X-Git-Url: http://git.cielonegro.org/gitweb.cgi?a=blobdiff_plain;f=youtube_dl%2FInfoExtractors.py;h=59f65aca37108945ac8e1c9aca0dbe9db227225b;hb=5c6760193199530da1e66a1e412b58e238786f51;hp=33ba0fdd1be218ae0a5e04348545defeb407b386;hpb=fb8f7280bce30d45009c429f0095a9d15cbc9de2;p=youtube-dl.git
diff --git a/youtube_dl/InfoExtractors.py b/youtube_dl/InfoExtractors.py
index 33ba0fdd1..59f65aca3 100755
--- a/youtube_dl/InfoExtractors.py
+++ b/youtube_dl/InfoExtractors.py
@@ -191,6 +191,47 @@ class InfoExtractor(object):
video_info['title'] = playlist_title
return video_info
+ def _search_regex(self, pattern, string, name, default=None, fatal=True, flags=0):
+ """
+ Perform a regex search on the given string, using a single or a list of
+ patterns returning the first matching group.
+ In case of failure return a default value or raise a WARNING or a
+ ExtractorError, depending on fatal, specifying the field name.
+ """
+ if isinstance(pattern, (str, compat_str, compiled_regex_type)):
+ mobj = re.search(pattern, string, flags)
+ else:
+ for p in pattern:
+ mobj = re.search(p, string, flags)
+ if mobj: break
+
+ if sys.stderr.isatty() and os.name != 'nt':
+ _name = u'\033[0;34m%s\033[0m' % name
+ else:
+ _name = name
+
+ if mobj:
+ # return the first matching group
+ return next(g for g in mobj.groups() if g is not None)
+ elif default is not None:
+ return default
+ elif fatal:
+ raise ExtractorError(u'Unable to extract %s' % _name)
+ else:
+ self._downloader.report_warning(u'unable to extract %s; '
+ u'please report this issue on GitHub.' % _name)
+ return None
+
+ def _html_search_regex(self, pattern, string, name, default=None, fatal=True, flags=0):
+ """
+ Like _search_regex, but strips HTML tags and unescapes entities.
+ """
+ res = self._search_regex(pattern, string, name, default, fatal, flags)
+ if res:
+ return clean_html(res).strip()
+ else:
+ return res
+
class SearchInfoExtractor(InfoExtractor):
"""
Base class for paged search queries extractors.
@@ -964,18 +1005,13 @@ class PhotobucketIE(InfoExtractor):
}]
# We try looking in other parts of the webpage
- mobj = re.search(r'', webpage)
- if mobj is None:
- raise ExtractorError(u'Unable to extract media URL')
- mediaURL = compat_urllib_parse.unquote(mobj.group(1))
-
- video_url = mediaURL
+ video_url = self._search_regex(r'',
+ webpage, u'video URL')
mobj = re.search(r'
(.*) video by (.*) - Photobucket', webpage)
if mobj is None:
raise ExtractorError(u'Unable to extract title')
video_title = mobj.group(1).decode('utf-8')
-
video_uploader = mobj.group(2).decode('utf-8')
return [{
@@ -1397,16 +1433,12 @@ class GenericIE(InfoExtractor):
# Site Name | Video Title
# Video Title - Tagline | Site Name
# and so on and so forth; it's just not practical
- mobj = re.search(r'(.*)', webpage)
- if mobj is None:
- raise ExtractorError(u'Unable to extract title')
- video_title = mobj.group(1)
+ video_title = self._html_search_regex(r'(.*)',
+ webpage, u'video title')
# video uploader is domain name
- mobj = re.match(r'(?:https?://)?([^/]*)/.*', url)
- if mobj is None:
- raise ExtractorError(u'Unable to extract title')
- video_uploader = mobj.group(1)
+ video_uploader = self._search_regex(r'(?:https?://)?([^/]*)/.*',
+ url, u'video uploader')
return [{
'id': video_id,
