Fix CollegHumorIE

Now it downloads the video over http in one file, it doesn't downloads in fragments
Added a test and use the methods in InfoExtractor for downloading webpages
This commit is contained in:
Jaime Marquínez Ferrándiz 2013-07-23 11:14:11 +02:00
parent ddbfd0f0c5
commit de48addae2

View File

@ -1,26 +1,26 @@
import re import re
import socket
import xml.etree.ElementTree import xml.etree.ElementTree
from .common import InfoExtractor from .common import InfoExtractor
from ..utils import ( from ..utils import (
compat_http_client,
compat_str,
compat_urllib_error,
compat_urllib_parse_urlparse, compat_urllib_parse_urlparse,
compat_urllib_request,
ExtractorError, ExtractorError,
) )
class CollegeHumorIE(InfoExtractor): class CollegeHumorIE(InfoExtractor):
_WORKING = False
_VALID_URL = r'^(?:https?://)?(?:www\.)?collegehumor\.com/video/(?P<videoid>[0-9]+)/(?P<shorttitle>.*)$' _VALID_URL = r'^(?:https?://)?(?:www\.)?collegehumor\.com/video/(?P<videoid>[0-9]+)/(?P<shorttitle>.*)$'
def report_manifest(self, video_id): _TEST = {
"""Report information extraction.""" u'url': u'http://www.collegehumor.com/video/6902724/comic-con-cosplay-catastrophe',
self.to_screen(u'%s: Downloading XML manifest' % video_id) u'file': u'6902724.mp4',
u'md5': u'1264c12ad95dca142a9f0bf7968105a0',
u'info_dict': {
u'title': u'Comic-Con Cosplay Catastrophe',
u'description': u'Fans get creative this year at San Diego. Too creative. And yes, that\'s really Joss Whedon.',
},
}
def _real_extract(self, url): def _real_extract(self, url):
mobj = re.match(self._VALID_URL, url) mobj = re.match(self._VALID_URL, url)
@ -36,10 +36,9 @@ def _real_extract(self, url):
self.report_extraction(video_id) self.report_extraction(video_id)
xmlUrl = 'http://www.collegehumor.com/moogaloop/video/' + video_id xmlUrl = 'http://www.collegehumor.com/moogaloop/video/' + video_id
try: metaXml = self._download_webpage(xmlUrl, video_id,
metaXml = compat_urllib_request.urlopen(xmlUrl).read() u'Downloading info XML',
except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: u'Unable to download video info XML')
raise ExtractorError(u'Unable to download video info XML: %s' % compat_str(err))
mdoc = xml.etree.ElementTree.fromstring(metaXml) mdoc = xml.etree.ElementTree.fromstring(metaXml)
try: try:
@ -52,11 +51,9 @@ def _real_extract(self, url):
raise ExtractorError(u'Invalid metadata XML file') raise ExtractorError(u'Invalid metadata XML file')
manifest_url += '?hdcore=2.10.3' manifest_url += '?hdcore=2.10.3'
self.report_manifest(video_id) manifestXml = self._download_webpage(manifest_url, video_id,
try: u'Downloading XML manifest',
manifestXml = compat_urllib_request.urlopen(manifest_url).read() u'Unable to download video info XML')
except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
raise ExtractorError(u'Unable to download video info XML: %s' % compat_str(err))
adoc = xml.etree.ElementTree.fromstring(manifestXml) adoc = xml.etree.ElementTree.fromstring(manifestXml)
try: try:
@ -66,9 +63,8 @@ def _real_extract(self, url):
except IndexError as err: except IndexError as err:
raise ExtractorError(u'Invalid manifest file') raise ExtractorError(u'Invalid manifest file')
url_pr = compat_urllib_parse_urlparse(manifest_url) url_pr = compat_urllib_parse_urlparse(info['thumbnail'])
url = url_pr.scheme + '://' + url_pr.netloc + '/z' + video_id[:-2] + '/' + node_id + 'Seg1-Frag1'
info['url'] = url info['url'] = url_pr.scheme + '://' + url_pr.netloc + video_id[:-2].replace('.csmil','').replace(',','')
info['ext'] = 'f4f' info['ext'] = 'mp4'
return [info] return [info]