From 21c924f4068692786e0c5435689d10f3d17ef612 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Jaime=20Marqui=CC=81nez=20Ferra=CC=81ndiz?= Date: Tue, 29 Oct 2013 20:58:49 +0100 Subject: [arte] Download the 'Originalversion' version if it's the only one available (fixes #1682) --- youtube_dl/extractor/arte.py | 17 ++++++++++++----- 1 file changed, 12 insertions(+), 5 deletions(-) (limited to 'youtube_dl/extractor/arte.py') diff --git a/youtube_dl/extractor/arte.py b/youtube_dl/extractor/arte.py index d39b48951..e10c74c11 100644 --- a/youtube_dl/extractor/arte.py +++ b/youtube_dl/extractor/arte.py @@ -158,7 +158,9 @@ class ArteTVPlus7IE(InfoExtractor): 'thumbnail': player_info.get('programImage') or player_info.get('VTU', {}).get('IUR'), } - formats = player_info['VSR'].values() + all_formats = player_info['VSR'].values() + # Some formats use the m3u8 protocol + all_formats = list(filter(lambda f: f.get('videoFormat') != 'M3U8', all_formats)) def _match_lang(f): if f.get('versionCode') is None: return True @@ -170,11 +172,16 @@ class ArteTVPlus7IE(InfoExtractor): regexes = [r'VO?%s' % l, r'VO?.-ST%s' % l] return any(re.match(r, f['versionCode']) for r in regexes) # Some formats may not be in the same language as the url - formats = filter(_match_lang, formats) - # Some formats use the m3u8 protocol - formats = filter(lambda f: f.get('videoFormat') != 'M3U8', formats) - # We order the formats by quality + formats = filter(_match_lang, all_formats) formats = list(formats) # in python3 filter returns an iterator + if not formats: + # Some videos are only available in the 'Originalversion' + # they aren't tagged as being in French or German + if all(f['versionCode'] == 'VO' for f in all_formats): + formats = all_formats + else: + raise ExtractorError(u'The formats list is empty') + # We order the formats by quality if re.match(r'[A-Z]Q', formats[0]['quality']) is not None: sort_key = lambda f: ['HQ', 'MQ', 'EQ', 'SQ'].index(f['quality']) else: -- cgit v1.2.3 From 566d4e0425245e46f211fec8cd38d369e2c8de2c Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Jaime=20Marqui=CC=81nez=20Ferra=CC=81ndiz?= Date: Sat, 9 Nov 2013 19:01:23 +0100 Subject: [arte] Make sure the format_id is unique (closes #1739) Include the bitrate and use the height instead of the quality field. --- youtube_dl/extractor/arte.py | 14 +++++++++----- 1 file changed, 9 insertions(+), 5 deletions(-) (limited to 'youtube_dl/extractor/arte.py') diff --git a/youtube_dl/extractor/arte.py b/youtube_dl/extractor/arte.py index e10c74c11..a636c6be5 100644 --- a/youtube_dl/extractor/arte.py +++ b/youtube_dl/extractor/arte.py @@ -10,6 +10,7 @@ from ..utils import ( unified_strdate, determine_ext, get_element_by_id, + compat_str, ) # There are different sources of video in arte.tv, the extraction process @@ -191,10 +192,13 @@ class ArteTVPlus7IE(InfoExtractor): formats = sorted(formats, key=lambda f: re.match(r'VO(F|A)-STM\1', f.get('versionCode', '')) is None) # Pick the best quality def _format(format_info): - quality = format_info['quality'] - m_quality = re.match(r'\w*? - (\d*)p', quality) - if m_quality is not None: - quality = m_quality.group(1) + quality = '' + height = format_info.get('height') + if height is not None: + quality = compat_str(height) + bitrate = format_info.get('bitrate') + if bitrate is not None: + quality += '-%d' % bitrate if format_info.get('versionCode') is not None: format_id = u'%s-%s' % (quality, format_info['versionCode']) else: @@ -203,7 +207,7 @@ class ArteTVPlus7IE(InfoExtractor): 'format_id': format_id, 'format_note': format_info.get('versionLibelle'), 'width': format_info.get('width'), - 'height': format_info.get('height'), + 'height': height, } if format_info['mediaType'] == u'rtmp': info['url'] = format_info['streamer'] -- cgit v1.2.3 From f470c6c812882d3726fe379d21069adac3cb2c69 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Jaime=20Marqui=CC=81nez=20Ferra=CC=81ndiz?= Date: Sat, 9 Nov 2013 19:05:19 +0100 Subject: [arte] Improve the format sorting Also use the bitrate. Prefer normal version and sourds/mal version over original version with subtitles. --- youtube_dl/extractor/arte.py | 19 +++++++++++++------ 1 file changed, 13 insertions(+), 6 deletions(-) (limited to 'youtube_dl/extractor/arte.py') diff --git a/youtube_dl/extractor/arte.py b/youtube_dl/extractor/arte.py index a636c6be5..b35a679e3 100644 --- a/youtube_dl/extractor/arte.py +++ b/youtube_dl/extractor/arte.py @@ -182,15 +182,22 @@ class ArteTVPlus7IE(InfoExtractor): formats = all_formats else: raise ExtractorError(u'The formats list is empty') - # We order the formats by quality + if re.match(r'[A-Z]Q', formats[0]['quality']) is not None: - sort_key = lambda f: ['HQ', 'MQ', 'EQ', 'SQ'].index(f['quality']) + def sort_key(f): + return ['HQ', 'MQ', 'EQ', 'SQ'].index(f['quality']) else: - sort_key = lambda f: int(f.get('height',-1)) + def sort_key(f): + return ( + # Sort first by quality + int(f.get('height',-1)), + int(f.get('bitrate',-1)), + # The original version with subtitles has lower relevance + re.match(r'VO-ST(F|A)', f.get('versionCode', '')) is None, + # The version with sourds/mal subtitles has also lower relevance + re.match(r'VO?(F|A)-STM\1', f.get('versionCode', '')) is None, + ) formats = sorted(formats, key=sort_key) - # Prefer videos without subtitles in the same language - formats = sorted(formats, key=lambda f: re.match(r'VO(F|A)-STM\1', f.get('versionCode', '')) is None) - # Pick the best quality def _format(format_info): quality = '' height = format_info.get('height') -- cgit v1.2.3 From ddf49c63445d236138846d778b18db6ede067fc8 Mon Sep 17 00:00:00 2001 From: Philipp Hagemeister Date: Sun, 17 Nov 2013 11:05:49 +0100 Subject: [arte] remove two typos --- youtube_dl/extractor/arte.py | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'youtube_dl/extractor/arte.py') diff --git a/youtube_dl/extractor/arte.py b/youtube_dl/extractor/arte.py index b35a679e3..44d0b5d70 100644 --- a/youtube_dl/extractor/arte.py +++ b/youtube_dl/extractor/arte.py @@ -69,7 +69,7 @@ class ArteTvIE(InfoExtractor): lang = mobj.group('lang') return self._extract_liveweb(url, name, lang) - if re.search(self._LIVE_URL, video_id) is not None: + if re.search(self._LIVE_URL, url) is not None: raise ExtractorError(u'Arte live streams are not yet supported, sorry') # self.extractLiveStream(url) # return @@ -115,7 +115,7 @@ class ArteTvIE(InfoExtractor): event_doc = config_doc.find('event') url_node = event_doc.find('video').find('urlHd') if url_node is None: - url_node = video_doc.find('urlSd') + url_node = event_doc.find('urlSd') return {'id': video_id, 'title': event_doc.find('name%s' % lang.capitalize()).text, -- cgit v1.2.3