[1tv] Improve extraction and add support for playlists (closes #11335)

youtube_dl/extractor/firsttv.py

@@ -2,7 +2,10 @@
 from __future__ import unicode_literals
 
 from .common import InfoExtractor
-from ..compat import compat_urlparse
+from ..compat import (
+    compat_str,
+    compat_urlparse,
+)
 from ..utils import (
     int_or_none,
     qualities,
@@ -22,8 +25,7 @@ class FirstTVIE(InfoExtractor):
         'info_dict': {
             'id': '40049',
             'ext': 'mp4',
-            'title': 'Гость Людмила Сенчина. Наедине со всеми. Выпуск от 12.02.2015',
-            'description': 'md5:36a39c1d19618fec57d12efe212a8370',
+            'title': 'Гость Людмила Сенчина. Наедине со всеми. Выпуск от 12.02.2015',
             'thumbnail': 're:^https?://.*\.(?:jpg|JPG)$',
             'upload_date': '20150212',
             'duration': 2694,
@@ -34,8 +36,7 @@ class FirstTVIE(InfoExtractor):
         'info_dict': {
             'id': '364746',
             'ext': 'mp4',
-            'title': 'Весенняя аллергия. Доброе утро. Фрагмент выпуска от 07.04.2016',
-            'description': 'md5:a242eea0031fd180a4497d52640a9572',
+            'title': 'Весенняя аллергия. Доброе утро. Фрагмент выпуска от 07.04.2016',
             'thumbnail': 're:^https?://.*\.(?:jpg|JPG)$',
             'upload_date': '20160407',
             'duration': 179,
@@ -44,6 +45,17 @@ class FirstTVIE(InfoExtractor):
         'params': {
             'skip_download': True,
         },
+    }, {
+        'url': 'http://www.1tv.ru/news/issue/2016-12-01/14:00',
+        'info_dict': {
+            'id': '14:00',
+            'title': 'Выпуск новостей в 14:00   1 декабря 2016 года. Новости. Первый канал',
+            'description': 'md5:2e921b948f8c1ff93901da78ebdb1dfd',
+        },
+        'playlist_count': 13,
+    }, {
+        'url': 'http://www.1tv.ru/shows/tochvtoch-supersezon/vystupleniya/evgeniy-dyatlov-vladimir-vysockiy-koni-priveredlivye-toch-v-toch-supersezon-fragment-vypuska-ot-06-11-2016',
+        'only_matching': True,
     }]
 
     def _real_extract(self, url):
@@ -51,43 +63,66 @@ class FirstTVIE(InfoExtractor):
 
         webpage = self._download_webpage(url, display_id)
         playlist_url = compat_urlparse.urljoin(url, self._search_regex(
-            r'data-playlist-url="([^"]+)', webpage, 'playlist url'))
+            r'data-playlist-url=(["\'])(?P<url>(?:(?!\1).)+)\1',
+            webpage, 'playlist url', group='url'))
+
+        parsed_url = compat_urlparse.urlparse(playlist_url)
+        qs = compat_urlparse.parse_qs(parsed_url.query)
+        item_ids = qs.get('videos_ids[]') or qs.get('news_ids[]')
 
-        item = self._download_json(playlist_url, display_id)[0]
-        video_id = item['id']
-        quality = qualities(('ld', 'sd', 'hd', ))
+        items = self._download_json(playlist_url, display_id)
+
+        if item_ids:
+            items = [
+                item for item in items
+                if item.get('uid') and compat_str(item['uid']) in item_ids]
+        else:
+            items = [items[0]]
+
+        entries = []
+        QUALITIES = ('ld', 'sd', 'hd', )
+
+        for item in items:
+            title = item['title']
+            quality = qualities(QUALITIES)
             formats = []
             for f in item.get('mbr', []):
                 src = f.get('src')
-            if not src:
+                if not src or not isinstance(src, compat_str):
                     continue
-            fname = f.get('name')
+                tbr = int_or_none(self._search_regex(
+                    r'_(\d{3,})\.mp4', src, 'tbr', default=None))
                 formats.append({
                     'url': src,
-                'format_id': fname,
-                'quality': quality(fname),
+                    'format_id': f.get('name'),
+                    'tbr': tbr,
+                    'quality': quality(f.get('name')),
                 })
             self._sort_formats(formats)
 
-        title = self._html_search_regex(
-            (r'<div class="tv_translation">\s*<h1><a href="[^"]+">([^<]*)</a>',
-             r"'title'\s*:\s*'([^']+)'"),
-            webpage, 'title', default=None) or item['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')
-        duration = int_or_none(self._html_search_meta(
+            thumbnail = item.get('poster') or self._og_search_thumbnail(webpage)
+            duration = int_or_none(item.get('duration') or self._html_search_meta(
                 'video:duration', webpage, 'video duration', fatal=False))
             upload_date = unified_strdate(self._html_search_meta(
-            'ya:ovs:upload_date', webpage, 'upload date', fatal=False))
+                'ya:ovs:upload_date', webpage, 'upload date', default=None))
 
-        return {
-            'id': video_id,
-            'thumbnail': item.get('poster') or self._og_search_thumbnail(webpage),
+            entries.append({
+                'id': item.get('id') or uid,
+                'thumbnail': thumbnail,
                 'title': title,
-            'description': description,
                 'upload_date': upload_date,
                 'duration': int_or_none(duration),
                 'formats': formats
-        }
+            })
+
+        title = self._html_search_regex(
+            (r'<div class="tv_translation">\s*<h1><a href="[^"]+">([^<]*)</a>',
+             r"'title'\s*:\s*'([^']+)'"),
+            webpage, 'title', default=None) or self._og_search_title(
+            webpage, default=None)
+        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', default=None)
+
+        return self.playlist_result(entries, display_id, title, description)