X-Git-Url: http://git.bitcoin.ninja/index.cgi?a=blobdiff_plain;ds=sidebyside;f=youtube_dl%2FInfoExtractors.py;h=e3998fbe88173ae1cb949d2f6256fa0fc06fb530;hb=2e5d60b7db7020b726cd54ee4cad8f2afbd1479d;hp=7545ae0b907bed1125e059fd6e12aabf57784619;hpb=25580f3251a8dcaa04a3edd9af328372c7caca4a;p=youtube-dl
diff --git a/youtube_dl/InfoExtractors.py b/youtube_dl/InfoExtractors.py
index 7545ae0b9..e3998fbe8 100755
--- a/youtube_dl/InfoExtractors.py
+++ b/youtube_dl/InfoExtractors.py
@@ -151,7 +151,7 @@ class YoutubeIE(InfoExtractor):
(?(1).+)? # if we found the ID, everything can follow
$"""
_LANG_URL = r'http://www.youtube.com/?hl=en&persist_hl=1&gl=US&persist_gl=1&opt_out_ackd=1'
- _LOGIN_URL = 'https://www.youtube.com/signup?next=/&gl=US&hl=en'
+ _LOGIN_URL = 'https://accounts.google.com/ServiceLogin'
_AGE_URL = 'http://www.youtube.com/verify_age?next_url=/&gl=US&hl=en'
_NEXT_URL_RE = r'[\?&]next_url=([^&]+)'
_NETRC_MACHINE = 'youtube'
@@ -228,23 +228,6 @@ class YoutubeIE(InfoExtractor):
"""Indicate the download will use the RTMP protocol."""
self._downloader.to_screen(u'[youtube] RTMP download detected')
- def _closed_captions_xml_to_srt(self, xml_string):
- srt = ''
- texts = re.findall(r'([^<]+)', xml_string, re.MULTILINE)
- # TODO parse xml instead of regex
- for n, (start, dur_tag, dur, caption) in enumerate(texts):
- if not dur: dur = '4'
- start = float(start)
- end = start + float(dur)
- start = "%02i:%02i:%02i,%03i" %(start/(60*60), start/60%60, start%60, start%1*1000)
- end = "%02i:%02i:%02i,%03i" %(end/(60*60), end/60%60, end%60, end%1*1000)
- caption = unescapeHTML(caption)
- caption = unescapeHTML(caption) # double cycle, intentional
- srt += str(n+1) + '\n'
- srt += start + ' --> ' + end + '\n'
- srt += caption + '\n\n'
- return srt
-
def _extract_subtitles(self, video_id):
self.report_video_subtitles_download(video_id)
request = compat_urllib_request.Request('http://video.google.com/timedtext?hl=en&type=list&v=%s' % video_id)
@@ -264,14 +247,20 @@ class YoutubeIE(InfoExtractor):
srt_lang = list(srt_lang_list.keys())[0]
if not srt_lang in srt_lang_list:
return (u'WARNING: no closed captions found in the specified language', None)
- request = compat_urllib_request.Request('http://www.youtube.com/api/timedtext?lang=%s&name=%s&v=%s' % (srt_lang, srt_lang_list[srt_lang], video_id))
+ params = compat_urllib_parse.urlencode({
+ 'lang': srt_lang,
+ 'name': srt_lang_list[srt_lang].encode('utf-8'),
+ 'v': video_id,
+ 'fmt': 'srt',
+ })
+ url = 'http://www.youtube.com/api/timedtext?' + params
try:
- srt_xml = compat_urllib_request.urlopen(request).read().decode('utf-8')
+ srt = compat_urllib_request.urlopen(url).read().decode('utf-8')
except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
return (u'WARNING: unable to download video subtitles: %s' % compat_str(err), None)
- if not srt_xml:
- return (u'WARNING: unable to download video subtitles', None)
- return (None, self._closed_captions_xml_to_srt(srt_xml))
+ if not srt:
+ return (u'WARNING: Did not fetch video subtitles', None)
+ return (None, srt)
def _print_formats(self, formats):
print('Available formats:')
@@ -315,19 +304,54 @@ class YoutubeIE(InfoExtractor):
if username is None:
return
+ request = compat_urllib_request.Request(self._LOGIN_URL)
+ try:
+ login_page = compat_urllib_request.urlopen(request).read().decode('utf-8')
+ except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
+ self._downloader.to_stderr(u'WARNING: unable to fetch login page: %s' % compat_str(err))
+ return
+
+ galx = None
+ dsh = None
+ match = re.search(re.compile(r']* name="loginForm"', login_results) is not None:
+ if re.search(r'(?i)]* id="gaia_loginform"', login_results) is not None:
self._downloader.to_stderr(u'WARNING: unable to log in: bad username or password')
return
except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
@@ -678,6 +702,7 @@ class DailymotionIE(InfoExtractor):
_VALID_URL = r'(?i)(?:https?://)?(?:www\.)?dailymotion\.[a-z]{2,3}/video/([^/]+)'
IE_NAME = u'dailymotion'
+ _WORKING = False
def __init__(self, downloader=None):
InfoExtractor.__init__(self, downloader)
@@ -973,7 +998,7 @@ class VimeoIE(InfoExtractor):
"""Information extractor for vimeo.com."""
