[infoq] fix format extraction(closes #25984)

Fix the thumbnail URL. The issue was here for many years, never fixed. It's done ! :-)

Example : https://www.france.tv/france-2/de-gaulle-l-eclat-et-le-secret/de-gaulle-l-eclat-et-le-secret-saison-1/2035247-solitude.html

failed thumbnail url generated : http://pluzz.francetv.fr/staticftv/ref_emissions/2020-11-02/EMI_1104da66f533cc7dc5d0d07a181a18c2e2fe1d81_20201014122553940.jpg

right thumbnail url fixed : https://sivideo.webservices.francetelevisions.fr/staticftv/ref_emissions/2020-11-02/EMI_1104da66f533cc7dc5d0d07a181a18c2e2fe1d81_20201014122553940.jpg

test/test_YoutubeDL.py

@@ -919,6 +919,76 @@ class TestYoutubeDL(unittest.TestCase):
         self.assertEqual(downloaded['extractor'], 'testex')
         self.assertEqual(downloaded['extractor_key'], 'TestEx')
 
+    # Test case for https://github.com/ytdl-org/youtube-dl/issues/27064
+    def test_ignoreerrors_for_playlist_with_url_transparent_iterable_entries(self):
+
+        class _YDL(YDL):
+            def __init__(self, *args, **kwargs):
+                super(_YDL, self).__init__(*args, **kwargs)
+
+            def trouble(self, s, tb=None):
+                pass
+
+        ydl = _YDL({
+            'format': 'extra',
+            'ignoreerrors': True,
+        })
+
+        class VideoIE(InfoExtractor):
+            _VALID_URL = r'video:(?P<id>\d+)'
+
+            def _real_extract(self, url):
+                video_id = self._match_id(url)
+                formats = [{
+                    'format_id': 'default',
+                    'url': 'url:',
+                }]
+                if video_id == '0':
+                    raise ExtractorError('foo')
+                if video_id == '2':
+                    formats.append({
+                        'format_id': 'extra',
+                        'url': TEST_URL,
+                    })
+                return {
+                    'id': video_id,
+                    'title': 'Video %s' % video_id,
+                    'formats': formats,
+                }
+
+        class PlaylistIE(InfoExtractor):
+            _VALID_URL = r'playlist:'
+
+            def _entries(self):
+                for n in range(3):
+                    video_id = compat_str(n)
+                    yield {
+                        '_type': 'url_transparent',
+                        'ie_key': VideoIE.ie_key(),
+                        'id': video_id,
+                        'url': 'video:%s' % video_id,
+                        'title': 'Video Transparent %s' % video_id,
+                    }
+
+            def _real_extract(self, url):
+                return self.playlist_result(self._entries())
+
+        ydl.add_info_extractor(VideoIE(ydl))
+        ydl.add_info_extractor(PlaylistIE(ydl))
+        info = ydl.extract_info('playlist:')
+        entries = info['entries']
+        self.assertEqual(len(entries), 3)
+        self.assertTrue(entries[0] is None)
+        self.assertTrue(entries[1] is None)
+        self.assertEqual(len(ydl.downloaded_info_dicts), 1)
+        downloaded = ydl.downloaded_info_dicts[0]
+        self.assertEqual(entries[2], downloaded)
+        self.assertEqual(downloaded['url'], TEST_URL)
+        self.assertEqual(downloaded['title'], 'Video Transparent 2')
+        self.assertEqual(downloaded['id'], '2')
+        self.assertEqual(downloaded['extractor'], 'Video')
+        self.assertEqual(downloaded['extractor_key'], 'Video')
+
 
 if __name__ == '__main__':
     unittest.main()

youtube_dl/YoutubeDL.py

@@ -793,10 +793,37 @@ class YoutubeDL(object):
                 self.report_warning('The program functionality for this site has been marked as broken, '
                                     'and will probably not work.')
 
