X-Git-Url: http://git.bitcoin.ninja/index.cgi?a=blobdiff_plain;f=youtube_dl%2Fextractor%2Fwistia.py;h=77febd2eb1b1cada3942c212739725135a36682b;hb=718393c632df5106df92c60c650f52d86a9a3510;hp=bc31c2e64f22999adf575e60d59bde3d903bb9cc;hpb=cc14dfb8ecb73be9905f5adab2d7f1f92d435e2f;p=youtube-dl diff --git a/youtube_dl/extractor/wistia.py b/youtube_dl/extractor/wistia.py index bc31c2e64..77febd2eb 100644 --- a/youtube_dl/extractor/wistia.py +++ b/youtube_dl/extractor/wistia.py @@ -1,57 +1,162 @@ -import json +from __future__ import unicode_literals + import re from .common import InfoExtractor +from ..utils import ( + ExtractorError, + int_or_none, + float_or_none, + unescapeHTML, +) class WistiaIE(InfoExtractor): - _VALID_URL = r'^https?://(?:fast\.)?wistia\.net/embed/iframe/(?P[a-z0-9]+)' - - _TEST = { - u"url": u"http://fast.wistia.net/embed/iframe/sh7fpupwlt", - u"file": u"sh7fpupwlt.mov", - u"md5": u"cafeb56ec0c53c18c97405eecb3133df", - u"info_dict": { - u"title": u"cfh_resourceful_zdkh_final_1" + _VALID_URL = r'(?:wistia:|https?://(?:fast\.)?wistia\.(?:net|com)/embed/(?:iframe|medias)/)(?P[a-z0-9]{10})' + _EMBED_BASE_URL = 'http://fast.wistia.com/embed/' + + _TESTS = [{ + 'url': 'http://fast.wistia.net/embed/iframe/sh7fpupwlt', + 'md5': 'cafeb56ec0c53c18c97405eecb3133df', + 'info_dict': { + 'id': 'sh7fpupwlt', + 'ext': 'mov', + 'title': 'Being Resourceful', + 'description': 'a Clients From Hell Video Series video from worldwidewebhosting', + 'upload_date': '20131204', + 'timestamp': 1386185018, + 'duration': 117, }, - } + }, { + 'url': 'wistia:sh7fpupwlt', + 'only_matching': True, + }, { + # with hls video + 'url': 'wistia:807fafadvk', + 'only_matching': True, + }, { + 'url': 'http://fast.wistia.com/embed/iframe/sh7fpupwlt', + 'only_matching': True, + }, { + 'url': 'http://fast.wistia.net/embed/medias/sh7fpupwlt.json', + 'only_matching': True, + }] + + # 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 + + @staticmethod + def _extract_urls(webpage): + urls = [] + for match in re.finditer( + r'<(?:meta[^>]+?content|(?:iframe|script)[^>]+?src)=["\'](?P(?: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) + ]+class=(["'])(?:(?!\1).)*?\bwistia_async_(?P[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[a-z0-9]{10})', webpage): + urls.append('wistia:%s' % match.group('id')) + return urls def _real_extract(self, url): - mobj = re.match(self._VALID_URL, url) - video_id = mobj.group('id') + video_id = self._match_id(url) - webpage = self._download_webpage(url, video_id) - data_json = self._html_search_regex( - r'Wistia.iframeInit\((.*?), {}\);', webpage, u'video data') + data_json = self._download_json( + self._EMBED_BASE_URL + 'medias/%s.json' % video_id, video_id, + # Some videos require this. + headers={ + 'Referer': url if url.startswith('http') else self._EMBED_BASE_URL + 'iframe/' + video_id, + }) + + if data_json.get('error'): + raise ExtractorError( + 'Error while getting the playlist', expected=True) - data = json.loads(data_json) + data = data_json['media'] + title = data['name'] formats = [] thumbnails = [] - for atype, a in data['assets'].items(): - if atype == 'still': - thumbnails.append({ - 'url': a['url'], - 'resolution': '%dx%d' % (a['width'], a['height']), - }) + for a in data['assets']: + aurl = a.get('url') + if not aurl: continue - if atype == 'preview': + astatus = a.get('status') + atype = a.get('type') + if (astatus is not None and astatus != 2) or atype in ('preview', 'storyboard'): continue - formats.append({ - 'format_id': atype, - 'url': a['url'], - 'width': a['width'], - 'height': a['height'], - 'filesize': a['size'], - 'ext': a['ext'], - 'preference': 1 if atype == 'original' else None, - }) + elif atype in ('still', 'still_image'): + thumbnails.append({ + '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, + 'url': aurl, + 'tbr': int_or_none(a.get('bitrate')) or 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': data['name'], + 'title': title, + 'description': data.get('seoDescription'), 'formats': formats, 'thumbnails': thumbnails, + 'duration': float_or_none(data.get('duration')), + 'timestamp': int_or_none(data.get('createdAt')), + 'subtitles': subtitles, }