0
0
Fork 0

implement fallbacks and defaults in _search_regex

main
Filippo Valsorda 2013-06-06 14:35:08 +02:00
parent ac3e9394e7
commit 468e2e926b
2 changed files with 48 additions and 41 deletions

View File

@ -191,19 +191,37 @@ class InfoExtractor(object):
video_info['title'] = playlist_title video_info['title'] = playlist_title
return video_info return video_info
def _search_regex(self, pattern, text, name, fatal=True, flags=0): def _search_regex(self, pattern, string, name, default=None, fatal=True, flags=0):
"""Extract a field from some text based on regex""" """
mobj = re.search(pattern, text, flags) Perform a regex search on the given string, using a single or a list of
if mobj is None and fatal: patterns returning the first matching group.
raise ExtractorError(u'Unable to extract %s; ' In case of failure return a default value or raise a WARNING or a
u'please report this issue on GitHub.' % name) ExtractorError, depending on fatal, specifying the field name.
elif mobj is None: """
self._downloader.report_warning(u'unable to extract %s; ' if isinstance(pattern, (str, compat_str, compiled_regex_type)):
u'please report this issue on GitHub.' % name) mobj = re.search(pattern, string, flags)
return None
else: else:
# return the first matched group 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) 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; '
u'please report this issue on GitHub.' % _name)
else:
self._downloader.report_warning(u'unable to extract %s; '
u'please report this issue on GitHub.' % _name)
return None
class SearchInfoExtractor(InfoExtractor): class SearchInfoExtractor(InfoExtractor):
""" """
@ -2820,12 +2838,8 @@ class StanfordOpenClassroomIE(InfoExtractor):
note='Downloading course info page', note='Downloading course info page',
errnote='Unable to download course info page') errnote='Unable to download course info page')
# TODO: implement default_value in search_regex info['title'] = self._search_regex('<h1>([^<]+)</h1>', coursepage, 'title', default=info['id'])
m = re.search('<h1>([^<]+)</h1>', coursepage) info['title'] = unescapeHTML(info['title'])
if m:
info['title'] = unescapeHTML(m.group(1))
else:
info['title'] = info['id']
info['description'] = self._search_regex('<description>([^<]+)</description>', info['description'] = self._search_regex('<description>([^<]+)</description>',
coursepage, u'description', fatal=False) coursepage, u'description', fatal=False)
@ -3108,12 +3122,8 @@ class GooglePlusIE(InfoExtractor):
# Extract title # Extract title
# Get the first line for title # Get the first line for title
# TODO: implement default_value in search_regex video_title = self._search_regex(r'<meta name\=\"Description\" content\=\"(.*?)[\n<"]',
video_title = u'NA' webpage, 'title', default=u'NA')
pattern = r'<meta name\=\"Description\" content\=\"(.*?)[\n<"]'
mobj = re.search(pattern, webpage)
if mobj:
video_title = mobj.group(1)
# Step 2, Stimulate clicking the image box to launch video # Step 2, Stimulate clicking the image box to launch video
video_page = self._search_regex('"(https\://plus\.google\.com/photos/.*?)",,"image/jpeg","video"\]', video_page = self._search_regex('"(https\://plus\.google\.com/photos/.*?)",,"image/jpeg","video"\]',
@ -3167,23 +3177,21 @@ class NBAIE(InfoExtractor):
video_url = u'http://ht-mobile.cdn.turner.com/nba/big' + video_id + '_nba_1280x720.mp4' video_url = u'http://ht-mobile.cdn.turner.com/nba/big' + video_id + '_nba_1280x720.mp4'
# TODO: implement default_value in search_regex
def _findProp(rexp, default=None):
m = re.search(rexp, webpage)
if m:
return unescapeHTML(m.group(1))
else:
return default
shortened_video_id = video_id.rpartition('/')[2] shortened_video_id = video_id.rpartition('/')[2]
title = _findProp(r'<meta property="og:title" content="(.*?)"', shortened_video_id).replace('NBA.com: ', '') title = self._search_regex(r'<meta property="og:title" content="(.*?)"',
webpage, 'title', default=shortened_video_id).replace('NBA.com: ', '')
uploader_date = self._search_regex(r'<b>Date:</b> (.*?)</div>', webpage, 'upload_date', fatal=False)
description = self._search_regex(r'<div class="description">(.*?)</h1>', webpage, 'description', fatal=False)
info = { info = {
'id': shortened_video_id, 'id': shortened_video_id,
'url': video_url, 'url': video_url,
'ext': 'mp4', 'ext': 'mp4',
'title': title, 'title': title,
'uploader_date': _findProp(r'<b>Date:</b> (.*?)</div>'), 'uploader_date': uploader_date,
'description': _findProp(r'<div class="description">(.*?)</h1>'), 'description': description,
} }
return [info] return [info]
@ -3335,13 +3343,9 @@ class FunnyOrDieIE(InfoExtractor):
webpage, u'video URL', flags=re.DOTALL) webpage, u'video URL', flags=re.DOTALL)
video_url = unescapeHTML(video_url) video_url = unescapeHTML(video_url)
# TODO: implement fallbacks in regex_search title = self._search_regex((r"<h1 class='player_page_h1'.*?>(?P<title>.*?)</h1>",
m = re.search(r"<h1 class='player_page_h1'.*?>(?P<title>.*?)</h1>", webpage, flags=re.DOTALL) r'<title>(?P<title>[^<]+?)</title>'), webpage, 'title', flags=re.DOTALL)
if not m: title = clean_html(title)
m = re.search(r'<title>(?P<title>[^<]+?)</title>', webpage)
if not m:
raise ExtractorError(u'Cannot find video title')
title = clean_html(m.group('title'))
video_description = self._search_regex(r'<meta property="og:description" content="(?P<desc>.*?)"', video_description = self._search_regex(r'<meta property="og:description" content="(?P<desc>.*?)"',
webpage, u'description', flags=re.DOTALL) webpage, u'description', flags=re.DOTALL)

View File

@ -154,6 +154,9 @@ def compat_ord(c):
if type(c) is int: return c if type(c) is int: return c
else: return ord(c) else: return ord(c)
# This is not clearly defined otherwise
compiled_regex_type = type(re.compile(''))
std_headers = { std_headers = {
'User-Agent': 'Mozilla/5.0 (X11; Linux x86_64; rv:10.0) Gecko/20100101 Firefox/10.0', 'User-Agent': 'Mozilla/5.0 (X11; Linux x86_64; rv:10.0) Gecko/20100101 Firefox/10.0',
'Accept-Charset': 'ISO-8859-1,utf-8;q=0.7,*;q=0.7', 'Accept-Charset': 'ISO-8859-1,utf-8;q=0.7,*;q=0.7',