[dailymotion] Extract all the available formats (closes #1028)

This commit is contained in:
Jaime Marquínez Ferrándiz 2013-10-23 17:33:38 +02:00
parent 2450bcb28b
commit cdec0190c4

View file

@ -28,6 +28,15 @@ class DailymotionIE(DailymotionBaseInfoExtractor, SubtitlesInfoExtractor):
_VALID_URL = r'(?i)(?:https?://)?(?:www\.)?dailymotion\.[a-z]{2,3}/(?:embed/)?video/([^/]+)' _VALID_URL = r'(?i)(?:https?://)?(?:www\.)?dailymotion\.[a-z]{2,3}/(?:embed/)?video/([^/]+)'
IE_NAME = u'dailymotion' IE_NAME = u'dailymotion'
_FORMATS = [
(u'stream_h264_ld_url', u'ld'),
(u'stream_h264_url', u'standard'),
(u'stream_h264_hq_url', u'hq'),
(u'stream_h264_hd_url', u'hd'),
(u'stream_h264_hd1080_url', u'hd180'),
]
_TESTS = [ _TESTS = [
{ {
u'url': u'http://www.dailymotion.com/video/x33vw9_tutoriel-de-youtubeur-dl-des-video_tech', u'url': u'http://www.dailymotion.com/video/x33vw9_tutoriel-de-youtubeur-dl-des-video_tech',
@ -60,7 +69,6 @@ def _real_extract(self, url):
video_id = mobj.group(1).split('_')[0].split('?')[0] video_id = mobj.group(1).split('_')[0].split('?')[0]
video_extension = 'mp4'
url = 'http://www.dailymotion.com/video/%s' % video_id url = 'http://www.dailymotion.com/video/%s' % video_id
# Retrieve video webpage to extract further information # Retrieve video webpage to extract further information
@ -99,18 +107,24 @@ def _real_extract(self, url):
msg = 'Couldn\'t get video, Dailymotion says: %s' % info['error']['title'] msg = 'Couldn\'t get video, Dailymotion says: %s' % info['error']['title']
raise ExtractorError(msg, expected=True) raise ExtractorError(msg, expected=True)
# TODO: support choosing qualities formats = []
for (key, format_id) in self._FORMATS:
for key in ['stream_h264_hd1080_url','stream_h264_hd_url', video_url = info.get(key)
'stream_h264_hq_url','stream_h264_url', if video_url is not None:
'stream_h264_ld_url']: m_size = re.search(r'H264-(\d+)x(\d+)', video_url)
if info.get(key):#key in info and info[key]: if m_size is not None:
max_quality = key width, height = m_size.group(1), m_size.group(2)
self.to_screen(u'Using %s' % key) else:
break width, height = None, None
else: formats.append({
'url': video_url,
'ext': 'mp4',
'format_id': format_id,
'width': width,
'height': height,
})
if not formats:
raise ExtractorError(u'Unable to extract video URL') raise ExtractorError(u'Unable to extract video URL')
video_url = info[max_quality]
# subtitles # subtitles
video_subtitles = self.extract_subtitles(video_id) video_subtitles = self.extract_subtitles(video_id)
@ -120,11 +134,10 @@ def _real_extract(self, url):
return [{ return [{
'id': video_id, 'id': video_id,
'url': video_url, 'formats': formats,
'uploader': video_uploader, 'uploader': video_uploader,
'upload_date': video_upload_date, 'upload_date': video_upload_date,
'title': self._og_search_title(webpage), 'title': self._og_search_title(webpage),
'ext': video_extension,
'subtitles': video_subtitles, 'subtitles': video_subtitles,
'thumbnail': info['thumbnail_url'] 'thumbnail': info['thumbnail_url']
}] }]