From e0f1fb0a27612c2398df59dd85194edfdf8cbc2a Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Sergey=20M=E2=80=A4?= Date: Sun, 23 Jul 2017 00:25:23 +0700 Subject: [PATCH] [mtv] Skip missing video parts (closes #13690) --- youtube_dl/extractor/mtv.py | 28 ++++++++++++++++++++++------ youtube_dl/extractor/vh1.py | 12 ++++++++---- 2 files changed, 30 insertions(+), 10 deletions(-) diff --git a/youtube_dl/extractor/mtv.py b/youtube_dl/extractor/mtv.py index 8acea1461..fc098cd13 100644 --- a/youtube_dl/extractor/mtv.py +++ b/youtube_dl/extractor/mtv.py @@ -83,7 +83,7 @@ class MTVServicesInfoExtractor(InfoExtractor): hls_url = rendition.find('./src').text formats.extend(self._extract_m3u8_formats( hls_url, video_id, ext='mp4', entry_protocol='m3u8_native', - m3u8_id='hls')) + m3u8_id='hls', fatal=False)) else: # fms try: @@ -106,7 +106,8 @@ class MTVServicesInfoExtractor(InfoExtractor): }]) except (KeyError, TypeError): raise ExtractorError('Invalid rendition field.') - self._sort_formats(formats) + if formats: + self._sort_formats(formats) return formats def _extract_subtitles(self, mdoc, mtvn_id): @@ -133,8 +134,11 @@ class MTVServicesInfoExtractor(InfoExtractor): mediagen_url += 'acceptMethods=' mediagen_url += 'hls' if use_hls else 'fms' - mediagen_doc = self._download_xml(mediagen_url, video_id, - 'Downloading video urls') + mediagen_doc = self._download_xml( + mediagen_url, video_id, 'Downloading video urls', fatal=False) + + if mediagen_doc is False: + return None item = mediagen_doc.find('./video/item') if item is not None and item.get('type') == 'text': @@ -174,6 +178,13 @@ class MTVServicesInfoExtractor(InfoExtractor): formats = self._extract_video_formats(mediagen_doc, mtvn_id, video_id) + # Some parts of complete video may be missing (e.g. missing Act 3 in + # http://www.southpark.de/alle-episoden/s14e01-sexual-healing) + if not formats: + return None + + self._sort_formats(formats) + return { 'title': title, 'formats': formats, @@ -205,9 +216,14 @@ class MTVServicesInfoExtractor(InfoExtractor): title = xpath_text(idoc, './channel/title') description = xpath_text(idoc, './channel/description') + entries = [] + for item in idoc.findall('.//item'): + info = self._get_video_info(item, use_hls) + if info: + entries.append(info) + return self.playlist_result( - [self._get_video_info(item, use_hls) for item in idoc.findall('.//item')], - playlist_title=title, playlist_description=description) + entries, playlist_title=title, playlist_description=description) def _extract_triforce_mgid(self, webpage, data_zone=None, video_id=None): triforce_feed = self._parse_json(self._search_regex( diff --git a/youtube_dl/extractor/vh1.py b/youtube_dl/extractor/vh1.py index 6be3774b7..570fa45ea 100644 --- a/youtube_dl/extractor/vh1.py +++ b/youtube_dl/extractor/vh1.py @@ -121,7 +121,11 @@ class VH1IE(MTVIE): idoc = self._download_xml( doc_url, video_id, 'Downloading info', transform_source=fix_xml_ampersands) - return self.playlist_result( - [self._get_video_info(item) for item in idoc.findall('.//item')], - playlist_id=video_id, - ) + + entries = [] + for item in idoc.findall('.//item'): + info = self._get_video_info(item) + if info: + entries.append(info) + + return self.playlist_result(entries, playlist_id=video_id)