From: Sergey M․ Date: Sat, 30 Apr 2016 22:18:56 +0000 (+0600) Subject: [tagesschau] Add support for playlists X-Git-Url: http://git.oshgnacknak.de/?a=commitdiff_plain;h=4c1b2e5c0ea6a041bfd773efd7c4ac78ac8f3b4b;p=youtube-dl [tagesschau] Add support for playlists --- diff --git a/youtube_dl/extractor/tagesschau.py b/youtube_dl/extractor/tagesschau.py index fcccb230c..e58385c57 100644 --- a/youtube_dl/extractor/tagesschau.py +++ b/youtube_dl/extractor/tagesschau.py @@ -8,11 +8,11 @@ from ..utils import parse_filesize class TagesschauIE(InfoExtractor): - _VALID_URL = r'https?://(?:www\.)?tagesschau\.de/multimedia/(?:[^/]+/)*?[^/#?]+?(?P-?[0-9]+)(?:~_?[^/#?]+?)?\.html' + _VALID_URL = r'https?://(?:www\.)?tagesschau\.de/[^/]+/(?:[^/]+/)*?[^/#?]+?(?P-?[0-9]+)(?:~_?[^/#?]+?)?\.html' _TESTS = [{ 'url': 'http://www.tagesschau.de/multimedia/video/video-102143.html', - 'md5': '917a228bc7df7850783bc47979673a09', + 'md5': 'f7c27a0eff3bfe8c7727e65f8fe1b1e6', 'info_dict': { 'id': '102143', 'ext': 'mp4', @@ -40,6 +40,13 @@ class TagesschauIE(InfoExtractor): 'description': 'Flüchtlingsdebatte: Hitzig, aber wenig hilfreich', 'thumbnail': 're:^https?:.*\.jpg$', }, + }, { + 'url': 'http://www.tagesschau.de/inland/afd-parteitag-135.html', + 'info_dict': { + 'id': '135', + 'title': 'Möchtegern-Underdog mit Machtanspruch', + }, + 'playlist_count': 2, }, { 'url': 'http://www.tagesschau.de/multimedia/sendung/tsg-3771.html', 'only_matching': True, @@ -75,6 +82,41 @@ class TagesschauIE(InfoExtractor): 'xxl': {'quality': 5}, } + def _extract_formats(self, download_text): + links = re.finditer( + r'', + download_text) + formats = [] + for l in links: + format_id = self._search_regex( + r'.*/[^/.]+\.([^/]+)\.[^/.]+', l.group('url'), 'format ID') + format = { + 'format_id': format_id, + 'url': l.group('url'), + 'format_name': l.group('name'), + } + m = re.match( + r'''(?x) + Video:\s*(?P[a-zA-Z0-9/._-]+)\s*&\#10; + (?P[0-9]+)x(?P[0-9]+)px&\#10; + (?P[0-9]+)kbps&\#10; + Audio:\s*(?P[0-9]+)kbps,\s*(?P[A-Za-z\.0-9]+)&\#10; + Größe:\s*(?P[0-9.,]+\s+[a-zA-Z]*B)''', + l.group('title')) + if m: + format.update({ + 'format_note': m.group('audio_desc'), + 'vcodec': m.group('vcodec'), + 'width': int(m.group('width')), + 'height': int(m.group('height')), + 'abr': int(m.group('abr')), + 'vbr': int(m.group('vbr')), + 'filesize_approx': parse_filesize(m.group('filesize_approx')), + }) + formats.append(format) + self._sort_formats(formats) + return formats + def _real_extract(self, url): video_id = self._match_id(url) display_id = video_id.lstrip('-') @@ -94,14 +136,14 @@ class TagesschauIE(InfoExtractor): (?:,\s*quality:(?P["\'])(?P.+?)(?P=q_quality))? ''', playerpage): url = media.group('url') - type_ = media.group('type') + webpage_type = media.group('type') ext = media.group('ext') res = media.group('quality') f = { 'format_id': '%s_%s' % (res, ext) if res else ext, 'url': url, 'ext': ext, - 'vcodec': 'none' if type_ == 'audio' else None, + 'vcodec': 'none' if webpage_type == 'audio' else None, } f.update(self._FORMATS.get(res, {})) formats.append(f) @@ -109,47 +151,31 @@ class TagesschauIE(InfoExtractor): title = self._og_search_title(webpage).strip() description = self._og_search_description(webpage).strip() else: - download_text = self._search_regex( - r'(?s)

Wir bieten dieses Video in folgenden Formaten zum Download an:

\s*
(.*?)
\s*

', - webpage, 'download links') - links = re.finditer( - r'

', - download_text) - formats = [] - for l in links: - format_id = self._search_regex( - r'.*/[^/.]+\.([^/]+)\.[^/.]+', l.group('url'), 'format ID') - format = { - 'format_id': format_id, - 'url': l.group('url'), - 'format_name': l.group('name'), - } - m = re.match( - r'''(?x) - Video:\s*(?P[a-zA-Z0-9/._-]+)\s*&\#10; - (?P[0-9]+)x(?P[0-9]+)px&\#10; - (?P[0-9]+)kbps&\#10; - Audio:\s*(?P[0-9]+)kbps,\s*(?P[A-Za-z\.0-9]+)&\#10; - Größe:\s*(?P[0-9.,]+\s+[a-zA-Z]*B)''', - l.group('title')) - if m: - format.update({ - 'format_note': m.group('audio_desc'), - 'vcodec': m.group('vcodec'), - 'width': int(m.group('width')), - 'height': int(m.group('height')), - 'abr': int(m.group('abr')), - 'vbr': int(m.group('vbr')), - 'filesize_approx': parse_filesize(m.group('filesize_approx')), - }) - formats.append(format) - thumbnail = self._og_search_thumbnail(webpage) - description = self._html_search_regex( - r'(?s)

(.*?)

', - webpage, 'description', default=None) title = self._html_search_regex( r'(.*?)', webpage, 'title') + DOWNLOAD_REGEX = r'(?s)

Wir bieten dieses Video in folgenden Formaten zum Download an:

\s*
(.*?)
\s*

' + + webpage_type = self._og_search_property('type', webpage, default=None) + if webpage_type == 'website': # Article + entries = [] + for num, (entry_title, download_text) in enumerate(re.findall( + r'(?s)]+class="infotext"[^>]*>.*?(.+?).*?

.*?%s' % DOWNLOAD_REGEX, + webpage)): + entries.append({ + 'id': display_id, + 'title': '%s-%d' % (entry_title, num), + 'formats': self._extract_formats(download_text), + }) + return self.playlist_result(entries, display_id, title) + else: # Assume single video + download_text = self._search_regex(DOWNLOAD_REGEX, webpage, 'download links') + formats = self._extract_formats(download_text) + thumbnail = self._og_search_thumbnail(webpage) + description = self._html_search_regex( + r'(?s)

(.*?)

', + webpage, 'description', default=None) + self._sort_formats(formats) return {