# _VALID_URL matches Vimeo URLs
- _VALID_URL = r'(?:https?://)?(?:(?:www|player).)?vimeo\.com/(?:(?:groups|album)/[^/]+/)?(?:videos?/)?([0-9]+)'
+ _VALID_URL = r'(?Phttps?://)?(?:(?:www|player)\.)?vimeo\.com/(?:(?:groups|album)/[^/]+/)?(?Pplay_redirect_hls\?clip_id=)?(?:videos?/)?(?P[0-9]+)'
IE_NAME = u'vimeo'
def __init__(self, downloader=None):
@@ -994,7 +1019,11 @@ class VimeoIE(InfoExtractor):
self._downloader.trouble(u'ERROR: Invalid URL: %s' % url)
return
- video_id = mobj.group(1)
+ video_id = mobj.group('id')
+ if not mobj.group('proto'):
+ url = 'https://' + url
+ if mobj.group('direct_link'):
+ url = 'https://vimeo.com/' + video_id
# Retrieve video webpage to extract further information
request = compat_urllib_request.Request(url, None, std_headers)
@@ -1285,7 +1314,7 @@ class GenericIE(InfoExtractor):
opener = compat_urllib_request.OpenerDirector()
for handler in [compat_urllib_request.HTTPHandler, compat_urllib_request.HTTPDefaultErrorHandler,
HTTPMethodFallback, HEADRedirectHandler,
- compat_urllib_error.HTTPErrorProcessor, compat_urllib_request.HTTPSHandler]:
+ compat_urllib_request.HTTPErrorProcessor, compat_urllib_request.HTTPSHandler]:
opener.add_handler(handler())
response = opener.open(HeadRequest(url))
@@ -1321,6 +1350,9 @@ class GenericIE(InfoExtractor):
if mobj is None:
# Broaden the search a little bit
mobj = re.search(r'[^A-Za-z0-9]?(?:file|source)=(http[^\'"&]*)', webpage)
+ if mobj is None:
+ # Broaden the search a little bit: JWPlayer JS loader
+ mobj = re.search(r'[^A-Za-z0-9]?file:\s*["\'](http[^\'"&]*)', webpage)
if mobj is None:
self._downloader.trouble(u'ERROR: Invalid URL: %s' % url)
return
@@ -2050,8 +2082,14 @@ class FacebookIE(InfoExtractor):
if not m:
raise ExtractorError(u'Cannot parse data')
data = dict(json.loads(m.group(1)))
- video_url = compat_urllib_parse.unquote(data['hd_src'])
- video_duration = int(data['video_duration'])
+ params_raw = compat_urllib_parse.unquote(data['params'])
+ params = json.loads(params_raw)
+ video_url = params['hd_src']
+ if not video_url:
+ video_url = params['sd_src']
+ if not video_url:
+ raise ExtractorError(u'Cannot find video URL')
+ video_duration = int(params['video_duration'])
m = re.search('', webpage)
if not m:
@@ -2064,7 +2102,7 @@ class FacebookIE(InfoExtractor):
'url': video_url,
'ext': 'mp4',
'duration': video_duration,
- 'thumbnail': data['thumbnail_src'],
+ 'thumbnail': params['thumbnail_src'],
}
return [info]
@@ -2186,7 +2224,7 @@ class MyVideoIE(InfoExtractor):
webpage = self._download_webpage(webpage_url, video_id)
self.report_extraction(video_id)
- mobj = re.search(r'',
+ mobj = re.search(r'',
webpage)
if mobj is None:
self._downloader.trouble(u'ERROR: unable to extract media URL')
@@ -2869,8 +2907,7 @@ class StanfordOpenClassroomIE(InfoExtractor):
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
+ raise ExtractorError(u'Invalid URL: %s' % url)
if mobj.group('course') and mobj.group('video'): # A specific video
course = mobj.group('course')
@@ -2907,12 +2944,9 @@ class StanfordOpenClassroomIE(InfoExtractor):
'upload_date': None,
}
- self.report_download_webpage(info['id'])
- try:
- coursepage = compat_urllib_request.urlopen(url).read()
- except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
- self._downloader.trouble(u'ERROR: unable to download course info page: ' + compat_str(err))
- return
+ coursepage = self._download_webpage(url, info['id'],
+ note='Downloading course info page',
+ errnote='Unable to download course info page')
m = re.search('([^<]+)
', coursepage)
if m:
@@ -2936,7 +2970,6 @@ class StanfordOpenClassroomIE(InfoExtractor):
assert entry['type'] == 'reference'
results += self.extract(entry['url'])
return results
-
else: # Root page
info = {
'id': 'Stanford OpenClassroom',
