Merge pull request #7045 from remitamine/ign

[youtube-dl] / youtube_dl / extractor / nba.py

diff --git a/youtube_dl/extractor/nba.py b/youtube_dl/extractor/nba.py
index 73116c7c611db13a81a530d423560ea5b43d1e8e..7c6b7841db251d3ee2178de680e60cbb6900b0fc 100644 (file)
--- a/youtube_dl/extractor/nba.py
+++ b/youtube_dl/extractor/nba.py
@@ -1,20 +1,24 @@
 from __future__ import unicode_literals
 
+import re
+
 from .common import InfoExtractor
 from ..utils import (
     parse_duration,
     int_or_none,
+    xpath_text,
+    xpath_attr,
 )
 
 
 class NBAIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:watch\.|www\.)?nba\.com/(?:nba/)?video/(?P<id>[^?]*?)/?(?:/index\.html)?(?:\?.*)?$'
+    _VALID_URL = r'https?://(?:watch\.|www\.)?nba\.com/(?P<path>(?:[^/]+/)?video/(?P<id>[^?]*?))/?(?:/index\.html)?(?:\?.*)?$'
     _TESTS = [{
         'url': 'http://www.nba.com/video/games/nets/2012/12/04/0021200253-okc-bkn-recap.nba/index.html',
-        'md5': '9d902940d2a127af3f7f9d2f3dc79c96',
+        'md5': '9e7729d3010a9c71506fd1248f74e4f4',
         'info_dict': {
             'id': '0021200253-okc-bkn-recap',
-            'ext': 'mp4',
+            'ext': 'flv',
             'title': 'Thunder vs. Nets',
             'description': 'Kevin Durant scores 32 points and dishes out six assists as the Thunder beat the Nets in Brooklyn.',
             'duration': 181,
@@ -24,8 +28,8 @@ class NBAIE(InfoExtractor):
     }, {
         'url': 'http://www.nba.com/video/games/hornets/2014/12/05/0021400276-nyk-cha-play5.nba/',
         'only_matching': True,
-    },{
-        'url': 'http://watch.nba.com/nba/video/channels/playoffs/2015/05/20/0041400301-cle-atl-recap.nba',
+    }, {
+        'url': 'http://watch.nba.com/video/channels/playoffs/2015/05/20/0041400301-cle-atl-recap.nba',
         'md5': 'b2b39b81cf28615ae0c3360a3f9668c4',
         'info_dict': {
             'id': '0041400301-cle-atl-recap',
@@ -38,72 +42,16 @@ class NBAIE(InfoExtractor):
         }
     }]
 
-    _QUALITIES = {
-        '420mp4': {
-            'width': 400,
-            'height': 224,
-            'preference': 1,
-        },
-        '416x234': {
-            'width': 416,
-            'height': 234,
-            'preference': 2,
-        },
-        '480x320_910': {
-            'width': 480,
-            'height': 320,
-            'preference': 3,
-        },
-        'nba_576x324': {
-            'width': 576,
-            'height': 324,
-            'preference': 4,
-        },
-        'nba_640x360': {
-            'width': 640,
-            'height': 360,
-            'preference': 5,
-        },
-        '640x360_664b': {
-            'width': 640,
-            'height': 360,
-            'preference': 6,
-        },
-        '640x360_664m': {
-            'width': 640,
-            'height': 360,
-            'preference': 7,
-        },
-        '768x432_996': {
-            'width': 768,
-            'height': 432,
-            'preference': 8,
-        },
-        '768x432_1404': {
-            'width': 768,
-            'height': 432,
-            'preference': 9,
-        },
-        '960x540_2104': {
-            'width': 960,
-            'height': 540,
-            'preference': 10,
-        },
-        '1280x720_3072': {
-            'width': 1280,
-            'height': 720,
-            'preference': 11,
-        },
-    }
-
     def _real_extract(self, url):
-        video_id = self._match_id(url)
-        video_info = self._download_xml('http://www.nba.com/video/%s.xml' % video_id, video_id)
-        video_id = video_info.find('slug').text
-        title = video_info.find('headline').text
-        description = video_info.find('description').text
-        duration = parse_duration(video_info.find('length').text)
-        timestamp = int_or_none(video_info.find('dateCreated').attrib.get('uts'))
+        path, video_id = re.match(self._VALID_URL, url).groups()
+        if path.startswith('nba/'):
+            path = path[3:]
+        video_info = self._download_xml('http://www.nba.com/%s.xml' % path, video_id)
+        video_id = xpath_text(video_info, 'slug')
+        title = xpath_text(video_info, 'headline')
+        description = xpath_text(video_info, 'description')
+        duration = parse_duration(xpath_text(video_info, 'length'))
+        timestamp = int_or_none(xpath_attr(video_info, 'dateCreated', 'uts'))
 
         thumbnails = []
         for image in video_info.find('images'):
@@ -115,24 +63,32 @@ class NBAIE(InfoExtractor):
             })
 
         formats = []
-        for video_file in video_info.find('files').iter('file'):
+        for video_file in video_info.findall('.//file'):
             video_url = video_file.text
             if video_url.startswith('/'):
                 continue
             if video_url.endswith('.m3u8'):
-                formats.extend(self._extract_m3u8_formats(video_url, video_id))
+                m3u8_formats = self._extract_m3u8_formats(video_url, video_id, m3u8_id='hls', fatal=False)
+                if m3u8_formats:
+                    formats.extend(m3u8_formats)
             elif video_url.endswith('.f4m'):
-                formats.extend(self._extract_f4m_formats(video_url + '?hdcore=3.4.1.1', video_id))
+                f4m_formats = self._extract_f4m_formats(video_url + '?hdcore=3.4.1.1', video_id, f4m_id='hds', fatal=False)
+                if f4m_formats:
+                    formats.extend(f4m_formats)
             else:
                 key = video_file.attrib.get('bitrate')
-                quality = self._QUALITIES[key]
-                formats.append({
+                format_info = {
                     'format_id': key,
                     'url': video_url,
-                    'width': quality['width'],
-                    'height': quality['height'],
-                    'preference': quality['preference'],
-                })
+                }
+                mobj = re.search(r'(\d+)x(\d+)(?:_(\d+))?', key)
+                if mobj:
+                    format_info.update({
+                        'width': int(mobj.group(1)),
+                        'height': int(mobj.group(2)),
+                        'tbr': int_or_none(mobj.group(3)),
+                    })
+                formats.append(format_info)
         self._sort_formats(formats)
 
         return {