Use canonical URL in Vcs-Git.

[youtubedl] / youtube_dl / extractor / hbo.py

diff --git a/youtube_dl/extractor/hbo.py b/youtube_dl/extractor/hbo.py
index 8116ad9bd42f840bc5875070d5f40e8d904b7abb..68df748f56ae306d37f7e961606e6538de018b53 100644 (file)
--- a/youtube_dl/extractor/hbo.py
+++ b/youtube_dl/extractor/hbo.py
@@ -9,19 +9,32 @@ from ..utils import (
     xpath_element,
     int_or_none,
     parse_duration,
+    urljoin,
 )
 
 
 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,
@@ -40,10 +53,13 @@ class HBOBaseIE(InfoExtractor):
         },
     }
 
-    def _extract_from_id(self, video_id):
-        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):
@@ -78,6 +94,19 @@ class HBOBaseIE(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({
@@ -86,7 +115,7 @@ class HBOBaseIE(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')
@@ -102,63 +131,45 @@ class HBOBaseIE(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_text(video_data, 'duration/tv14')),
+            'series': series,
+            'episode': episode_title,
             'formats': formats,
             'thumbnails': thumbnails,
+            'subtitles': subtitles,
         }
 
 
 class HBOIE(HBOBaseIE):
-    _VALID_URL = r'https?://(?:www\.)?hbo\.com/video/video\.html\?.*vid=(?P<id>[0-9]+)'
+    IE_NAME = 'hbo'
+    _VALID_URL = r'https?://(?:www\.)?hbo\.com/(?:video|embed)(?:/[^/]+)*/(?P<id>[^/?#]+)'
     _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',
-            'thumbnail': r're:https?://.*\.jpg$',
-            'duration': 1072,
-        }
-    }
-
-    def _real_extract(self, url):
-        video_id = self._match_id(url)
-        return self._extract_from_id(video_id)
-
-
-class HBOEpisodeIE(HBOBaseIE):
-    _VALID_URL = r'https?://(?:www\.)?hbo\.com/(?!video)([^/]+/)+video/(?P<id>[0-9a-z-]+)\.html'
-
-    _TESTS = [{
-        'url': 'http://www.hbo.com/girls/episodes/5/52-i-love-you-baby/video/ep-52-inside-the-episode.html?autoplay=true',
-        'md5': '689132b253cc0ab7434237fc3a293210',
+        'url': 'https://www.hbo.com/video/game-of-thrones/seasons/season-8/videos/trailer',
+        'md5': '8126210656f433c452a21367f9ad85b3',
         'info_dict': {
-            'id': '1439518',
-            'display_id': 'ep-52-inside-the-episode',
+            'id': '22113301',
             'ext': 'mp4',
-            'title': 'Ep. 52: Inside the Episode',
-            'thumbnail': r're:https?://.*\.jpg$',
-            'duration': 240,
+            'title': 'Game of Thrones - Trailer',
         },
-    }, {
-        'url': 'http://www.hbo.com/game-of-thrones/about/video/season-5-invitation-to-the-set.html?autoplay=true',
-        'only_matching': True,
-    }]
+        '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)
-
-        video_id = self._search_regex(
-            r'(?P<q1>[\'"])videoId(?P=q1)\s*:\s*(?P<q2>[\'"])(?P<video_id>\d+)(?P=q2)',
-            webpage, 'video ID', group='video_id')
-
-        info_dict = self._extract_from_id(video_id)
-        info_dict['display_id'] = display_id
-
-        return info_dict
+        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)