diff options
Diffstat (limited to 'youtube_dl/extractor/wistia.py')
| -rw-r--r-- | youtube_dl/extractor/wistia.py | 99 |
1 files changed, 32 insertions, 67 deletions
diff --git a/youtube_dl/extractor/wistia.py b/youtube_dl/extractor/wistia.py index 77febd2eb..0fbc888ec 100644 --- a/youtube_dl/extractor/wistia.py +++ b/youtube_dl/extractor/wistia.py @@ -13,7 +13,8 @@ from ..utils import ( class WistiaIE(InfoExtractor): _VALID_URL = r'(?:wistia:|https?://(?:fast\.)?wistia\.(?:net|com)/embed/(?:iframe|medias)/)(?P<id>[a-z0-9]{10})' - _EMBED_BASE_URL = 'http://fast.wistia.com/embed/' + _API_URL = 'http://fast.wistia.com/embed/medias/%s.json' + _IFRAME_URL = 'http://fast.wistia.net/embed/iframe/%s' _TESTS = [{ 'url': 'http://fast.wistia.net/embed/iframe/sh7fpupwlt', @@ -45,32 +46,31 @@ class WistiaIE(InfoExtractor): # https://wistia.com/support/embed-and-share/video-on-your-website @staticmethod def _extract_url(webpage): - urls = WistiaIE._extract_urls(webpage) - return urls[0] if urls else None + match = re.search( + r'<(?:meta[^>]+?content|(?:iframe|script)[^>]+?src)=["\'](?P<url>(?:https?:)?//(?:fast\.)?wistia\.(?:net|com)/embed/(?:iframe|medias)/[a-z0-9]{10})', webpage) + if match: + return unescapeHTML(match.group('url')) - @staticmethod - def _extract_urls(webpage): - urls = [] - for match in re.finditer( - r'<(?:meta[^>]+?content|(?:iframe|script)[^>]+?src)=["\'](?P<url>(?:https?:)?//(?:fast\.)?wistia\.(?:net|com)/embed/(?:iframe|medias)/[a-z0-9]{10})', webpage): - urls.append(unescapeHTML(match.group('url'))) - for match in re.finditer( - r'''(?sx) - <div[^>]+class=(["'])(?:(?!\1).)*?\bwistia_async_(?P<id>[a-z0-9]{10})\b(?:(?!\1).)*?\1 - ''', webpage): - urls.append('wistia:%s' % match.group('id')) - for match in re.finditer(r'(?:data-wistia-?id=["\']|Wistia\.embed\(["\']|id=["\']wistia_)(?P<id>[a-z0-9]{10})', webpage): - urls.append('wistia:%s' % match.group('id')) - return urls + match = re.search( + r'''(?sx) + <script[^>]+src=(["'])(?:https?:)?//fast\.wistia\.com/assets/external/E-v1\.js\1[^>]*>.*? + <div[^>]+class=(["']).*?\bwistia_async_(?P<id>[a-z0-9]{10})\b.*?\2 + ''', webpage) + if match: + return 'wistia:%s' % match.group('id') + + match = re.search(r'(?:data-wistia-?id=["\']|Wistia\.embed\(["\']|id=["\']wistia_)(?P<id>[a-z0-9]{10})', webpage) + if match: + return 'wistia:%s' % match.group('id') def _real_extract(self, url): video_id = self._match_id(url) data_json = self._download_json( - self._EMBED_BASE_URL + 'medias/%s.json' % video_id, video_id, + self._API_URL % video_id, video_id, # Some videos require this. headers={ - 'Referer': url if url.startswith('http') else self._EMBED_BASE_URL + 'iframe/' + video_id, + 'Referer': url if url.startswith('http') else self._IFRAME_URL % video_id, }) if data_json.get('error'): @@ -95,61 +95,27 @@ class WistiaIE(InfoExtractor): 'url': aurl, 'width': int_or_none(a.get('width')), 'height': int_or_none(a.get('height')), - 'filesize': int_or_none(a.get('size')), }) else: aext = a.get('ext') - display_name = a.get('display_name') - format_id = atype - if atype and atype.endswith('_video') and display_name: - format_id = '%s-%s' % (atype[:-6], display_name) - f = { - 'format_id': format_id, + is_m3u8 = a.get('container') == 'm3u8' or aext == 'm3u8' + formats.append({ + 'format_id': atype, 'url': aurl, - 'tbr': int_or_none(a.get('bitrate')) or None, + 'tbr': int_or_none(a.get('bitrate')), + 'vbr': int_or_none(a.get('opt_vbitrate')), + 'width': int_or_none(a.get('width')), + 'height': int_or_none(a.get('height')), + 'filesize': int_or_none(a.get('size')), + 'vcodec': a.get('codec'), + 'container': a.get('container'), + 'ext': 'mp4' if is_m3u8 else aext, + 'protocol': 'm3u8' if is_m3u8 else None, 'preference': 1 if atype == 'original' else None, - } - if display_name == 'Audio': - f.update({ - 'vcodec': 'none', - }) - else: - f.update({ - 'width': int_or_none(a.get('width')), - 'height': int_or_none(a.get('height')), - 'vcodec': a.get('codec'), - }) - if a.get('container') == 'm3u8' or aext == 'm3u8': - ts_f = f.copy() - ts_f.update({ - 'ext': 'ts', - 'format_id': f['format_id'].replace('hls-', 'ts-'), - 'url': f['url'].replace('.bin', '.ts'), - }) - formats.append(ts_f) - f.update({ - 'ext': 'mp4', - 'protocol': 'm3u8_native', - }) - else: - f.update({ - 'container': a.get('container'), - 'ext': aext, - 'filesize': int_or_none(a.get('size')), - }) - formats.append(f) + }) self._sort_formats(formats) - subtitles = {} - for caption in data.get('captions', []): - language = caption.get('language') - if not language: - continue - subtitles[language] = [{ - 'url': self._EMBED_BASE_URL + 'captions/' + video_id + '.vtt?language=' + language, - }] - return { 'id': video_id, 'title': title, @@ -158,5 +124,4 @@ class WistiaIE(InfoExtractor): 'thumbnails': thumbnails, 'duration': float_or_none(data.get('duration')), 'timestamp': int_or_none(data.get('createdAt')), - 'subtitles': subtitles, } |