@@ -1544,7 +1576,7 @@ class YoutubePlaylistIE(InfoExtractor):
|
((?:PL|EC|UU)[0-9A-Za-z-_]{10,})
)"""
- _TEMPLATE_URL = 'https://gdata.youtube.com/feeds/api/playlists/%s?max-results=%i&start-index=%i&v=2&alt=json'
+ _TEMPLATE_URL = 'https://gdata.youtube.com/feeds/api/playlists/%s?max-results=%i&start-index=%i&v=2&alt=json&safeSearch=none'
_MAX_RESULTS = 50
IE_NAME = u'youtube:playlist'
@@ -1805,10 +1837,7 @@ class DepositFilesIE(InfoExtractor):
file_extension = os.path.splitext(file_url)[1][1:]
# Search for file title
- mobj = re.search(r'', webpage)
- if mobj is None:
- raise ExtractorError(u'Unable to extract title')
- file_title = mobj.group(1).decode('utf-8')
+ file_title = self._search_regex(r'', webpage, u'title')
return [{
'id': file_id.decode('utf-8'),
@@ -1902,10 +1931,8 @@ class FacebookIE(InfoExtractor):
video_duration = int(video_data['video_duration'])
thumbnail = video_data['thumbnail_src']
- m = re.search('
([^<]+)
', webpage)
- if not m:
- raise ExtractorError(u'Cannot find title in webpage')
- video_title = unescapeHTML(m.group(1))
+ video_title = self._html_search_regex('
([^<]+)
',
+ webpage, u'title')
info = {
'id': video_id,
@@ -2067,15 +2094,10 @@ class MyVideoIE(InfoExtractor):
self.report_extraction(video_id)
video_url = mobj.group(1) + '.flv'
- mobj = re.search('([^<]+)', webpage)
- if mobj is None:
- raise ExtractorError(u'Unable to extract title')
- video_title = mobj.group(1)
+ video_title = self._html_search_regex('([^<]+)',
+ webpage, u'title')
- mobj = re.search('[.](.+?)$', video_url)
- if mobj is None:
- raise ExtractorError(u'Unable to extract extention')
- video_ext = mobj.group(1)
+ video_ext = self._search_regex('[.](.+?)$', video_url, u'extension')
return [{
'id': video_id,
@@ -2123,25 +2145,23 @@ class MyVideoIE(InfoExtractor):
# extracting infos
self.report_extraction(video_id)
+ video_url = None
mobj = re.search('connectionurl=\'(.*?)\'', dec_data)
- if mobj is None:
- raise ExtractorError(u'unable to extract rtmpurl')
- video_rtmpurl = compat_urllib_parse.unquote(mobj.group(1))
- if 'myvideo2flash' in video_rtmpurl:
- self._downloader.report_warning(u'forcing RTMPT ...')
- video_rtmpurl = video_rtmpurl.replace('rtmpe://', 'rtmpt://')
-
- # extract non rtmp videos
- if (video_rtmpurl is None) or (video_rtmpurl == ''):
+ if mobj:
+ video_url = compat_urllib_parse.unquote(mobj.group(1))
+ if 'myvideo2flash' in video_url:
+ self._downloader.report_warning(u'forcing RTMPT ...')
+ video_url = video_url.replace('rtmpe://', 'rtmpt://')
+
+ if not video_url:
+ # extract non rtmp videos
mobj = re.search('path=\'(http.*?)\' source=\'(.*?)\'', dec_data)
if mobj is None:
raise ExtractorError(u'unable to extract url')
- video_rtmpurl = compat_urllib_parse.unquote(mobj.group(1)) + compat_urllib_parse.unquote(mobj.group(2))
+ video_url = compat_urllib_parse.unquote(mobj.group(1)) + compat_urllib_parse.unquote(mobj.group(2))
- mobj = re.search('source=\'(.*?)\'', dec_data)
- if mobj is None:
- raise ExtractorError(u'unable to extract swfobj')
- video_file = compat_urllib_parse.unquote(mobj.group(1))
+ video_file = self._search_regex('source=\'(.*?)\'', dec_data, u'video file')
+ video_file = compat_urllib_parse.unquote(video_file)
if not video_file.endswith('f4m'):
ppath, prefix = video_file.split('.')
