diff options
author | Unknown <blackjack4494@web.de> | 2020-09-02 20:25:25 +0200 |
---|---|---|
committer | Unknown <blackjack4494@web.de> | 2020-09-02 20:25:25 +0200 |
commit | cefecac12cd3c70f9c7a30992c60b05c2eb5d34e (patch) | |
tree | f7b8e3f8ca2f6e402c83a501f72c09854ae04887 /youtube_dlc/extractor/googlesearch.py | |
parent | 9688f237163b6aa546fde00bb3fd1e3445dd4c31 (diff) | |
download | hypervideo-pre-cefecac12cd3c70f9c7a30992c60b05c2eb5d34e.tar.lz hypervideo-pre-cefecac12cd3c70f9c7a30992c60b05c2eb5d34e.tar.xz hypervideo-pre-cefecac12cd3c70f9c7a30992c60b05c2eb5d34e.zip |
[skip travis] renaming
to avoid using same folder when using pip install for example
Diffstat (limited to 'youtube_dlc/extractor/googlesearch.py')
-rw-r--r-- | youtube_dlc/extractor/googlesearch.py | 59 |
1 files changed, 59 insertions, 0 deletions
diff --git a/youtube_dlc/extractor/googlesearch.py b/youtube_dlc/extractor/googlesearch.py new file mode 100644 index 000000000..5279fa807 --- /dev/null +++ b/youtube_dlc/extractor/googlesearch.py @@ -0,0 +1,59 @@ +from __future__ import unicode_literals + +import itertools +import re + +from .common import SearchInfoExtractor + + +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(): + webpage = self._download_webpage( + 'http://www.google.com/search', + 'gvsearch:' + query, + note='Downloading result page %s' % (pagenum + 1), + query={ + 'tbm': 'vid', + 'q': query, + 'start': pagenum * 10, + 'hl': 'en', + }) + + 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 |