diff options
| -rw-r--r-- | youtube_dl/extractor/comedycentral.py | 63 | ||||
| -rw-r--r-- | youtube_dl/extractor/common.py | 17 | ||||
| -rw-r--r-- | youtube_dl/extractor/mtv.py | 47 | ||||
| -rw-r--r-- | youtube_dl/extractor/ted.py | 22 |
4 files changed, 86 insertions, 63 deletions
diff --git a/youtube_dl/extractor/comedycentral.py b/youtube_dl/extractor/comedycentral.py index bf8d711ee..69b2beece 100644 --- a/youtube_dl/extractor/comedycentral.py +++ b/youtube_dl/extractor/comedycentral.py @@ -51,12 +51,12 @@ class ComedyCentralIE(InfoExtractor): '400': 'mp4', } _video_dimensions = { - '3500': '1280x720', - '2200': '960x540', - '1700': '768x432', - '1200': '640x360', - '750': '512x288', - '400': '384x216', + '3500': (1280, 720), + '2200': (960, 540), + '1700': (768, 432), + '1200': (640, 360), + '750': (512, 288), + '400': (384, 216), } @classmethod @@ -64,11 +64,13 @@ class ComedyCentralIE(InfoExtractor): """Receives a URL and returns True if suitable for this IE.""" return re.match(cls._VALID_URL, url, re.VERBOSE) is not None - def _print_formats(self, formats): - print('Available formats:') - for x in formats: - print('%s\t:\t%s\t[%s]' %(x, self._video_extensions.get(x, 'mp4'), self._video_dimensions.get(x, '???'))) - + @staticmethod + def _transform_rtmp_url(rtmp_video_url): + m = re.match(r'^rtmpe?://.*?/(?P<finalid>gsp.comedystor/.*)$', rtmp_video_url) + if not m: + raise ExtractorError(u'Cannot transform RTMP url') + base = 'http://mtvnmobile.vo.llnwd.net/kip0/_pxn=1+_pxI0=Ripod-h264+_pxL0=undefined+_pxM0=+_pxK=18639+_pxE=mp4/44620/mtvnorigin/' + return base + m.group('finalid') def _real_extract(self, url): mobj = re.match(self._VALID_URL, url, re.VERBOSE) @@ -155,40 +157,31 @@ class ComedyCentralIE(InfoExtractor): self._downloader.report_error(u'unable to download ' + mediaId + ': No videos found') continue - if self._downloader.params.get('listformats', None): - self._print_formats([i[0] for i in turls]) - return - - # For now, just pick the highest bitrate - format,rtmp_video_url = turls[-1] - - # Get the format arg from the arg stream - req_format = self._downloader.params.get('format', None) - - # Select format if we can find one - for f,v in turls: - if f == req_format: - format, rtmp_video_url = f, v - break - - m = re.match(r'^rtmpe?://.*?/(?P<finalid>gsp.comedystor/.*)$', rtmp_video_url) - if not m: - raise ExtractorError(u'Cannot transform RTMP url') - base = 'http://mtvnmobile.vo.llnwd.net/kip0/_pxn=1+_pxI0=Ripod-h264+_pxL0=undefined+_pxM0=+_pxK=18639+_pxE=mp4/44620/mtvnorigin/' - video_url = base + m.group('finalid') + formats = [] + for format, rtmp_video_url in turls: + w, h = self._video_dimensions.get(format, (None, None)) + formats.append({ + 'url': self._transform_rtmp_url(rtmp_video_url), + 'ext': self._video_extensions.get(format, 'mp4'), + 'format_id': format, + 'height': h, + 'width': w, + }) effTitle = showId + u'-' + epTitle + u' part ' + compat_str(partNum+1) info = { 'id': shortMediaId, - 'url': video_url, + 'formats': formats, 'uploader': showId, 'upload_date': officialDate, 'title': effTitle, - 'ext': 'mp4', - 'format': format, 'thumbnail': None, 'description': compat_str(officialTitle), } + + # TODO: Remove when #980 has been merged + info.update(info['formats'][-1]) + results.append(info) return results diff --git a/youtube_dl/extractor/common.py b/youtube_dl/extractor/common.py index 77726ee24..69cdcdc1b 100644 --- a/youtube_dl/extractor/common.py +++ b/youtube_dl/extractor/common.py @@ -35,6 +35,8 @@ class InfoExtractor(object): title: Video title, unescaped. ext: Video filename extension. + Instead of url and ext, formats can also specified. + The following fields are optional: format: The video format, defaults to ext (used for --get-format) @@ -52,8 +54,19 @@ class InfoExtractor(object): view_count: How many users have watched the video on the platform. urlhandle: [internal] The urlHandle to be used to download the file, like returned by urllib.request.urlopen - - The fields should all be Unicode strings. + formats: A list of dictionaries for each format