+            return self.__extract_info(url, ie, download, extra_info, process)
+        else:
+            self.report_error('no suitable InfoExtractor for URL %s' % url)
+
+    def __handle_extraction_exceptions(func):
+        def wrapper(self, *args, **kwargs):
             try:
+                return func(self, *args, **kwargs)
+            except GeoRestrictedError as e:
+                msg = e.msg
+                if e.countries:
+                    msg += '\nThis video is available in %s.' % ', '.join(
+                        map(ISO3166Utils.short2full, e.countries))
+                msg += '\nYou might want to use a VPN or a proxy server (with --proxy) to workaround.'
+                self.report_error(msg)
+            except ExtractorError as e:  # An error we somewhat expected
+                self.report_error(compat_str(e), e.format_traceback())
+            except MaxDownloadsReached:
+                raise
+            except Exception as e:
+                if self.params.get('ignoreerrors', False):
+                    self.report_error(error_to_compat_str(e), tb=encode_compat_str(traceback.format_exc()))
+                else:
+                    raise
+        return wrapper
+
+    @__handle_extraction_exceptions
+    def __extract_info(self, url, ie, download, extra_info, process):
         ie_result = ie.extract(url)
         if ie_result is None:  # Finished already (backwards compatibility; listformats and friends should be moved here)
-                    break
+            return
         if isinstance(ie_result, list):
             # Backwards compatibility: old IE result format
             ie_result = {
@@ -808,27 +835,6 @@ class YoutubeDL(object):
             return self.process_ie_result(ie_result, download, extra_info)
         else:
             return ie_result
-            except GeoRestrictedError as e:
-                msg = e.msg
-                if e.countries:
-                    msg += '\nThis video is available in %s.' % ', '.join(
-                        map(ISO3166Utils.short2full, e.countries))
-                msg += '\nYou might want to use a VPN or a proxy server (with --proxy) to workaround.'
-                self.report_error(msg)
-                break
-            except ExtractorError as e:  # An error we somewhat expected
-                self.report_error(compat_str(e), e.format_traceback())
-                break
-            except MaxDownloadsReached:
-                raise
-            except Exception as e:
-                if self.params.get('ignoreerrors', False):
-                    self.report_error(error_to_compat_str(e), tb=encode_compat_str(traceback.format_exc()))
-                    break
-                else:
-                    raise
-        else:
-            self.report_error('no suitable InfoExtractor for URL %s' % url)
 
     def add_default_extra_info(self, ie_result, ie, url):
         self.add_extra_info(ie_result, {
@@ -1003,9 +1009,8 @@ class YoutubeDL(object):
                     self.to_screen('[download] ' + reason)
                     continue
 
-                entry_result = self.process_ie_result(entry,
+                entry_result = self.__process_iterable_entry(entry, download, extra)
-                                                      download=download,
+                # TODO: skip failed (empty) entries?
-                                                      extra_info=extra)
                 playlist_results.append(entry_result)
             ie_result['entries'] = playlist_results
             self.to_screen('[download] Finished downloading playlist: %s' % playlist)
@@ -1034,6 +1039,11 @@ class YoutubeDL(object):
         else:
             raise Exception('Invalid result type: %s' % result_type)
 
+    @__handle_extraction_exceptions
+    def __process_iterable_entry(self, entry, download, extra_info):
+        return self.process_ie_result(
+            entry, download=download, extra_info=extra_info)
+
     def _build_format_filter(self, filter_spec):
         " Returns a function to filter the formats according to the filter_spec "
 

youtube_dl/downloader/http.py

@@ -109,7 +109,9 @@ class HttpFD(FileDownloader):
                 try:
                     ctx.data = self.ydl.urlopen(request)
                 except (compat_urllib_error.URLError, ) as err:
-                    if isinstance(err.reason, socket.timeout):
+                    # reason may not be available, e.g. for urllib2.HTTPError on python 2.6
+                    reason = getattr(err, 'reason', None)
+                    if isinstance(reason, socket.timeout):
                         raise RetryDownload(err)
                     raise err
                 # When trying to resume, Content-Range HTTP header of response has to be checked

youtube_dl/extractor/francetv.py

@@ -211,7 +211,7 @@ class FranceTVIE(InfoExtractor):
             'id': video_id,
             'title': self._live_title(title) if is_live else title,
             'description': clean_html(info.get('synopsis')),
-            'thumbnail': urljoin('http://pluzz.francetv.fr', info.get('image')),
+            'thumbnail': urljoin('https://sivideo.webservices.francetelevisions.fr', info.get('image')),
             'duration': int_or_none(info.get('real_duration')) or parse_duration(info.get('duree')),
             'timestamp': int_or_none(try_get(info, lambda x: x['diffusion']['timestamp'])),
             'is_live': is_live,

youtube_dl/extractor/infoq.py

@@ -54,7 +54,7 @@ class InfoQIE(BokeCCBaseIE):
 
     def _extract_rtmp_video(self, webpage):
         # The server URL is hardcoded
-        video_url = 'rtmpe://video.infoq.com/cfx/st/'
+        video_url = 'rtmpe://videof.infoq.com/cfx/st/'
 