@@ -3683,13 +3716,13 @@ class YouPornIE(InfoExtractor):
webpage = self._download_webpage(req, video_id)
# Get the video title
- result = re.search(r'videoTitleArea">(?P.*)</h1>', webpage)
+ result = re.search(r'<h1.*?>(?P<title>.*)</h1>', webpage)
if result is None:
- raise ExtractorError(u'ERROR: unable to extract video title')
+ raise ExtractorError(u'Unable to extract video title')
video_title = result.group('title').strip()
# Get the video date
- result = re.search(r'Date:</b>(?P<date>.*)</li>', webpage)
+ result = re.search(r'Date:</label>(?P<date>.*) </li>', webpage)
if result is None:
self._downloader.to_stderr(u'WARNING: unable to extract video date')
upload_date = None
@@ -3697,9 +3730,9 @@ class YouPornIE(InfoExtractor):
upload_date = result.group('date').strip()
# Get the video uploader
- result = re.search(r'Submitted:</b>(?P<uploader>.*)</li>', webpage)
+ result = re.search(r'Submitted:</label>(?P<uploader>.*)</li>', webpage)
if result is None:
- self._downloader.to_stderr(u'ERROR: unable to extract uploader')
+ self._downloader.to_stderr(u'WARNING: unable to extract uploader')
video_uploader = None
else:
video_uploader = result.group('uploader').strip()
@@ -3871,7 +3904,7 @@ class EightTracksIE(InfoExtractor):
webpage = self._download_webpage(url, playlist_id)
- m = re.search(r"new TRAX.Mix\((.*?)\);\n*\s*TRAX.initSearchAutocomplete\('#search'\);", webpage, flags=re.DOTALL)
+ m = re.search(r"PAGE.mix = (.*?);\n", webpage, flags=re.DOTALL)
if not m:
raise ExtractorError(u'Cannot find trax information')
json_like = m.group(1)
@@ -3903,6 +3936,158 @@ class EightTracksIE(InfoExtractor):
next_url = 'http://8tracks.com/sets/%s/next?player=sm&mix_id=%s&format=jsonh&track_id=%s' % (session, mix_id, track_data['id'])
return res
+class KeekIE(InfoExtractor):
+ _VALID_URL = r'http://(?:www\.)?keek\.com/(?:!|\w+/keeks/)(?P<videoID>\w+)'
+ IE_NAME = u'keek'
+
+ def _real_extract(self, url):
+ m = re.match(self._VALID_URL, url)
+ video_id = m.group('videoID')
+ video_url = u'http://cdn.keek.com/keek/video/%s' % video_id
+ thumbnail = u'http://cdn.keek.com/keek/thumbnail/%s/w100/h75' % video_id
+ webpage = self._download_webpage(url, video_id)
+ m = re.search(r'<meta property="og:title" content="(?P<title>.+)"', webpage)
+ title = unescapeHTML(m.group('title'))
+ m = re.search(r'<div class="bio-names-and-report">[\s\n]+<h4>(?P<uploader>\w+)</h4>', webpage)
+ uploader = unescapeHTML(m.group('uploader'))
+ info = {
+ 'id':video_id,
+ 'url':video_url,
+ 'ext': 'mp4',
+ 'title': title,
+ 'thumbnail': thumbnail,
+ 'uploader': uploader
+ }
+ return [info]
+
+class TEDIE(InfoExtractor):
+ _VALID_URL=r'''http://www.ted.com/
+ (
+ ((?P<type_playlist>playlists)/(?P<playlist_id>\d+)) # We have a playlist
+ |
+ ((?P<type_talk>talks)) # We have a simple talk
+ )
+ /(?P<name>\w+) # Here goes the name and then ".html"
+ '''
+
+ def suitable(self, url):
+ """Receives a URL and returns True if suitable for this IE."""
+ return re.match(self._VALID_URL, url, re.VERBOSE) is not None
+
+ def _real_extract(self, url):
+ m=re.match(self._VALID_URL, url, re.VERBOSE)
+ if m.group('type_talk'):
+ return [self._talk_info(url)]
+ else :
+ playlist_id=m.group('playlist_id')
+ name=m.group('name')
+ self._downloader.to_screen(u'[%s] Getting info of playlist %s: "%s"' % (self.IE_NAME,playlist_id,name))
+ return self._playlist_videos_info(url,name,playlist_id)
+
+ def _talk_video_link(self,mediaSlug):
+ '''Returns the video link for that mediaSlug'''
+ return 'http://download.ted.com/talks/%s.mp4' % mediaSlug
+
+ def _playlist_videos_info(self,url,name,playlist_id=0):
+ '''Returns the videos of the playlist'''
+ video_RE=r'''
+ <li\ id="talk_(\d+)"([.\s]*?)data-id="(?P<video_id>\d+)"
+ ([.\s]*?)data-playlist_item_id="(\d+)"
+ ([.\s]*?)data-mediaslug="(?P<mediaSlug>.+?)"
