This commit is contained in:
Philipp Hagemeister 2012-03-15 07:56:32 +01:00
parent afbaa80b8b
commit 7e45ec57a8

View file

@ -2340,8 +2340,8 @@ class GoogleSearchIE(InfoExtractor):
"""Information Extractor for Google Video search queries.""" """Information Extractor for Google Video search queries."""
_VALID_URL = r'gvsearch(\d+|all)?:[\s\S]+' _VALID_URL = r'gvsearch(\d+|all)?:[\s\S]+'
_TEMPLATE_URL = 'http://video.google.com/videosearch?q=%s+site:video.google.com&start=%s&hl=en' _TEMPLATE_URL = 'http://video.google.com/videosearch?q=%s+site:video.google.com&start=%s&hl=en'
_VIDEO_INDICATOR = r'videoplay\?docid=([^\&>]+)\&' _VIDEO_INDICATOR = r'<a href="http://video\.google\.com/videoplay\?docid=([^"\&]+)'
_MORE_PAGES_INDICATOR = r'<span>Next</span>' _MORE_PAGES_INDICATOR = r'class="pn" id="pnnext"'
_google_ie = None _google_ie = None
_max_google_results = 1000 _max_google_results = 1000
IE_NAME = u'video.google:search' IE_NAME = u'video.google:search'
@ -2392,12 +2392,11 @@ class GoogleSearchIE(InfoExtractor):
"""Downloads a specified number of results for a query""" """Downloads a specified number of results for a query"""
video_ids = [] video_ids = []
already_seen = set() pagenum = 0
pagenum = 1
while True: while True:
self.report_download_page(query, pagenum) self.report_download_page(query, pagenum)
result_url = self._TEMPLATE_URL % (urllib.quote_plus(query), pagenum) result_url = self._TEMPLATE_URL % (urllib.quote_plus(query), pagenum*10)
request = urllib2.Request(result_url) request = urllib2.Request(result_url)
try: try:
page = urllib2.urlopen(request).read() page = urllib2.urlopen(request).read()
@ -2408,9 +2407,8 @@ class GoogleSearchIE(InfoExtractor):
# Extract video identifiers # Extract video identifiers
for mobj in re.finditer(self._VIDEO_INDICATOR, page): for mobj in re.finditer(self._VIDEO_INDICATOR, page):
video_id = mobj.group(1) video_id = mobj.group(1)
if video_id not in already_seen: if video_id not in video_ids:
video_ids.append(video_id) video_ids.append(video_id)
already_seen.add(video_id)
if len(video_ids) == n: if len(video_ids) == n:
# Specified n videos reached # Specified n videos reached
for id in video_ids: for id in video_ids: