[1tv] Fix extraction (Closes #9103)

youtube_dl/extractor/firsttv.py

@@ -2,78 +2,133 @@
 from __future__ import unicode_literals
 
 from .common import InfoExtractor
-from ..utils import int_or_none
+from ..compat import compat_xpath
+from ..utils import (
+    int_or_none,
+    qualities,
+    unified_strdate,
+    xpath_attr,
+    xpath_element,
+    xpath_text,
+    xpath_with_ns,
+)
 
 
 class FirstTVIE(InfoExtractor):
     IE_NAME = '1tv'
     IE_DESC = 'Первый канал'
-    _VALID_URL = r'https?://(?:www\.)?1tv\.ru/(?:[^/]+/)+(?P<id>.+)'
+    _VALID_URL = r'https?://(?:www\.)?1tv\.ru/(?:[^/]+/)+p?(?P<id>\d+)'
 
     _TESTS = [{
-        'url': 'http://www.1tv.ru/videoarchive/73390',
+        # single format via video_materials.json API
-        'md5': '777f525feeec4806130f4f764bc18a4f',
-        'info_dict': {
-            'id': '73390',
-            'ext': 'mp4',
-            'title': 'Олимпийские канатные дороги',
-            'description': 'md5:d41d8cd98f00b204e9800998ecf8427e',
-            'thumbnail': 're:^https?://.*\.(?:jpg|JPG)$',
-            'duration': 149,
-            'like_count': int,
-            'dislike_count': int,
-        },
-        'skip': 'Only works from Russia',
-    }, {
         'url': 'http://www.1tv.ru/prj/inprivate/vypusk/35930',
-        'md5': 'a1b6b60d530ebcf8daacf4565762bbaf',
+        'md5': '82a2777648acae812d58b3f5bd42882b',
         'info_dict': {
             'id': '35930',
             'ext': 'mp4',
-            'title': 'Наедине со всеми. Людмила Сенчина',
+            'title': 'Гость Людмила Сенчина. Наедине со всеми. Выпуск от 12.02.2015',
-            'description': 'md5:89553aed1d641416001fe8d450f06cb9',
+            'description': 'md5:357933adeede13b202c7c21f91b871b2',
             'thumbnail': 're:^https?://.*\.(?:jpg|JPG)$',
+            'upload_date': '20150212',
             'duration': 2694,
         },
-        'skip': 'Only works from Russia',
+    }, {
+        # multiple formats via video_materials.json API
+        'url': 'http://www.1tv.ru/video_archive/projects/dobroeutro/p113641',
+        'info_dict': {
+            'id': '113641',
+            'ext': 'mp4',
+            'title': 'Весенняя аллергия. Доброе утро. Фрагмент выпуска от 07.04.2016',
+            'description': 'md5:8dcebb3dded0ff20fade39087fd1fee2',
+            'thumbnail': 're:^https?://.*\.(?:jpg|JPG)$',
+            'upload_date': '20160407',
+            'duration': 179,
+            'formats': 'mincount:3',
+        },
+        'params': {
+            'skip_download': True,
+        },
+    }, {
+        # single format only available via ONE_ONLINE_VIDEOS.archive_single_xml API
+        'url': 'http://www.1tv.ru/video_archive/series/f7552/p47038',
+        'md5': '519d306c5b5669761fd8906c39dbee23',
+        'info_dict': {
+            'id': '47038',
+            'ext': 'mp4',
+            'title': '"Побег". Второй сезон. 3 серия',
+            'description': 'md5:3abf8f6b9bce88201c33e9a3d794a00b',
+            'thumbnail': 're:^https?://.*\.(?:jpg|JPG)$',
+            'upload_date': '20120516',
+            'duration': 3080,
+        },
+    }, {
+        'url': 'http://www.1tv.ru/videoarchive/9967',
+        'only_matching': True,
     }]
 
     def _real_extract(self, url):
         video_id = self._match_id(url)
 
-        webpage = self._download_webpage(url, video_id, 'Downloading page')
+        # Videos with multiple formats only available via this API
+        video = self._download_json(
+            'http://www.1tv.ru/video_materials.json?legacy_id=%s' % video_id,
+            video_id, fatal=False)
 
-        video_url = self._html_search_regex(
+        description, thumbnail, upload_date, duration = [None] * 4
-            r'''(?s)(?:jwplayer\('flashvideoportal_1'\)\.setup\({|var\s+playlistObj\s*=).*?'file'\s*:\s*'([^']+)'.*?}\);''',
-            webpage, 'video URL')
 
-        title = self._html_search_regex(
+        if video:
-            [r'<div class="tv_translation">\s*<h1><a href="[^"]+">([^<]*)</a>',
+            item = video[0]
-             r"'title'\s*:\s*'([^']+)'"], webpage, 'title')
+            title = item['title']
-        description = self._html_search_regex(
+            quality = qualities(('ld', 'sd', 'hd', ))
-            r'<div class="descr">\s*<div>&nbsp;</div>\s*<p>([^<]*)</p></div>',
+            formats = [{
-            webpage, 'description', default=None) or self._html_search_meta(
+                'url': f['src'],
+                'format_id': f.get('name'),
+                'quality': quality(f.get('name')),
+            } for f in item['mbr'] if f.get('src')]
+            thumbnail = item.get('poster')
+        else:
+            # Some videos are not available via video_materials.json
+            video = self._download_xml(
+                'http://www.1tv.ru/owa/win/ONE_ONLINE_VIDEOS.archive_single_xml?pid=%s' % video_id,
+                video_id)
+
+            NS_MAP = {
+                'media': 'http://search.yahoo.com/mrss/',
+            }
+
+            item = xpath_element(video, './channel/item', fatal=True)
+            title = xpath_text(item, './title', fatal=True)
+            formats = [{
+                'url': content.attrib['url'],
+            } for content in item.findall(
+                compat_xpath(xpath_with_ns('./media:content', NS_MAP))) if content.attrib.get('url')]
+            thumbnail = xpath_attr(
+                item, xpath_with_ns('./media:thumbnail', NS_MAP), 'url')
+
+        self._sort_formats(formats)
+
+        webpage = self._download_webpage(url, video_id, 'Downloading page', fatal=False)
+        if webpage:
+            title = self._html_search_regex(
+                (r'<div class="tv_translation">\s*<h1><a href="[^"]+">([^<]*)</a>',
+                 r"'title'\s*:\s*'([^']+)'"),
+                webpage, 'title', default=None) or title
+            description = self._html_search_regex(
+                r'<div class="descr">\s*<div>&nbsp;</div>\s*<p>([^<]*)</p></div>',
+                webpage, 'description', default=None) or self._html_search_meta(
                 'description', webpage, 'description')
-
+            thumbnail = thumbnail or self._og_search_thumbnail(webpage)
-        thumbnail = self._og_search_thumbnail(webpage)
+            duration = int_or_none(self._html_search_meta(
-        duration = self._og_search_property(
+                'video:duration', webpage, 'video duration', fatal=False))
-            'video:duration', webpage,
+            upload_date = unified_strdate(self._html_search_meta(
-            'video duration', fatal=False)
+                'ya:ovs:upload_date', webpage, 'upload date', fatal=False))
-
-        like_count = self._html_search_regex(
-            r'title="Понравилось".*?/></label> \[(\d+)\]',
-            webpage, 'like count', default=None)
-        dislike_count = self._html_search_regex(
-            r'title="Не понравилось".*?/></label> \[(\d+)\]',
-            webpage, 'dislike count', default=None)
 
         return {
             'id': video_id,
-            'url': video_url,
             'thumbnail': thumbnail,
             'title': title,
             'description': description,
+            'upload_date': upload_date,
             'duration': int_or_none(duration),
-            'like_count': int_or_none(like_count),
+            'formats': formats
-            'dislike_count': int_or_none(dislike_count),
         }