X-Git-Url: https://git.rapsys.eu/youtubedl/blobdiff_plain/9dc487f48b50767cf540fa36c3de2c386fd74c04..929d540d5979815dd13aa119b6fdb41454907d0c:/youtube_dl/extractor/hbo.py diff --git a/youtube_dl/extractor/hbo.py b/youtube_dl/extractor/hbo.py index dad0f39..68df748 100644 --- a/youtube_dl/extractor/hbo.py +++ b/youtube_dl/extractor/hbo.py @@ -9,29 +9,32 @@ from ..utils import ( xpath_element, int_or_none, parse_duration, + urljoin, ) -class HBOIE(InfoExtractor): - _VALID_URL = r'https?://(?:www\.)?hbo\.com/video/video\.html\?.*vid=(?P[0-9]+)' - _TEST = { - 'url': 'http://www.hbo.com/video/video.html?autoplay=true&g=u&vid=1437839', - 'md5': '1c33253f0c7782142c993c0ba62a8753', - 'info_dict': { - 'id': '1437839', - 'ext': 'mp4', - 'title': 'Ep. 64 Clip: Encryption', - } - } +class HBOBaseIE(InfoExtractor): _FORMATS_INFO = { + 'pro7': { + 'width': 1280, + 'height': 720, + }, '1920': { 'width': 1280, 'height': 720, }, + 'pro6': { + 'width': 768, + 'height': 432, + }, '640': { 'width': 768, 'height': 432, }, + 'pro5': { + 'width': 640, + 'height': 360, + }, 'highwifi': { 'width': 640, 'height': 360, @@ -50,11 +53,13 @@ class HBOIE(InfoExtractor): }, } - def _real_extract(self, url): - video_id = self._match_id(url) - video_data = self._download_xml( - 'http://render.lv3.hbo.com/data/content/global/videos/data/%s.xml' % video_id, video_id) - title = xpath_text(video_data, 'title', 'title', True) + def _extract_info(self, url, display_id): + video_data = self._download_xml(url, display_id) + video_id = xpath_text(video_data, 'id', fatal=True) + episode_title = title = xpath_text(video_data, 'title', fatal=True) + series = xpath_text(video_data, 'program') + if series: + title = '%s - %s' % (series, title) formats = [] for source in xpath_element(video_data, 'videos', 'sources', True): @@ -89,6 +94,19 @@ class HBOIE(InfoExtractor): formats.extend(self._extract_m3u8_formats( video_url.replace('.tar', '/base_index_w8.m3u8'), video_id, 'mp4', 'm3u8_native', m3u8_id='hls', fatal=False)) + elif source.tag == 'hls': + m3u8_formats = self._extract_m3u8_formats( + video_url.replace('.tar', '/base_index.m3u8'), + video_id, 'mp4', 'm3u8_native', m3u8_id='hls', fatal=False) + for f in m3u8_formats: + if f.get('vcodec') == 'none' and not f.get('tbr'): + f['tbr'] = int_or_none(self._search_regex( + r'-(\d+)k/', f['url'], 'tbr', default=None)) + formats.extend(m3u8_formats) + elif source.tag == 'dash': + formats.extend(self._extract_mpd_formats( + video_url.replace('.tar', '/manifest.mpd'), + video_id, mpd_id='dash', fatal=False)) else: format_info = self._FORMATS_INFO.get(source.tag, {}) formats.append({ @@ -97,7 +115,7 @@ class HBOIE(InfoExtractor): 'width': format_info.get('width'), 'height': format_info.get('height'), }) - self._sort_formats(formats, ('width', 'height', 'tbr', 'format_id')) + self._sort_formats(formats) thumbnails = [] card_sizes = xpath_element(video_data, 'titleCardSizes') @@ -113,10 +131,45 @@ class HBOIE(InfoExtractor): 'width': width, }) + subtitles = None + caption_url = xpath_text(video_data, 'captionUrl') + if caption_url: + subtitles = { + 'en': [{ + 'url': caption_url, + 'ext': 'ttml' + }], + } + return { 'id': video_id, 'title': title, - 'duration': parse_duration(xpath_element(video_data, 'duration/tv14')), + 'duration': parse_duration(xpath_text(video_data, 'duration/tv14')), + 'series': series, + 'episode': episode_title, 'formats': formats, 'thumbnails': thumbnails, + 'subtitles': subtitles, } + + +class HBOIE(HBOBaseIE): + IE_NAME = 'hbo' + _VALID_URL = r'https?://(?:www\.)?hbo\.com/(?:video|embed)(?:/[^/]+)*/(?P[^/?#]+)' + _TEST = { + 'url': 'https://www.hbo.com/video/game-of-thrones/seasons/season-8/videos/trailer', + 'md5': '8126210656f433c452a21367f9ad85b3', + 'info_dict': { + 'id': '22113301', + 'ext': 'mp4', + 'title': 'Game of Thrones - Trailer', + }, + 'expected_warnings': ['Unknown MIME type application/mp4 in DASH manifest'], + } + + def _real_extract(self, url): + display_id = self._match_id(url) + webpage = self._download_webpage(url, display_id) + location_path = self._parse_json(self._html_search_regex( + r'data-state="({.+?})"', webpage, 'state'), display_id)['video']['locationUrl'] + return self._extract_info(urljoin(url, location_path), display_id)