[periscope] Extract width and height (closes #20015)

youtube_dl/downloader/external.py

@@ -121,7 +121,11 @@ class CurlFD(ExternalFD):
         cmd += self._valueless_option('--silent', 'noprogress')
         cmd += self._valueless_option('--verbose', 'verbose')
         cmd += self._option('--limit-rate', 'ratelimit')
-        cmd += self._option('--retry', 'retries')
+        retry = self._option('--retry', 'retries')
+        if len(retry) == 2:
+            if retry[1] in ('inf', 'infinite'):
+                retry[1] = '2147483647'
+            cmd += retry
         cmd += self._option('--max-filesize', 'max_filesize')
         cmd += self._option('--interface', 'source_address')
         cmd += self._option('--proxy', 'proxy')
@@ -160,6 +164,12 @@ class WgetFD(ExternalFD):
         cmd = [self.exe, '-O', tmpfilename, '-nv', '--no-cookies']
         for key, val in info_dict['http_headers'].items():
             cmd += ['--header', '%s: %s' % (key, val)]
+        cmd += self._option('--limit-rate', 'ratelimit')
+        retry = self._option('--tries', 'retries')
+        if len(retry) == 2:
+            if retry[1] in ('inf', 'infinite'):
+                retry[1] = '0'
+            cmd += retry
         cmd += self._option('--bind-address', 'source_address')
         cmd += self._option('--proxy', 'proxy')
         cmd += self._valueless_option('--no-check-certificate', 'nocheckcertificate')

youtube_dl/extractor/bbc.py

@@ -1,8 +1,9 @@
 # coding: utf-8
 from __future__ import unicode_literals
 
-import re
 import itertools
+import re
+import xml
 
 from .common import InfoExtractor
 from ..utils import (
@@ -17,6 +18,7 @@ from ..utils import (
     parse_iso8601,
     try_get,
     unescapeHTML,
+    url_or_none,
     urlencode_postdata,
     urljoin,
 )
@@ -310,7 +312,13 @@ class BBCCoUkIE(InfoExtractor):
     def _get_subtitles(self, media, programme_id):
         subtitles = {}
         for connection in self._extract_connections(media):
-            captions = self._download_xml(connection.get('href'), programme_id, 'Downloading captions')
+            cc_url = url_or_none(connection.get('href'))
+            if not cc_url:
+                continue
+            captions = self._download_xml(
+                cc_url, programme_id, 'Downloading captions', fatal=False)
+            if not isinstance(captions, xml.etree.ElementTree.Element):
+                continue
             lang = captions.get('{http://www.w3.org/XML/1998/namespace}lang', 'en')
             subtitles[lang] = [
                 {

youtube_dl/extractor/metacafe.py

@@ -1,12 +1,13 @@
 from __future__ import unicode_literals
 
+import json
 import re
 
 from .common import InfoExtractor
 from ..compat import (
     compat_parse_qs,
+    compat_urllib_parse,
     compat_urllib_parse_unquote,
-    compat_urllib_parse_urlencode,
 )
 from ..utils import (
     determine_ext,
@@ -144,7 +145,7 @@ class MetacafeIE(InfoExtractor):
 
         headers = {
             # Disable family filter
-            'Cookie': 'user=%s; ' % compat_urllib_parse_urlencode({'ffilter': False})
+            'Cookie': 'user=%s; ' % compat_urllib_parse.quote(json.dumps({'ffilter': False}))
         }
 
         # AnyClip videos require the flashversion cookie so that we get the link

youtube_dl/extractor/periscope.py

@@ -5,6 +5,7 @@ import re
 
 from .common import InfoExtractor
 from ..utils import (
+    int_or_none,
     parse_iso8601,
     unescapeHTML,
 )
@@ -75,6 +76,14 @@ class PeriscopeIE(PeriscopeBaseIE):
             'url': broadcast[image],
         } for image in ('image_url', 'image_url_small') if broadcast.get(image)]
 
