diff options
Diffstat (limited to 'hypervideo_dl/extractor/googlesearch.py')
| -rw-r--r-- | hypervideo_dl/extractor/googlesearch.py | 21 |
1 files changed, 9 insertions, 12 deletions
diff --git a/hypervideo_dl/extractor/googlesearch.py b/hypervideo_dl/extractor/googlesearch.py index f605c0c..4b8b1bc 100644 --- a/hypervideo_dl/extractor/googlesearch.py +++ b/hypervideo_dl/extractor/googlesearch.py @@ -8,36 +8,33 @@ from .common import SearchInfoExtractor class GoogleSearchIE(SearchInfoExtractor): IE_DESC = 'Google Video search' - _MAX_RESULTS = 1000 IE_NAME = 'video.google:search' _SEARCH_KEY = 'gvsearch' - _WORKING = False - _TEST = { + _TESTS = [{ 'url': 'gvsearch15:python language', 'info_dict': { 'id': 'python language', 'title': 'python language', }, 'playlist_count': 15, - } + }] + _PAGE_SIZE = 100 def _search_results(self, query): for pagenum in itertools.count(): webpage = self._download_webpage( - 'http://www.google.com/search', - 'gvsearch:' + query, - note='Downloading result page %s' % (pagenum + 1), + 'http://www.google.com/search', f'gvsearch:{query}', + note=f'Downloading result page {pagenum + 1}', query={ 'tbm': 'vid', 'q': query, - 'start': pagenum * 10, + 'start': pagenum * self._PAGE_SIZE, + 'num': self._PAGE_SIZE, 'hl': 'en', }) - for hit_idx, mobj in enumerate(re.finditer( - r'<h3 class="r"><a href="([^"]+)"', webpage)): - if re.search(f'id="vidthumb{hit_idx + 1}"', webpage): - yield self.url_result(mobj.group(1)) + for url in re.findall(r'<div[^>]* class="dXiKIc"[^>]*><a href="([^"]+)"', webpage): + yield self.url_result(url) if not re.search(r'id="pnnext"', webpage): return |