diff options
| -rw-r--r-- | test/tests.json | 10 | ||||
| -rw-r--r-- | youtube_dl/FileDownloader.py | 7 | ||||
| -rwxr-xr-x | youtube_dl/InfoExtractors.py | 41 | ||||
| -rw-r--r-- | youtube_dl/__init__.py | 1 |
4 files changed, 57 insertions, 2 deletions
diff --git a/test/tests.json b/test/tests.json index 0c94c65bd..0c3b24054 100644 --- a/test/tests.json +++ b/test/tests.json @@ -328,5 +328,15 @@ "info_dict": { "title": "Video: KO Of The Week: MMA Fighter Gets Knocked Out By Swift Head Kick! " } + }, + { + "name": "ARD", + "url": "http://www.ardmediathek.de/das-erste/tagesschau-in-100-sek?documentId=14077640", + "file": "14077640.mp4", + "md5": "6ca8824255460c787376353f9e20bbd8", + "info_dict": { + "title": "11.04.2013 09:23 Uhr - Tagesschau in 100 Sekunden" + } } + ] diff --git a/youtube_dl/FileDownloader.py b/youtube_dl/FileDownloader.py index 7c5a52be1..e801db00a 100644 --- a/youtube_dl/FileDownloader.py +++ b/youtube_dl/FileDownloader.py @@ -629,7 +629,7 @@ class FileDownloader(object): except (IOError, OSError): self.report_warning(u'Unable to remove downloaded video file') - def _download_with_rtmpdump(self, filename, url, player_url, page_url): + def _download_with_rtmpdump(self, filename, url, player_url, page_url, play_path): self.report_destination(filename) tmpfilename = self.temp_name(filename) @@ -648,6 +648,8 @@ class FileDownloader(object): basic_args += ['-W', player_url] if page_url is not None: basic_args += ['--pageUrl', page_url] + if play_path is not None: + basic_args += ['-y', play_path] args = basic_args + [[], ['-e', '-k', '1']][self.params.get('continuedl', False)] if self.params.get('verbose', False): try: @@ -702,7 +704,8 @@ class FileDownloader(object): if url.startswith('rtmp'): return self._download_with_rtmpdump(filename, url, info_dict.get('player_url', None), - info_dict.get('page_url', None)) + info_dict.get('page_url', None), + info_dict.get('play_path', None)) tmpfilename = self.temp_name(filename) stream = None diff --git a/youtube_dl/InfoExtractors.py b/youtube_dl/InfoExtractors.py index aa8074a9e..51aca5497 100755 --- a/youtube_dl/InfoExtractors.py +++ b/youtube_dl/InfoExtractors.py @@ -4356,6 +4356,46 @@ class LiveLeakIE(InfoExtractor): return [info] +class ARDIE(InfoExtractor): + _VALID_URL = r'^(?:https?://)?(?:(?:www\.)?ardmediathek\.de|mediathek\.daserste\.de)/(?:.*/)(?P<video_id>[^/\?]+)(?:\?.*)?' + _TITLE = r'<h1(?: class="boxTopHeadline")?>(?P<title>.*)</h1>' + _MEDIA_STREAM = r'mediaCollection\.addMediaStream\((?P<media_type>\d+), (?P<quality>\d+), "(?P<rtmp_url>[^"]*)", "(?P<video_url>[^"]*)", "[^"]*"\)' + + def _real_extract(self, url): + # determine video id from url + m = re.match(self._VALID_URL, url) + + numid = re.search(r'documentId=([0-9]+)', url) + if numid: + video_id = numid.group(1) + else: + video_id = m.group('video_id') + + # determine title and media streams from webpage + html = self._download_webpage(url, video_id) + title = re.search(self._TITLE, html).group('title') + streams = [m.groupdict() for m in re.finditer(self._MEDIA_STREAM, html)] + if not streams: + assert '"fsk"' in html + self._downloader.report_error(u'this video is only available after 8:00 pm') + return + + # choose default media type and highest quality for now + stream = max([s for s in streams if int(s["media_type"]) == 0], + key=lambda s: int(s["quality"])) + + # there's two possibilities: RTMP stream or HTTP download + info = {'id': video_id, 'title': title, 'ext': 'mp4'} + if stream['rtmp_url']: + self._downloader.to_screen(u'[%s] RTMP download detected' % self.IE_NAME) + assert stream['video_url'].startswith('mp4:') + info["url"] = stream["rtmp_url"] + info["play_path"] = stream['video_url'] + else: + assert stream["video_url"].endswith('.mp4') + info["url"] = stream["video_url"] + return [info] + def gen_extractors(): """ Return a list of an instance of every supported extractor. @@ -4409,5 +4449,6 @@ def gen_extractors(): MySpassIE(), SpiegelIE(), LiveLeakIE(), + ARDIE(), GenericIE() ] diff --git a/youtube_dl/__init__.py b/youtube_dl/__init__.py index 807b73541..6a1a9bb52 100644 --- a/youtube_dl/__init__.py +++ b/youtube_dl/__init__.py @@ -24,6 +24,7 @@ __authors__ = ( 'Jaime Marquínez Ferrándiz', 'Jeff Crouse', 'Osama Khalid', + 'Michael Walter', ) __license__ = 'Public Domain' |