from __future__ import unicode_literals import itertools import re from .common import SearchInfoExtractor from ..compat import ( compat_urllib_parse, ) class GoogleSearchIE(SearchInfoExtractor): IE_DESC = 'Google Video search' _MAX_RESULTS = 1000 IE_NAME = 'video.google:search' _SEARCH_KEY = 'gvsearch' _TEST = { 'url': 'gvsearch15:python language', 'info_dict': { 'id': 'python language', 'title': 'python language', }, 'playlist_count': 15, } def _get_n_results(self, query, n): """Get a specified number of results for a query""" entries = [] res = { '_type': 'playlist', 'id': query, 'title': query, } for pagenum in itertools.count(): result_url = ( 'http://www.google.com/search?tbm=vid&q=%s&start=%s&hl=en' % (compat_urllib_parse.quote_plus(query), pagenum * 10)) webpage = self._download_webpage( result_url, 'gvsearch:' + query, note='Downloading result page ' + str(pagenum + 1)) for hit_idx, mobj in enumerate(re.finditer( r'<h3 class="r"><a href="([^"]+)"', webpage)): # Skip playlists if not re.search(r'id="vidthumb%d"' % (hit_idx + 1), webpage): continue entries.append({ '_type': 'url', 'url': mobj.group(1) }) if (len(entries) >= n) or not re.search(r'id="pnnext"', webpage): res['entries'] = entries[:n] return res