available, it must + be ordered from worst to best quality. Potential fields: + * url Mandatory. The URL of the video file + * ext Will be calculated from url if missing + * format A human-readable description of the format + ("mp4 container with h264/opus"). + Calculated from width and height if missing. + * format_id A short description of the format + ("mp4_h264_opus" or "19") + * width Width of the video, if known + * height Height of the video, if known + + Unless mentioned otherwise, the fields should be Unicode strings. Subclasses of this one should re-define the _real_initialize() and _real_extract() methods and define a _VALID_URL regexp. diff --git a/youtube_dl/extractor/mtv.py b/youtube_dl/extractor/mtv.py index 8f956571d..001a576a8 100644 --- a/youtube_dl/extractor/mtv.py +++ b/youtube_dl/extractor/mtv.py @@ -54,23 +54,26 @@ class MTVIE(InfoExtractor): def _get_thumbnail_url(self, uri, itemdoc): return 'http://mtv.mtvnimages.com/uri/' + uri - def _extract_video_url(self, metadataXml): + def _extract_video_formats(self, metadataXml): if '/error_country_block.swf' in metadataXml: raise ExtractorError(u'This video is not available from your country.', expected=True) mdoc = xml.etree.ElementTree.fromstring(metadataXml.encode('utf-8')) renditions = mdoc.findall('.//rendition') - # For now, always pick the highest quality. - rendition = renditions[-1] - - try: - _,_,ext = rendition.attrib['type'].partition('/') - format = ext + '-' + rendition.attrib['width'] + 'x' + rendition.attrib['height'] + '_' + rendition.attrib['bitrate'] - rtmp_video_url = rendition.find('./src').text - except KeyError: - raise ExtractorError('Invalid rendition field.') - video_url = self._transform_rtmp_url(rtmp_video_url) - return {'ext': ext, 'url': video_url, 'format': format} + formats = [] + for rendition in mdoc.findall('.//rendition'): + try: + _, _, ext = rendition.attrib['type'].partition('/') + rtmp_video_url = rendition.find('./src').text + formats.append({'ext': ext, + 'url': self._transform_rtmp_url(rtmp_video_url), + 'format_id': rendition.get('bitrate'), + 'width': int(rendition.get('width')), + 'height': int(rendition.get('height')), + }) + except (KeyError, TypeError): + raise ExtractorError('Invalid rendition field.') + return formats def _get_video_info(self, itemdoc): uri = itemdoc.find('guid').text @@ -81,19 +84,25 @@ class MTVIE(InfoExtractor): mediagen_url += '&acceptMethods=fms' mediagen_page = self._download_webpage(mediagen_url, video_id, u'Downloading video urls') - video_info = self._extract_video_url(mediagen_page) description_node = itemdoc.find('description') if description_node is not None: description = description_node.text else: description = None - video_info.update({'title': itemdoc.find('title').text, - 'id': video_id, - 'thumbnail': self._get_thumbnail_url(uri, itemdoc), - 'description': description, - }) - return video_info + + info = { + 'title': itemdoc.find('title').text, + 'formats': self._extract_video_formats(mediagen_page), + 'id': video_id, + 'thumbnail': self._get_thumbnail_url(uri, itemdoc), + 'description': description, + } + + # TODO: Remove when #980 has been merged + info.update(info['formats'][-1]) + + return info def _get_videos_info(self, uri): video_id = self._id_from_uri(uri) diff --git a/youtube_dl/extractor/ted.py b/youtube_dl/extractor/ted.py index 4c11f7a03..dfa1176a3 100644 --- a/youtube_dl/extractor/ted.py +++ b/youtube_dl/extractor/ted.py @@ -77,12 +77,20 @@ class TEDIE(InfoExtractor): thumbnail = self._search_regex(r'</span>[\s.]*</div>[\s.]*<img src="(.*?)"', webpage, 'thumbnail') + formats = [{ + 'ext': 'mp4', + 'url': stream['file'], + 'format': stream['id'] + } for stream in info['htmlStreams']] info = { - 'id': info['id'], - 'url': info['htmlStreams'][-1]['file'], - 'ext': 'mp4', - 'title': title, - 'thumbnail': thumbnail, - 'description': desc, - } + 'id': info['id'], + 'title': title, + 'thumbnail': thumbnail, + 'description': desc, + 'formats': formats, + } + + # TODO: Remove when #980 has been merged + info.update(info['formats'][-1]) + return info |