nojhan/weboob-devel
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity

better code and check youtube.com is in url

Browse source
This commit is contained in:
Roger Philibert 2010-04-15 01:31:51 +02:00
commit 07b54f47a0
2 changed files with 15 additions and 9 deletions
Download patch file Download diff file Expand all files Collapse all files

3
weboob/backends/youtube/backend.py
View file

@ -38,6 +38,9 @@ class YoutubeBackend(Backend, ICapVideoProvider):
def inner(self, *args, **kwargs): def inner(self, *args, **kwargs):
if not self.browser: if not self.browser:
self.browser = YoutubeBrowser() self.browser = YoutubeBrowser()
url = args[0]
if u'youtube.com' not in url:
return None
return func(self, *args, **kwargs) return func(self, *args, **kwargs)
return inner return inner

21
weboob/backends/youtube/browser.py
View file

@ -26,7 +26,7 @@ from weboob.tools.parser import LxmlHtmlParser
from .pages import VideoPage from .pages import VideoPage
class YoutubeBrowser(Browser): class YoutubeBrowser(Browser):
regex = re.compile(r'&t=([^ ,&]*)') video_signature_regex = re.compile(r'&t=([^ ,&]*)')
def __init__(self, *args, **kwargs): def __init__(self, *args, **kwargs):
kwargs['parser'] = LxmlHtmlParser() kwargs['parser'] = LxmlHtmlParser()
@ -38,13 +38,16 @@ class YoutubeBrowser(Browser):
return self.page.title return self.page.title
def get_video_url(self, page_url): def get_video_url(self, page_url):
result = self.openurl(page_url).read() def find_video_signature(data):
for _signature in re.finditer(self.regex, result): for video_signature in re.finditer(self.video_signature_regex, data):
signature = _signature.group(1) return video_signature.group(1)
break return None
data = self.openurl(page_url).read()
video_signature = find_video_signature(data)
m = re.match(r'http://.*\.youtube\.com/watch\?v=(.+)', page_url)
if m:
video_id = m.group(1)
url = 'http://www.youtube.com/get_video?video_id=%s&t=%s&fmt=18' % (video_id, video_signature)
return url
else: else:
return None return None
m = re.match(r'http://.*\.youtube\.com/watch\?v=(.+)', page_url)
video_id = m.group(1)
url = 'http://www.youtube.com/get_video?video_id=%s&t=%s&fmt=18' % (video_id, signature)
return url