + 'view_count': int, + 'like_count': int, } }, ] @@ -47,13 +108,14 @@ class XimalayaIE(XimalayaBaseIE): def _real_extract(self, url): is_m = 'm.ximalaya' in url + scheme = 'https' if url.startswith('https') else 'http' audio_id = self._match_id(url) webpage = self._download_webpage(url, audio_id, note='Download sound page for %s' % audio_id, errnote='Unable to get sound page') - audio_info_file = 'http://m.ximalaya.com/tracks/%s.json' % audio_id + audio_info_file = '%s://m.ximalaya.com/tracks/%s.json' % (scheme, audio_id) audio_info = self._download_json(audio_info_file, audio_id, 'Downloading info json %s' % audio_info_file, 'Unable to download info file') @@ -64,23 +126,29 @@ class XimalayaIE(XimalayaBaseIE): formats.append({ 'format_id': bps, 'url': audio_info[k], - 'ext': 'm4a', }) - # cover pics kyes like: cover_url', 'cover_url_142' - thumbnails = [{'name': k, 'url': audio_info.get(k)} for k in audio_info.keys() if k.startswith('cover_url')] + thumbnails = [] + for k in audio_info.keys(): + # cover pics kyes like: cover_url', 'cover_url_142' + if k.startswith('cover_url'): + thumbnail = {'name': k, 'url': audio_info[k]} + if k == 'cover_url_142': + thumbnail['width'] = 180 + thumbnail['height'] = 180 + thumbnails.append(thumbnail) audio_uploader_id = audio_info.get('uid') if is_m: - intro = re.search(r'(?s)]+>(.+)', webpage) + audio_description= self._html_search_regex(r'(?s)]+>(.+?)', + webpage, 'audio_description', fatal=False, group=1) else: - intro = re.search(r'(?s)]*>(.+?)', webpage) + audio_description= self._html_search_regex(r'(?s)]*>(.+?)', + webpage, 'audio_description', fatal=False, group=1) - if intro: - audio_description = intro.group(1).strip() - else: - audio_description_file = 'http://www.ximalaya.com/sounds/%s/rich_intro' % audio_id + if not audio_description: + audio_description_file = '%s://www.ximalaya.com/sounds/%s/rich_intro' % (scheme, audio_id) audio_description = self._download_webpage(audio_description_file, audio_id, note='Downloading description file %s' % audio_description_file, errnote='Unable to download descrip file, try to parse web page', @@ -91,11 +159,11 @@ class XimalayaIE(XimalayaBaseIE): 'id': audio_id, 'uploader': audio_info.get('nickname'), 'uploader_id': audio_uploader_id, - 'uploader_url': self._USER_URL_FORMAT % audio_uploader_id, + 'uploader_url': self._USER_URL_FORMAT % (scheme, audio_uploader_id), 'title': audio_info['title'], 'thumbnails': thumbnails, 'description': audio_description, - 'categories': audio_info.get('category_title'), + 'categories': list((audio_info.get('category_name'), audio_info.get('category_title'))), 'duration': audio_info.get('duration'), 'view_count': audio_info.get('play_count'), 'like_count': audio_info.get('favorites_count'), @@ -104,11 +172,11 @@ class XimalayaIE(XimalayaBaseIE): class XimalayaAlbumIE(XimalayaBaseIE): - IE_NAME = 'ximalaya.com:album' - IE_DESC = 'ximalaya album' - _VALID_URL = r'https?://(?:www\.|m\.)?ximalaya\.com/(?P[0-9]+)/album/(?P[0-9]+)/?' - _TEMPLATE_URL = 'http://www.ximalaya.com/%s/album/%s/' - _BASE_URL_TEMPL = 'http://www.ximalaya.com%s' + IE_NAME = 'ximalaya:album' + IE_DESC = '喜马拉雅FM 专辑' + _VALID_URL = r'https?://(?:www\.|m\.)?ximalaya\.com/(?P[0-9]+)/album/(?P[0-9]+)' + _TEMPLATE_URL = '%s://www.ximalaya.com/%s/album/%s/' + _BASE_URL_TEMPL = '%s://www.ximalaya.com%s' _LIST_VIDEO_RE = r']+?href="(?P/%s/sound/(?P\d+)/?)"[^>]+?title="(?P[^>]+)">' _TESTS = [{ 'url': 'http://www.ximalaya.com/61425525/album/5534601/', @@ -128,15 +196,18 @@ class XimalayaAlbumIE(XimalayaBaseIE): ] def _real_extract(self, url): + self.scheme = scheme = 'https' if url.startswith('https') else 'http' + mobj = re.match(self._VALID_URL, url) uid, playlist_id = mobj.group('uid'), mobj.group('id') - webpage = self._download_webpage(self._TEMPLATE_URL % (uid, playlist_id), playlist_id, + webpage = self._download_webpage(self._TEMPLATE_URL % (scheme, uid, playlist_id), playlist_id, note='Download album page for %s' % playlist_id, errnote='Unable to get album info') - mobj = re.search(r'detailContent_title(?:[^>]+)?><h1(?:[^>]+)?>([^<]+)</h1>', webpage) - title = mobj.group(1) if mobj else self._html_search_meta('title') + title = self._html_search_regex(r'detailContent_title(?:[^>]+)?><h1(?:[^>]+)?>([^<]+)</h1>', + webpage, 'title',fatal=False, group=1) + title = title or self._html_search_meta('title') return self.playlist_result(self._entries(webpage, playlist_id, uid), playlist_id, title) @@ -146,29 +217,20 @@ class XimalayaAlbumIE(XimalayaBaseIE): for entry in self._process_page(html, uid): yield entry - mobj = re.search(r'<a\s+href=(["\'])(?P<more>[^\'"]+)\1' - r'[^>]+rel=(["\'])next\3', html) - if not mobj: + next_url = self._search_regex(r'<a\s+href=(["\'])(?P<more>[\S]+)\1[^>]+rel=(["\'])next\3', + html, 'list_next_url', default=None, group='more') + if not next_url: break - next_url = self._BASE_URL_TEMPL % mobj.group('more') - self.report_download_webpage('%d %s' % (page_num, next_url)) - html = self._download_webpage(next_url, playlist_id) - if not html.strip(): - # Some webpages show a "Load more" button but they don't - # have more videos - break + next_full_url = self._BASE_URL_TEMPL % (self.scheme, next_url) + html = self._download_webpage(next_full_url, playlist_id) def _process_page(self, html, uid): find_from = html.index('album_soundlist') for mobj in re.finditer(self._LIST_VIDEO_RE % uid, html[find_from:]): - if 'url' in mobj.groupdict(): - yield self.url_result(self._BASE_URL_TEMPL % mobj.group('url'), - 'Ximalaya', - mobj.group('id'), - mobj.group('title')) + yield self.url_result(self._BASE_URL_TEMPL % (self.scheme, mobj.group('url')), + XimalayaIE.ie_key(), + mobj.group('id'), + mobj.group('title')) + - @classmethod - def _match_uid_an_id(cls, url): - if '_VALID_URL_RE' not in cls.__dict__: - cls._VALID_URL_RE = re.compile(cls._VALID_URL)