+ '''
+ video_name_RE=r'<p\ class="talk-title"><a href="/talks/(.+).html">(?P<fullname>.+?)</a></p>'
+ webpage=self._download_webpage(url, playlist_id, 'Downloading playlist webpage')
+ m_videos=re.finditer(video_RE,webpage,re.VERBOSE)
+ m_names=re.finditer(video_name_RE,webpage)
+ info=[]
+ for m_video, m_name in zip(m_videos,m_names):
+ video_dic={
+ 'id': m_video.group('video_id'),
+ 'url': self._talk_video_link(m_video.group('mediaSlug')),
+ 'ext': 'mp4',
+ 'title': m_name.group('fullname')
+ }
+ info.append(video_dic)
+ return info
+ def _talk_info(self, url, video_id=0):
+ """Return the video for the talk in the url"""
+ m=re.match(self._VALID_URL, url,re.VERBOSE)
+ videoName=m.group('name')
+ webpage=self._download_webpage(url, video_id, 'Downloading \"%s\" page' % videoName)
+ # If the url includes the language we get the title translated
+ title_RE=r'<h1><span id="altHeadline" >(?P<title>[\s\w:/\.\?=\+-\\\']*)</span></h1>'
+ title=re.search(title_RE, webpage).group('title')
+ info_RE=r'''<script\ type="text/javascript">var\ talkDetails\ =(.*?)
+ "id":(?P<videoID>[\d]+).*?
+ "mediaSlug":"(?P<mediaSlug>[\w\d]+?)"'''
+ info_match=re.search(info_RE,webpage,re.VERBOSE)
+ video_id=info_match.group('videoID')
+ mediaSlug=info_match.group('mediaSlug')
+ video_url=self._talk_video_link(mediaSlug)
+ info = {
+ 'id': video_id,
+ 'url': video_url,
+ 'ext': 'mp4',
+ 'title': title
+ }
+ return info
+
+class MySpassIE(InfoExtractor):
+ _VALID_URL = r'http://www.myspass.de/.*'
+
+ def _real_extract(self, url):
+ META_DATA_URL_TEMPLATE = 'http://www.myspass.de/myspass/includes/apps/video/getvideometadataxml.php?id=%s'
+
+ # video id is the last path element of the URL
+ # usually there is a trailing slash, so also try the second but last
+ url_path = compat_urllib_parse_urlparse(url).path
+ url_parent_path, video_id = os.path.split(url_path)
+ if not video_id:
+ _, video_id = os.path.split(url_parent_path)
+
+ # get metadata
+ metadata_url = META_DATA_URL_TEMPLATE % video_id
+ metadata_text = self._download_webpage(metadata_url, video_id)
+ metadata = xml.etree.ElementTree.fromstring(metadata_text.encode('utf-8'))
+
+ # extract values from metadata
+ url_flv_el = metadata.find('url_flv')
+ if url_flv_el is None:
+ self._downloader.trouble(u'ERROR: unable to extract download url')
+ return
+ video_url = url_flv_el.text
+ extension = os.path.splitext(video_url)[1][1:]
+ title_el = metadata.find('title')
+ if title_el is None:
+ self._downloader.trouble(u'ERROR: unable to extract title')
+ return
+ title = title_el.text
+ format_id_el = metadata.find('format_id')
+ if format_id_el is None:
+ format = ext
+ else:
+ format = format_id_el.text
+ description_el = metadata.find('description')
+ if description_el is not None:
+ description = description_el.text
+ else:
+ description = None
+ imagePreview_el = metadata.find('imagePreview')
+ if imagePreview_el is not None:
+ thumbnail = imagePreview_el.text
+ else:
+ thumbnail = None
+ info = {
+ 'id': video_id,
+ 'url': video_url,
+ 'title': title,
+ 'ext': extension,
+ 'format': format,
+ 'thumbnail': thumbnail,
+ 'description': description
+ }
+ return [info]
+
def gen_extractors():
""" Return a list of an instance of every supported extractor.
The order does matter; the first extractor matched is the one handling the URL.
@@ -3949,6 +4134,9 @@ def gen_extractors():
UstreamIE(),
RBMARadioIE(),
EightTracksIE(),
+ KeekIE(),
+ TEDIE(),
+ MySpassIE(),
GenericIE()
]