youtube_dl/extractor/googlesearch.py

   1 from __future__ import unicode_literals
   2
   3 import itertools
   4 import re
   5
   6 from .common import SearchInfoExtractor
   7 from ..utils import (
   8     compat_urllib_parse,
   9 )
  10
  11
  12 class GoogleSearchIE(SearchInfoExtractor):
  13     IE_DESC = 'Google Video search'
  14     _MAX_RESULTS = 1000
  15     IE_NAME = 'video.google:search'
  16     _SEARCH_KEY = 'gvsearch'
  17
  18     def _get_n_results(self, query, n):
  19         """Get a specified number of results for a query"""
  20
  21         entries = []
  22         res = {
  23             '_type': 'playlist',
  24             'id': query,
  25             'title': query,
  26         }
  27
  28         for pagenum in itertools.count():
  29             result_url = (
  30                 'http://www.google.com/search?tbm=vid&q=%s&start=%s&hl=en'
  31                 % (compat_urllib_parse.quote_plus(query), pagenum * 10))
  32
  33             webpage = self._download_webpage(
  34                 result_url, 'gvsearch:' + query,
  35                 note='Downloading result page ' + str(pagenum + 1))
  36
  37             for hit_idx, mobj in enumerate(re.finditer(
  38                     r'<h3 class="r"><a href="([^"]+)"', webpage)):
  39
  40                 # Skip playlists
  41                 if not re.search(r'id="vidthumb%d"' % (hit_idx + 1), webpage):
  42                     continue
  43
  44                 entries.append({
  45                     '_type': 'url',
  46                     'url': mobj.group(1)
  47                 })
  48
  49             if (len(entries) >= n) or not re.search(r'id="pnnext"', webpage):
  50                 res['entries'] = entries[:n]
  51                 return res