         # Extract video URL
         encoded_id = self._search_regex(
@@ -86,17 +86,18 @@ class InfoQIE(BokeCCBaseIE):
         return [{
             'format_id': 'http_video',
             'url': http_video_url,
+            'http_headers': {'Referer': 'https://www.infoq.com/'},
         }]
 
     def _extract_http_audio(self, webpage, video_id):
-        fields = self._hidden_inputs(webpage)
+        fields = self._form_hidden_inputs('mp3Form', webpage)
         http_audio_url = fields.get('filename')
         if not http_audio_url:
             return []
 
         # base URL is found in the Location header in the response returned by
         # GET https://www.infoq.com/mp3download.action?filename=... when logged in.
-        http_audio_url = compat_urlparse.urljoin('http://res.infoq.com/downloads/mp3downloads/', http_audio_url)
+        http_audio_url = compat_urlparse.urljoin('http://ress.infoq.com/downloads/mp3downloads/', http_audio_url)
         http_audio_url = update_url_query(http_audio_url, self._extract_cf_auth(webpage))
 
         # audio file seem to be missing some times even if there is a download link

youtube_dl/extractor/youtube.py

@@ -283,6 +283,8 @@ class YoutubeBaseInfoExtractor(InfoExtractor):
         },
     }
 
+    _YT_INITIAL_DATA_RE = r'(?:window\s*\[\s*["\']ytInitialData["\']\s*\]|ytInitialData)\s*=\s*({.+?})\s*;'
+
     def _call_api(self, ep, query, video_id):
         data = self._DEFAULT_API_DATA.copy()
         data.update(query)
@@ -299,8 +301,8 @@ class YoutubeBaseInfoExtractor(InfoExtractor):
     def _extract_yt_initial_data(self, video_id, webpage):
         return self._parse_json(
             self._search_regex(
-                r'(?:window\s*\[\s*["\']ytInitialData["\']\s*\]|ytInitialData)\s*=\s*({.+?})\s*;',
+                (r'%s\s*\n' % self._YT_INITIAL_DATA_RE,
-                webpage, 'yt initial data'),
+                 self._YT_INITIAL_DATA_RE), webpage, 'yt initial data'),
             video_id)
 
 
@@ -1066,6 +1068,22 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
                 'skip_download': True,
             },
         },
+        {
+            # with '};' inside yt initial data (see https://github.com/ytdl-org/youtube-dl/issues/27093)
+            'url': 'https://www.youtube.com/watch?v=CHqg6qOn4no',
+            'info_dict': {
+                'id': 'CHqg6qOn4no',
+                'ext': 'mp4',
+                'title': 'Part 77   Sort a list of simple types in c#',
+                'description': 'md5:b8746fa52e10cdbf47997903f13b20dc',
+                'upload_date': '20130831',
+                'uploader_id': 'kudvenkat',
+                'uploader': 'kudvenkat',
+            },
+            'params': {
+                'skip_download': True,
+            },
+        },
     ]
 
     def __init__(self, *args, **kwargs):
@@ -2456,13 +2474,13 @@ class YoutubeTabIE(YoutubeBaseInfoExtractor):
         },
         'playlist_mincount': 138,
     }, {
-        'url': 'https://invidio.us/channel/UC23qupoDRn9YOAVzeoxjOQA',
+        'url': 'https://invidio.us/channel/UCmlqkdCBesrv2Lak1mF_MxA',
         'only_matching': True,
     }, {
-        'url': 'https://www.youtubekids.com/channel/UCyu8StPfZWapR6rfW_JgqcA',
+        'url': 'https://www.youtubekids.com/channel/UCmlqkdCBesrv2Lak1mF_MxA',
         'only_matching': True,
     }, {
-        'url': 'https://music.youtube.com/channel/UCT-K0qO8z6NzWrywqefBPBQ',
+        'url': 'https://music.youtube.com/channel/UCmlqkdCBesrv2Lak1mF_MxA',
         'only_matching': True,
     }, {
         'note': 'Playlist with deleted videos (#651). As a bonus, the video #51 is also twice in this list.',
@@ -2509,7 +2527,7 @@ class YoutubeTabIE(YoutubeBaseInfoExtractor):
         },
         'playlist_mincount': 11,
     }, {
-        'url': 'https://invidio.us/playlist?list=PLDIoUOhQQPlXr63I_vwF9GD8sAKh77dWU',
+        'url': 'https://invidio.us/playlist?list=PL4lCao7KL_QFVb7Iudeipvc2BCavECqzc',
         'only_matching': True,
     }, {
         # Playlist URL that does not actually serve a playlist