diff options
Diffstat (limited to 'youtube_dl/extractor/spiegel.py')
| -rw-r--r-- | youtube_dl/extractor/spiegel.py | 122 |
1 files changed, 94 insertions, 28 deletions
diff --git a/youtube_dl/extractor/spiegel.py b/youtube_dl/extractor/spiegel.py index 340a38440..f345883c7 100644 --- a/youtube_dl/extractor/spiegel.py +++ b/youtube_dl/extractor/spiegel.py @@ -4,10 +4,19 @@ from __future__ import unicode_literals import re from .common import InfoExtractor +from ..compat import ( + compat_urlparse, + compat_HTTPError, +) +from ..utils import ( + HEADRequest, + ExtractorError, +) +from .spiegeltv import SpiegeltvIE class SpiegelIE(InfoExtractor): - _VALID_URL = r'https?://(?:www\.)?spiegel\.de/video/[^/]*-(?P<videoID>[0-9]+)(?:\.html)?(?:#.*)?$' + _VALID_URL = r'https?://(?:www\.)?spiegel\.de/video/[^/]*-(?P<id>[0-9]+)(?:-embed)?(?:\.html)?(?:#.*)?$' _TESTS = [{ 'url': 'http://www.spiegel.de/video/vulkan-tungurahua-in-ecuador-ist-wieder-aktiv-video-1259285.html', 'md5': '2c2754212136f35fb4b19767d242f66e', @@ -29,25 +38,27 @@ class SpiegelIE(InfoExtractor): 'duration': 983, }, }, { - 'url': 'http://www.spiegel.de/video/johann-westhauser-videobotschaft-des-hoehlenforschers-video-1502367.html', - 'md5': '54f58ba0e752e3c07bc2a26222dd0acf', + 'url': 'http://www.spiegel.de/video/astronaut-alexander-gerst-von-der-iss-station-beantwortet-fragen-video-1519126-embed.html', + 'md5': 'd8eeca6bfc8f1cd6f490eb1f44695d51', 'info_dict': { - 'id': '1502367', + 'id': '1519126', 'ext': 'mp4', - 'title': 'Videobotschaft: Höhlenforscher Westhauser dankt seinen Rettern', - 'description': 'md5:c6f1ec11413ebd1088b6813943e5fc91', - 'duration': 42, - }, + 'description': 'SPIEGEL ONLINE-Nutzer durften den deutschen Astronauten Alexander Gerst über sein Leben auf der ISS-Station befragen. Hier kommen seine Antworten auf die besten sechs Fragen.', + 'title': 'Fragen an Astronaut Alexander Gerst: "Bekommen Sie die Tageszeiten mit?"', + } }] def _real_extract(self, url): - m = re.match(self._VALID_URL, url) - video_id = m.group('videoID') + video_id = self._match_id(url) + webpage, handle = self._download_webpage_handle(url, video_id) - webpage = self._download_webpage(url, video_id) + # 302 to spiegel.tv, like http://www.spiegel.de/video/der-film-zum-wochenende-die-wahrheit-ueber-maenner-video-99003272.html + if SpiegeltvIE.suitable(handle.geturl()): + return self.url_result(handle.geturl(), 'Spiegeltv') - title = self._html_search_regex( - r'<div class="module-title">(.*?)</div>', webpage, 'title') + title = re.sub(r'\s+', ' ', self._html_search_regex( + r'(?s)<(?:h1|div) class="module-title"[^>]*>(.*?)</(?:h1|div)>', + webpage, 'title')) description = self._html_search_meta('description', webpage, 'description') base_url = self._search_regex( @@ -56,21 +67,31 @@ class SpiegelIE(InfoExtractor): xml_url = base_url + video_id + '.xml' idoc = self._download_xml(xml_url, video_id) - formats = [ - { - 'format_id': n.tag.rpartition('type')[2], - 'url': base_url + n.find('./filename').text, - 'width': int(n.find('./width').text), - 'height': int(n.find('./height').text), - 'abr': int(n.find('./audiobitrate').text), - 'vbr': int(n.find('./videobitrate').text), - 'vcodec': n.find('./codec').text, - 'acodec': 'MP4A', - } - for n in list(idoc) - # Blacklist type 6, it's extremely LQ and not available on the same server - if n.tag.startswith('type') and n.tag != 'type6' - ] + formats = [] + for n in list(idoc): + if n.tag.startswith('type') and n.tag != 'type6': + format_id = n.tag.rpartition('type')[2] + video_url = base_url + n.find('./filename').text + # Test video URLs beforehand as some of them are invalid + try: + self._request_webpage( + HEADRequest(video_url), video_id, + 'Checking %s video URL' % format_id) + except ExtractorError as e: + if isinstance(e.cause, compat_HTTPError) and e.cause.code == 404: + self.report_warning( + '%s video URL is invalid, skipping' % format_id, video_id) + continue + formats.append({ + 'format_id': format_id, + 'url': video_url, + 'width': int(n.find('./width').text), + 'height': int(n.find('./height').text), + 'abr': int(n.find('./audiobitrate').text), + 'vbr': int(n.find('./videobitrate').text), + 'vcodec': n.find('./codec').text, + 'acodec': 'MP4A', + }) duration = float(idoc[0].findall('./duration')[0].text) self._sort_formats(formats) @@ -82,3 +103,48 @@ class SpiegelIE(InfoExtractor): 'duration': duration, 'formats': formats, } + + +class SpiegelArticleIE(InfoExtractor): + _VALID_URL = 'https?://www\.spiegel\.de/(?!video/)[^?#]*?-(?P<id>[0-9]+)\.html' + IE_NAME = 'Spiegel:Article' + IE_DESC = 'Articles on spiegel.de' + _TESTS = [{ + 'url': 'http://www.spiegel.de/sport/sonst/badminton-wm-die-randsportart-soll-populaerer-werden-a-987092.html', + 'info_dict': { + 'id': '1516455', + 'ext': 'mp4', + 'title': 'Faszination Badminton: Nennt es bloß nicht Federball', + 'description': 're:^Patrick Kämnitz gehört.{100,}', + }, + }, { + 'url': 'http://www.spiegel.de/wissenschaft/weltall/astronaut-alexander-gerst-antwortet-spiegel-online-lesern-a-989876.html', + 'info_dict': { + + }, + 'playlist_count': 6, + }] + + def _real_extract(self, url): + video_id = self._match_id(url) + webpage = self._download_webpage(url, video_id) + + # Single video on top of the page + video_link = self._search_regex( + r'<a href="([^"]+)" onclick="return spOpenVideo\(this,', webpage, + 'video page URL', default=None) + if video_link: + video_url = compat_urlparse.urljoin( + self.http_scheme() + '//spiegel.de/', video_link) + return self.url_result(video_url) + + # Multiple embedded videos + embeds = re.findall( + r'<div class="vid_holder[0-9]+.*?</div>\s*.*?url\s*=\s*"([^"]+)"', + webpage) + entries = [ + self.url_result(compat_urlparse.urljoin( + self.http_scheme() + '//spiegel.de/', embed_path)) + for embed_path in embeds + ] + return self.playlist_result(entries) |