[generic] Detect direct video links (Fixes #4149, #4313)

This commit is contained in:
Philipp Hagemeister 2014-11-26 10:44:39 +01:00
parent 63c9b2c1d9
commit 4e262a8838
2 changed files with 41 additions and 5 deletions

View File

@ -296,9 +296,11 @@ class InfoExtractor(object):
content = self._webpage_read_content(urlh, url_or_request, video_id, note, errnote, fatal)
return (content, urlh)
def _webpage_read_content(self, urlh, url_or_request, video_id, note=None, errnote=None, fatal=True):
def _webpage_read_content(self, urlh, url_or_request, video_id, note=None, errnote=None, fatal=True, prefix=None):
content_type = urlh.headers.get('Content-Type', '')
webpage_bytes = urlh.read()
if prefix is not None:
webpage_bytes = prefix + webpage_bytes
m = re.match(r'[a-zA-Z0-9_.-]+/[a-zA-Z0-9_.-]+\s*;\s*charset=(.+)', content_type)
if m:
encoding = m.group(1)

View File

@ -452,7 +452,23 @@ class GenericIE(InfoExtractor):
'title': 'Unity 8 desktop-mode windows on Mir! - Ubuntu Discourse',
},
'playlist_mincount': 2,
},
# Direct link with incorrect MIME type
{
'url': 'http://ftp.nluug.nl/video/nluug/2014-11-20_nj14/zaal-2/5_Lennart_Poettering_-_Systemd.webm',
'md5': '4ccbebe5f36706d85221f204d7eb5913',
'info_dict': {
'url': 'http://ftp.nluug.nl/video/nluug/2014-11-20_nj14/zaal-2/5_Lennart_Poettering_-_Systemd.webm',
'id': '5_Lennart_Poettering_-_Systemd',
'ext': 'webm',
'title': '5_Lennart_Poettering_-_Systemd',
'upload_date': '20141120',
},
'expected_warnings': [
'URL could be a direct video link, returning it as such.'
]
}
]
def report_following_redirect(self, new_url):
@ -606,10 +622,28 @@ class GenericIE(InfoExtractor):
if not self._downloader.params.get('test', False) and not is_intentional:
self._downloader.report_warning('Falling back on generic information extractor.')
if full_response:
webpage = self._webpage_read_content(full_response, url, video_id)
else:
webpage = self._download_webpage(url, video_id)
if not full_response:
full_response = self._request_webpage(url, video_id)
# Maybe it's a direct link to a video?
# Be careful not to download the whole thing!
first_bytes = full_response.read(512)
if not re.match(r'^\s*<', first_bytes.decode('utf-8', 'replace')):
self._downloader.report_warning(
'URL could be a direct video link, returning it as such.')
upload_date = unified_strdate(
head_response.headers.get('Last-Modified'))
return {
'id': video_id,
'title': os.path.splitext(url_basename(url))[0],
'direct': True,
'url': url,
'upload_date': upload_date,
}
webpage = self._webpage_read_content(
full_response, url, video_id, prefix=first_bytes)
self.report_extraction(video_id)
# Is it an RSS feed?