Madeorsk/youtube-dl
1
0
Fork 0
mirror of https://codeberg.org/polarisfm/youtube-dl synced 2024-11-26 10:24:33 +01:00
Code Issues Releases Wiki Activity

Support more than 100 videos for justin.tv

Browse Source
This commit is contained in:
Dave Vasilevsky 2012-12-16 04:05:39 -05:00
parent 0b40544f29
commit 2ab1c5ed1a
1 changed files with 42 additions and 21 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

61
youtube_dl/InfoExtractors.py
View File

@ -3637,37 +3637,27 @@ class NBAIE(InfoExtractor):
class JustinTVIE(InfoExtractor): class JustinTVIE(InfoExtractor):
"""Information extractor for justin.tv and twitch.tv""" """Information extractor for justin.tv and twitch.tv"""
# TODO: One broadcast may be split into multiple videos. The key
# 'broadcast_id' is the same for all parts, and 'broadcast_part'
# starts at 1 and increases. Can we treat all parts as one video?
# _VALID_URL = r"""^(?:http(?:s?)://)?www\.(?:justin|twitch)\.tv/ # _VALID_URL = r"""^(?:http(?:s?)://)?www\.(?:justin|twitch)\.tv/
# ([^/]+)(?:/b/([^/]+))?/?(?:#.*)?$""" # ([^/]+)(?:/b/([^/]+))?/?(?:#.*)?$"""
_VALID_URL = r'^http://www.twitch.tv/(.*)$' _VALID_URL = r'^http://www.twitch.tv/(.*)$'
IE_NAME = u'justin.tv' IE_NAME = u'justin.tv'
_max_justin_results = 1000
_justin_page_limit = 100
def report_extraction(self, file_id): def report_extraction(self, file_id):
"""Report information extraction.""" """Report information extraction."""
self._downloader.to_screen(u'[%s] %s: Extracting information' % (self.IE_NAME, file_id)) self._downloader.to_screen(u'[%s] %s: Extracting information' % (self.IE_NAME, file_id))
def _real_extract(self, url): # Return count of items, list of *valid* items
mobj = re.match(self._VALID_URL, url) def _parse_page(self, url):
if mobj is None: print url
self._downloader.trouble(u'ERROR: invalid URL: %s' % url)
return
api = 'http://api.justin.tv'
video_id = mobj.group(mobj.lastindex)
if mobj.lastindex == 1:
api += '/channel/archives/%s.json?limit=100'
else:
api += '/clip/show/%s.json'
api = api % (video_id,)
self.report_extraction(video_id)
# TODO: multiple pages
# TODO: One broadcast may be split into multiple videos. The key
# 'broadcast_id' is the same for all parts, and 'broadcast_part'
# starts at 1 and increases. Can we treat all parts as one video?
try: try:
urlh = compat_urllib_request.urlopen(api) urlh = compat_urllib_request.urlopen(url)
webpage_bytes = urlh.read() webpage_bytes = urlh.read()
webpage = webpage_bytes.decode('utf-8', 'ignore') webpage = webpage_bytes.decode('utf-8', 'ignore')
except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
@ -3689,4 +3679,35 @@ class JustinTVIE(InfoExtractor):
'upload_date': video_date, 'upload_date': video_date,
'ext': video_extension, 'ext': video_extension,
}) })
print len(response)
return (len(response), info)
def _real_extract(self, url):
mobj = re.match(self._VALID_URL, url)
if mobj is None:
self._downloader.trouble(u'ERROR: invalid URL: %s' % url)
return
api = 'http://api.justin.tv'
video_id = mobj.group(mobj.lastindex)
paged = False
if mobj.lastindex == 1:
paged = True
api += '/channel/archives/%s.json'
else:
api += '/clip/show/%s.json'
api = api % (video_id,)
self.report_extraction(video_id)
info = []
offset = 0
limit = self._justin_page_limit
while offset < self._max_justin_results:
page_url = api + ('?offset=%d&limit=%d' % (offset, limit))
page_count, page_info = self._parse_page(page_url)
info.extend(page_info)
if not paged or page_count != limit:
break
offset += limit
return info return info