w/youtube-dl
1
0
Fork 0
Code Issues Pull requests Releases Wiki Activity

[soundcloud] Prefer HTTP over RTMP (#1798)

Browse source
This commit is contained in:
Philipp Hagemeister 2013-11-25 20:30:41 +01:00
parent d46cc192d7
commit 2a15e7063b
1 changed files with 48 additions and 18 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

66
youtube_dl/extractor/soundcloud.py
View file

@ -76,44 +76,74 @@ class SoundcloudIE(InfoExtractor):
def _extract_info_dict(self, info, full_title=None, quiet=False): def _extract_info_dict(self, info, full_title=None, quiet=False):
track_id = compat_str(info['id']) track_id = compat_str(info['id'])
name = full_title or track_id name = full_title or track_id
if quiet == False: if quiet:
self.report_extraction(name) self.report_extraction(name)
thumbnail = info['artwork_url'] thumbnail = info['artwork_url']
if thumbnail is not None: if thumbnail is not None:
thumbnail = thumbnail.replace('-large', '-t500x500') thumbnail = thumbnail.replace('-large', '-t500x500')
ext = info.get('original_format', u'mp3')
result = { result = {
'id': track_id, 'id': track_id,
'uploader': info['user']['username'], 'uploader': info['user']['username'],
'upload_date': unified_strdate(info['created_at']), 'upload_date': unified_strdate(info['created_at']),
'title': info['title'], 'title': info['title'],
'ext': info.get('original_format', u'mp3'),
'description': info['description'], 'description': info['description'],
'thumbnail': thumbnail, 'thumbnail': thumbnail,
} }
if info.get('downloadable', False): if info.get('downloadable', False):
# We can build a direct link to the song # We can build a direct link to the song
result['url'] = 'https://api.soundcloud.com/tracks/{0}/download?client_id={1}'.format(track_id, self._CLIENT_ID) format_url = (
u'https://api.soundcloud.com/tracks/{0}/download?client_id={1}'.format(
track_id, self._CLIENT_ID))
result['formats'] = [{
'format_id': 'download',
'ext': ext,
'url': format_url,
}]
else: else:
# We have to retrieve the url # We have to retrieve the url
stream_json = self._download_webpage( stream_json = self._download_webpage(
'http://api.soundcloud.com/i1/tracks/{0}/streams?client_id={1}'.format(track_id, self._IPHONE_CLIENT_ID), 'http://api.soundcloud.com/i1/tracks/{0}/streams?client_id={1}'.format(track_id, self._IPHONE_CLIENT_ID),
track_id, u'Downloading track url') track_id, u'Downloading track url')
# There should be only one entry in the dictionary
key, stream_url = list(json.loads(stream_json).items())[0] formats = []
if key.startswith(u'http'): format_dict = json.loads(stream_json)
result['url'] = stream_url for key, stream_url in format_dict.items():
elif key.startswith(u'rtmp'): if key.startswith(u'http'):
# The url doesn't have an rtmp app, we have to extract the playpath formats.append({
url, path = stream_url.split('mp3:', 1) 'format_id': key,
result.update({ 'ext': ext,
'url': url, 'url': stream_url,
'play_path': 'mp3:' + path, })
}) elif key.startswith(u'rtmp'):
else: # The url doesn't have an rtmp app, we have to extract the playpath
url, path = stream_url.split('mp3:', 1)
formats.append({
'format_id': key,
'url': url,
'play_path': 'mp3:' + path,
'ext': ext,
})
if not formats:
# We fallback to the stream_url in the original info, this # We fallback to the stream_url in the original info, this
# cannot be always used, sometimes it can give an HTTP 404 error # cannot be always used, sometimes it can give an HTTP 404 error
result['url'] = info['stream_url'] + '?client_id=' + self._CLIENT_ID, formats.append({
'format_id': u'fallback',
'url': info['stream_url'] + '?client_id=' + self._CLIENT_ID,
'ext': ext,
})
def format_pref(f):
if f['format_id'].startswith('http'):
return 2
if f['format_id'].startswith('rtmp'):
return 1
return 0
formats.sort(key=format_pref)
result['formats'] = formats
return result return result