youtube-dl

commit decf86044d17a8ec04e43a4805a0092622d976ae
parent 94b817edebb63c3d8485e1ae27cc394dd9e21f9d
Author: Sergey M․ <dstftw@gmail.com>
Date:   Sun, 16 Jul 2017 03:06:04 +0700

[pearvideo] Improve (closes #13031)

Diffstat:
M
youtube_dl/extractor/extractors.py
 | 
2
+-
D
youtube_dl/extractor/pear.py
 | 
34
----------------------------------
A
youtube_dl/extractor/pearvideo.py
 | 
63
+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

3 files changed, 64 insertions(+), 35 deletions(-)
diff --git a/youtube_dl/extractor/extractors.py b/youtube_dl/extractor/extractors.py
@@ -762,7 +762,7 @@ from .pandoratv import PandoraTVIE
 from .parliamentliveuk import ParliamentLiveUKIE
 from .patreon import PatreonIE
 from .pbs import PBSIE
-from .pear import PearIE
+from .pearvideo import PearVideoIE
 from .people import PeopleIE
 from .periscope import (
     PeriscopeIE,
diff --git a/youtube_dl/extractor/pear.py b/youtube_dl/extractor/pear.py
@@ -1,34 +0,0 @@
-# coding: utf-8
-from __future__ import unicode_literals
-
-from .common import InfoExtractor
-
-
-class PearIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?pearvideo\.com/video_(?P<id>[0-9]+)'
-    _TEST = {
-        'url': 'http://www.pearvideo.com/video_1076290',
-        'info_dict': {
-            'id': '1076290',
-            'ext': 'mp4',
-            'title': '小浣熊在主人家玻璃上滚石头：没砸',
-            'description': '小浣熊找到一个小石头，仿佛发现了一个宝贝。它不停地用石头按在玻璃上，滚来滚去，吸引主人注意。',
-            'url': 'http://video.pearvideo.com/mp4/short/20170508/cont-1076290-10438018-hd.mp4'
-        }
-    }
-
-    def _real_extract(self, url):
-        video_id = self._match_id(url)
-        webpage = self._download_webpage(url, video_id)
-
-        title = self._html_search_regex(r'<h1[^>]+class="video-tt">(.+)</h1>', webpage, 'title', fatal=False)
-        description = self._html_search_regex(r'<div[^>]+class="summary"[^>]*>([^<]+)<', webpage, 'description', fatal=False)
-        url = self._html_search_regex(r'hdUrl="(.*?)"', webpage, 'url', fatal=False)
-
-        return {
-            'id': video_id,
-            'ext': 'mp4',
-            'title': title,
-            'description': description,
-            'url': url
-        }
diff --git a/youtube_dl/extractor/pearvideo.py b/youtube_dl/extractor/pearvideo.py
@@ -0,0 +1,63 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+import re
+
+from .common import InfoExtractor
+from ..utils import (
+    qualities,
+    unified_timestamp,
+)
+
+
+class PearVideoIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:www\.)?pearvideo\.com/video_(?P<id>\d+)'
+    _TEST = {
+        'url': 'http://www.pearvideo.com/video_1076290',
+        'info_dict': {
+            'id': '1076290',
+            'ext': 'mp4',
+            'title': '小浣熊在主人家玻璃上滚石头：没砸',
+            'description': 'md5:01d576b747de71be0ee85eb7cac25f9d',
+            'timestamp': 1494275280,
+            'upload_date': '20170508',
+        }
+    }
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+
+        webpage = self._download_webpage(url, video_id)
+
+        quality = qualities(
+            ('ldflv', 'ld', 'sdflv', 'sd', 'hdflv', 'hd', 'src'))
+
+        formats = [{
+            'url': mobj.group('url'),
+            'format_id': mobj.group('id'),
+            'quality': quality(mobj.group('id')),
+        } for mobj in re.finditer(
+            r'(?P<id>[a-zA-Z]+)Url\s*=\s*(["\'])(?P<url>(?:https?:)?//.+?)\2',
+            webpage)]
+        self._sort_formats(formats)
+
+        title = self._search_regex(
+            (r'<h1[^>]+\bclass=(["\'])video-tt\1[^>]*>(?P<value>[^<]+)',
+             r'<[^>]+\bdata-title=(["\'])(?P<value>(?:(?!\1).)+)\1'),
+            webpage, 'title', group='value')
+        description = self._search_regex(
+            (r'<div[^>]+\bclass=(["\'])summary\1[^>]*>(?P<value>[^<]+)',
+             r'<[^>]+\bdata-summary=(["\'])(?P<value>(?:(?!\1).)+)\1'),
+            webpage, 'description', default=None,
+            group='value') or self._html_search_meta('Description', webpage)
+        timestamp = unified_timestamp(self._search_regex(
+            r'<div[^>]+\bclass=["\']date["\'][^>]*>([^<]+)',
+            webpage, 'timestamp', fatal=False))
+
+        return {
+            'id': video_id,
+            'title': title,
+            'description': description,
+            'timestamp': timestamp,
+            'formats': formats,
+        }