youtube-dl

commit 640788f6f476f4344c13ed7c369c73575d3e705c
parent 1ae0f0a21dec0cfce85b6f8285cc64cd702846b5
Author: Sergey M․ <dstftw@gmail.com>
Date:   Wed, 27 Dec 2017 23:26:48 +0700

[internazionale] Improve extraction (closes #14973)

Diffstat:
M
youtube_dl/extractor/internazionale.py
 | 
56
+++++++++++++++++++++++++++++++++++++-------------------

1 file changed, 37 insertions(+), 19 deletions(-)
diff --git a/youtube_dl/extractor/internazionale.py b/youtube_dl/extractor/internazionale.py
@@ -2,45 +2,63 @@
 from __future__ import unicode_literals
 
 from .common import InfoExtractor
+from ..utils import unified_timestamp
 
 
 class InternazionaleIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?internazionale\.it/video/(?P<id>.*)'
-    _TESTS = [{
+    _VALID_URL = r'https?://(?:www\.)?internazionale\.it/video/(?:[^/]+/)*(?P<id>[^/?#&]+)'
+    _TEST = {
         'url': 'https://www.internazionale.it/video/2015/02/19/richard-linklater-racconta-una-scena-di-boyhood',
-        'md5': '11b54a3d3333e455c00684e50a65c58e',
+        'md5': '3e39d32b66882c1218e305acbf8348ca',
         'info_dict': {
             'id': '265968',
+            'display_id': 'richard-linklater-racconta-una-scena-di-boyhood',
             'ext': 'mp4',
-            'description': 'md5:efb7e5bbfb1a54ae2ed5a4a015f0e665',
             'title': 'Richard Linklater racconta una scena di Boyhood',
+            'description': 'md5:efb7e5bbfb1a54ae2ed5a4a015f0e665',
+            'timestamp': 1424354635,
+            'upload_date': '20150219',
             'thumbnail': r're:^https?://.*\.jpg$',
-        }
-    }]
+        },
+        'params': {
+            'format': 'bestvideo',
+        },
+    }
 
     def _real_extract(self, url):
-        video_id = self._match_id(url)
-        webpage = self._download_webpage(url, video_id)
+        display_id = self._match_id(url)
 
-        data_job_id = self._html_search_regex(r'data-job-id="([^"]+)"', webpage, 'data-job-id')
-        data_video_path = self._html_search_regex(r'data-video-path="([^"]+)"', webpage, 'data-video-path')
+        webpage = self._download_webpage(url, display_id)
 
-        formats = []
+        DATA_RE = r'data-%s=(["\'])(?P<value>(?:(?!\1).)+)\1'
 
-        formats.extend(self._extract_m3u8_formats(
-            'https://video.internazionale.it/%s/%s.m3u8' % (data_video_path, data_job_id),
-            video_id))
+        title = self._search_regex(
+            DATA_RE % 'video-title', webpage, 'title', default=None,
+            group='value') or self._og_search_title(webpage)
 
-        formats.extend(self._extract_mpd_formats(
-            'https://video.internazionale.it/%s/%s.mpd' % (data_video_path, data_job_id),
-            video_id))
+        video_id = self._search_regex(
+            DATA_RE % 'job-id', webpage, 'video id', group='value')
+        video_path = self._search_regex(
+            DATA_RE % 'video-path', webpage, 'video path', group='value')
 
+        video_base = 'https://video.internazionale.it/%s/%s.' % (video_path, video_id)
+
+        formats = self._extract_m3u8_formats(
+            video_base + 'm3u8', display_id, 'mp4',
+            entry_protocol='m3u8_native', m3u8_id='hls', fatal=False)
+        formats.extend(self._extract_mpd_formats(
+            video_base + 'mpd', display_id, mpd_id='dash', fatal=False))
         self._sort_formats(formats)
 
+        timestamp = unified_timestamp(self._html_search_meta(
+            'article:published_time', webpage, 'timestamp'))
+
         return {
-            'id': data_job_id,
-            'title': self._og_search_title(webpage),
+            'id': video_id,
+            'display_id': display_id,
+            'title': title,
             'thumbnail': self._og_search_thumbnail(webpage),
             'description': self._og_search_description(webpage),
+            'timestamp': timestamp,
             'formats': formats,
         }