@@ -2153,20 +2173,16 @@ class MyVideoIE(InfoExtractor):
video_filepath + video_file
).replace('.f4m', '.m3u8')
- mobj = re.search('swfobject.embedSWF\(\'(.+?)\'', webpage)
- if mobj is None:
- raise ExtractorError(u'unable to extract swfobj')
- video_swfobj = compat_urllib_parse.unquote(mobj.group(1))
+ video_swfobj = self._search_regex('swfobject.embedSWF\(\'(.+?)\'', webpage, u'swfobj')
+ video_swfobj = compat_urllib_parse.unquote(video_swfobj)
- mobj = re.search("
(.*?)
", webpage)
- if mobj is None:
- raise ExtractorError(u'unable to extract title')
- video_title = mobj.group(1)
+ video_title = self._html_search_regex("
(.*?)
",
+ webpage, u'title')
return [{
'id': video_id,
- 'url': video_rtmpurl,
- 'tc_url': video_rtmpurl,
+ 'url': video_url,
+ 'tc_url': video_url,
'uploader': None,
'upload_date': None,
'title': video_title,
@@ -2177,6 +2193,7 @@ class MyVideoIE(InfoExtractor):
'player_url': video_swfobj,
}]
+
class ComedyCentralIE(InfoExtractor):
"""Information extractor for The Daily Show and Colbert Report """
@@ -2358,19 +2375,25 @@ class EscapistIE(InfoExtractor):
showName = mobj.group('showname')
videoId = mobj.group('episode')
- self.report_extraction(showName)
- webPage = self._download_webpage(url, showName)
+ self.report_extraction(videoId)
+ webpage = self._download_webpage(url, videoId)
+
+ videoDesc = self._html_search_regex('(.*?)\s+-\s+XVID', webpage)
- if mobj is None:
- raise ExtractorError(u'Unable to extract video title')
- video_title = mobj.group(1)
-
+ video_title = self._html_search_regex(r'(.*?)\s+-\s+XVID',
+ webpage, u'title')
# Extract video thumbnail
- mobj = re.search(r'http://(?:img.*?\.)xvideos.com/videos/thumbs/[a-fA-F0-9]+/[a-fA-F0-9]+/[a-fA-F0-9]+/[a-fA-F0-9]+/([a-fA-F0-9.]+jpg)', webpage)
- if mobj is None:
- raise ExtractorError(u'Unable to extract video thumbnail')
- video_thumbnail = mobj.group(0)
+ video_thumbnail = self._search_regex(r'http://(?:img.*?\.)xvideos.com/videos/thumbs/[a-fA-F0-9]+/[a-fA-F0-9]+/[a-fA-F0-9]+/[a-fA-F0-9]+/([a-fA-F0-9.]+jpg)',
+ webpage, u'thumbnail', fatal=False)
info = {
'id': video_id,
@@ -2654,16 +2668,12 @@ class InfoQIE(InfoExtractor):
video_url = 'rtmpe://video.infoq.com/cfx/st/' + real_id
# Extract title
- mobj = re.search(r'contentTitle = "(.*?)";', webpage)
- if mobj is None:
- raise ExtractorError(u'Unable to extract video title')
- video_title = mobj.group(1)
+ video_title = self._search_regex(r'contentTitle = "(.*?)";',
+ webpage, u'title')
# Extract description
- video_description = u'No description available.'
- mobj = re.search(r'', webpage)
- if mobj is not None:
- video_description = mobj.group(1)
+ video_description = self._html_search_regex(r'',
+ webpage, u'description', fatal=False)
video_filename = video_url.split('/')[-1]
video_id, extension = video_filename.split('.')
@@ -2834,15 +2844,10 @@ class StanfordOpenClassroomIE(InfoExtractor):
note='Downloading course info page',
errnote='Unable to download course info page')
- m = re.search('