+        width = int_or_none(broadcast.get('width'))
+        height = int_or_none(broadcast.get('height'))
+
+        def add_width_and_height(f):
+            for key, val in (('width', width), ('height', height)):
+                if not f.get(key):
+                    f[key] = val
+
         video_urls = set()
         formats = []
         for format_id in ('replay', 'rtmp', 'hls', 'https_hls', 'lhls', 'lhlsweb'):
@@ -83,16 +92,21 @@ class PeriscopeIE(PeriscopeBaseIE):
                 continue
             video_urls.add(video_url)
             if format_id != 'rtmp':
-                formats.extend(self._extract_m3u8_formats(
+                m3u8_formats = self._extract_m3u8_formats(
                     video_url, token, 'mp4',
                     entry_protocol='m3u8_native'
                     if state in ('ended', 'timed_out') else 'm3u8',
-                    m3u8_id=format_id, fatal=False))
+                    m3u8_id=format_id, fatal=False)
+                if len(m3u8_formats) == 1:
+                    add_width_and_height(m3u8_formats[0])
+                formats.extend(m3u8_formats)
                 continue
-            formats.append({
+            rtmp_format = {
                 'url': video_url,
                 'ext': 'flv' if format_id == 'rtmp' else 'mp4',
-            })
+            }
+            add_width_and_height(rtmp_format)
+            formats.append(rtmp_format)
         self._sort_formats(formats)
 
         return {

youtube_dl/extractor/servus.py

@@ -1,31 +1,44 @@
 # coding: utf-8
 from __future__ import unicode_literals
 
+import re
+
 from .common import InfoExtractor
 
 
 class ServusIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?servus\.com/(?:at|de)/p/[^/]+/(?P<id>AA-\w+|\d+-\d+)'
+    _VALID_URL = r'https?://(?:www\.)?servus\.com/(?:(?:at|de)/p/[^/]+|tv/videos)/(?P<id>[aA]{2}-\w+|\d+-\d+)'
     _TESTS = [{
         'url': 'https://www.servus.com/de/p/Die-Gr%C3%BCnen-aus-Sicht-des-Volkes/AA-1T6VBU5PW1W12/',
-        'md5': '046dee641cda1c4cabe13baef3be2c1c',
+        'md5': '3e1dd16775aa8d5cbef23628cfffc1f4',
         'info_dict': {
             'id': 'AA-1T6VBU5PW1W12',
             'ext': 'mp4',
-            'title': 'Die Grünen aus Volkssicht',
-            'description': 'md5:052b5da1cb2cd7d562ef1f19be5a5cba',
-            'thumbnail': r're:^https?://.*\.jpg$',
+            'title': 'Die Grünen aus Sicht des Volkes',
+            'description': 'md5:1247204d85783afe3682644398ff2ec4',
+            'thumbnail': r're:^https?://.*\.jpg',
         }
     }, {
         'url': 'https://www.servus.com/at/p/Wie-das-Leben-beginnt/1309984137314-381415152/',
         'only_matching': True,
+    }, {
+        'url': 'https://www.servus.com/tv/videos/aa-1t6vbu5pw1w12/',
+        'only_matching': True,
+    }, {
+        'url': 'https://www.servus.com/tv/videos/1380889096408-1235196658/',
+        'only_matching': True,
     }]
 
     def _real_extract(self, url):
-        video_id = self._match_id(url)
+        video_id = self._match_id(url).upper()
         webpage = self._download_webpage(url, video_id)
 
-        title = self._og_search_title(webpage)
+        title = self._search_regex(
+            (r'videoLabel\s*=\s*(["\'])(?P<title>(?:(?!\1).)+)\1',
+             r'<h\d+[^>]+\bclass=["\']heading--(?:one|two)["\'][^>]*>(?P<title>[^<]+)'),
+            webpage, 'title', default=None,
+            group='title') or self._og_search_title(webpage)
+        title = re.sub(r'\s*-\s*Servus TV\s*$', '', title)
         description = self._og_search_description(webpage)
         thumbnail = self._og_search_thumbnail(webpage)