Switch Vimeo to scraping: fixes #285

This commit is contained in:
Filippo Valsorda 2012-03-05 17:51:16 +01:00
parent 0f6e296a8e
commit 2934c2ce43

View file

@ -2058,7 +2058,7 @@ class VimeoIE(InfoExtractor):
video_id = mobj.group(1) video_id = mobj.group(1)
# Retrieve video webpage to extract further information # Retrieve video webpage to extract further information
request = urllib2.Request("http://vimeo.com/moogaloop/load/clip:%s" % video_id, None, std_headers) request = urllib2.Request(url, None, std_headers)
try: try:
self.report_download_webpage(video_id) self.report_download_webpage(video_id)
webpage = urllib2.urlopen(request).read() webpage = urllib2.urlopen(request).read()
@ -2071,77 +2071,66 @@ class VimeoIE(InfoExtractor):
# and latter we extract those that are Vimeo specific. # and latter we extract those that are Vimeo specific.
self.report_extraction(video_id) self.report_extraction(video_id)
# Extract title # Extract the config JSON
mobj = re.search(r'<caption>(.*?)</caption>', webpage) config = webpage.split(' = {config:')[1].split(',assets:')[0]
if mobj is None: try:
self._downloader.trouble(u'ERROR: unable to extract video title') config = json.loads(config)
except:
self._downloader.trouble(u'ERROR: unable to extract info section')
return return
video_title = mobj.group(1).decode('utf-8')
# Extract title
video_title = config["video"]["title"]
simple_title = _simplify_title(video_title) simple_title = _simplify_title(video_title)
# Extract uploader # Extract uploader
mobj = re.search(r'<uploader_url>http://vimeo.com/(.*?)</uploader_url>', webpage) video_uploader = config["video"]["owner"]["name"]
if mobj is None:
self._downloader.trouble(u'ERROR: unable to extract video uploader')
return
video_uploader = mobj.group(1).decode('utf-8')
# Extract video thumbnail # Extract video thumbnail
mobj = re.search(r'<thumbnail>(.*?)</thumbnail>', webpage) video_thumbnail = config["video"]["thumbnail"]
if mobj is None:
self._downloader.trouble(u'ERROR: unable to extract video thumbnail')
return
video_thumbnail = mobj.group(1).decode('utf-8')
# # Extract video description # Extract video description
# mobj = re.search(r'<meta property="og:description" content="(.*)" />', webpage) try:
# if mobj is None: lxml.etree
# self._downloader.trouble(u'ERROR: unable to extract video description') except NameError:
# return video_description = u'No description available.'
# video_description = mobj.group(1).decode('utf-8') mobj = re.search(r'<meta name="description" content="(.*?)" />', webpage, re.MULTILINE)
# if not video_description: video_description = 'No description available.' if mobj is not None:
video_description = 'Foo.' video_description = mobj.group(1)
else:
html_parser = lxml.etree.HTMLParser()
vwebpage_doc = lxml.etree.parse(StringIO.StringIO(webpage), html_parser)
video_description = u''.join(vwebpage_doc.xpath('id("description")//text()')).strip()
# TODO use another parser
# Vimeo specific: extract request signature # Extract upload date
mobj = re.search(r'<request_signature>(.*?)</request_signature>', webpage) video_upload_date = u'NA'
if mobj is None: mobj = re.search(r'<span id="clip-date" style="display:none">[^:]*: (.*?)( \([^\(]*\))?</span>', webpage)
self._downloader.trouble(u'ERROR: unable to extract request signature') if mobj is not None:
return video_upload_date = mobj.group(1)
sig = mobj.group(1).decode('utf-8')
# Vimeo specific: extract request signature and timestamp
sig = config['request']['signature']
timestamp = config['request']['timestamp']
# Vimeo specific: extract video quality information # Vimeo specific: extract video quality information
mobj = re.search(r'<isHD>(\d+)</isHD>', webpage) # TODO bind to format param
if mobj is None: if 'hd' in config["video"]["files"]["h264"]: quality = 'hd'
self._downloader.trouble(u'ERROR: unable to extract video quality information') else: quality = 'sd'
return
quality = mobj.group(1).decode('utf-8')
if int(quality) == 1: video_url = "http://player.vimeo.com/play_redirect?clip_id=%s&sig=%s&time=%s&quality=%s&codecs=H264&type=moogaloop_local&embed_location=" \
quality = 'hd' %(video_id, sig, timestamp, quality)
else:
quality = 'sd'
# Vimeo specific: Extract request signature expiration
mobj = re.search(r'<request_signature_expires>(.*?)</request_signature_expires>', webpage)
if mobj is None:
self._downloader.trouble(u'ERROR: unable to extract request signature expiration')
return
sig_exp = mobj.group(1).decode('utf-8')
video_url = "http://vimeo.com/moogaloop/play/clip:%s/%s/%s/?q=%s" % (video_id, sig, sig_exp, quality)
try: try:
# Process video information # Process video information
self._downloader.process_info({ self._downloader.process_info({
'id': video_id.decode('utf-8'), 'id': video_id,
'url': video_url, 'url': video_url,
'uploader': video_uploader, 'uploader': video_uploader,
'upload_date': u'NA', 'upload_date': video_upload_date,
'title': video_title, 'title': video_title,
'stitle': simple_title, 'stitle': simple_title,
'ext': u'mp4', 'ext': u'mp4',
'thumbnail': video_thumbnail.decode('utf-8'),
'description': video_description,
'thumbnail': video_thumbnail, 'thumbnail': video_thumbnail,
'description': video_description, 'description': video_description,
'player_url': None, 'player_url': None,