[hbo] add support for free episode urls and new formats extraction(closes #12519)

youtube_dl/extractor/hbo.py

@@ -4,6 +4,7 @@
 import re
 
 from .common import InfoExtractor
+from ..compat import compat_str
 from ..utils import (
     xpath_text,
     xpath_element,
@@ -14,14 +15,26 @@
 
 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,
@@ -78,6 +91,17 @@ def _extract_from_id(self, video_id):
                     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':
+                    # #EXT-X-BYTERANGE is not supported by native hls downloader
+                    # and ffmpeg (#10955)
+                    # formats.extend(self._extract_m3u8_formats(
+                    #     video_url.replace('.tar', '/base_index.m3u8'),
+                    #     video_id, 'mp4', 'm3u8_native', m3u8_id='hls', fatal=False))
+                    continue
+                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({
@@ -112,10 +136,11 @@ def _extract_from_id(self, video_id):
 
 
 class HBOIE(HBOBaseIE):
+    IE_NAME = 'hbo'
     _VALID_URL = r'https?://(?:www\.)?hbo\.com/video/video\.html\?.*vid=(?P<id>[0-9]+)'
     _TEST = {
         'url': 'http://www.hbo.com/video/video.html?autoplay=true&g=u&vid=1437839',
-        'md5': '1c33253f0c7782142c993c0ba62a8753',
+        'md5': '2c6a6bc1222c7e91cb3334dad1746e5a',
         'info_dict': {
             'id': '1437839',
             'ext': 'mp4',
@@ -131,11 +156,12 @@ def _real_extract(self, url):
 
 
 class HBOEpisodeIE(HBOBaseIE):
-    _VALID_URL = r'https?://(?:www\.)?hbo\.com/(?!video)([^/]+/)+video/(?P<id>[0-9a-z-]+)\.html'
+    IE_NAME = 'hbo:episode'
+    _VALID_URL = r'https?://(?:www\.)?hbo\.com/(?P<path>(?!video)(?:(?:[^/]+/)+video|watch-free-episodes)/(?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',
+        'md5': '61ead79b9c0dfa8d3d4b07ef4ac556fb',
         'info_dict': {
             'id': '1439518',
             'display_id': 'ep-52-inside-the-episode',
@@ -147,16 +173,19 @@ class HBOEpisodeIE(HBOBaseIE):
     }, {
         'url': 'http://www.hbo.com/game-of-thrones/about/video/season-5-invitation-to-the-set.html?autoplay=true',
         'only_matching': True,
+    }, {
+        'url': 'http://www.hbo.com/watch-free-episodes/last-week-tonight-with-john-oliver',
+        'only_matching': True,
     }]
 
     def _real_extract(self, url):
-        display_id = self._match_id(url)
+        path, display_id = re.match(self._VALID_URL, url).groups()
 
-        webpage = self._download_webpage(url, display_id)
+        content = self._download_json(
+            'http://www.hbo.com/api/content/' + path, display_id)['content']
 
-        video_id = self._search_regex(
+        video_id = compat_str((content.get('parsed', {}).get(
-            r'(?P<q1>[\'"])videoId(?P=q1)\s*:\s*(?P<q2>[\'"])(?P<video_id>\d+)(?P=q2)',
+            'common:FullBleedVideo', {}) or content['selectedEpisode'])['videoId'])
-            webpage, 'video ID', group='video_id')
 
         info_dict = self._extract_from_id(video_id)
         info_dict['display_id'] = display_id