diff options
Diffstat (limited to 'youtube_dl/extractor')
| -rw-r--r-- | youtube_dl/extractor/youtube.py | 44 |
1 files changed, 43 insertions, 1 deletions
diff --git a/youtube_dl/extractor/youtube.py b/youtube_dl/extractor/youtube.py index 4aef85d0c..fbe8d63a3 100644 --- a/youtube_dl/extractor/youtube.py +++ b/youtube_dl/extractor/youtube.py @@ -5,7 +5,7 @@ import netrc import re import socket -from .common import InfoExtractor +from .common import InfoExtractor, SearchInfoExtractor from ..utils import ( compat_http_client, compat_parse_qs, @@ -751,3 +751,45 @@ class YoutubeUserIE(InfoExtractor): urls = ['http://www.youtube.com/watch?v=%s' % video_id for video_id in video_ids] url_results = [self.url_result(url, 'Youtube') for url in urls] return [self.playlist_result(url_results, playlist_title = username)] + +class YoutubeSearchIE(SearchInfoExtractor): + """Information Extractor for YouTube search queries.""" + _API_URL = 'https://gdata.youtube.com/feeds/api/videos?q=%s&start-index=%i&max-results=50&v=2&alt=jsonc' + _MAX_RESULTS = 1000 + IE_NAME = u'youtube:search' + _SEARCH_KEY = 'ytsearch' + + def report_download_page(self, query, pagenum): + """Report attempt to download search page with given number.""" + self._downloader.to_screen(u'[youtube] query "%s": Downloading page %s' % (query, pagenum)) + + def _get_n_results(self, query, n): + """Get a specified number of results for a query""" + + video_ids = [] + pagenum = 0 + limit = n + + while (50 * pagenum) < limit: + self.report_download_page(query, pagenum+1) + result_url = self._API_URL % (compat_urllib_parse.quote_plus(query), (50*pagenum)+1) + request = compat_urllib_request.Request(result_url) + try: + data = compat_urllib_request.urlopen(request).read().decode('utf-8') + except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: + raise ExtractorError(u'Unable to download API page: %s' % compat_str(err)) + api_response = json.loads(data)['data'] + + if not 'items' in api_response: + raise ExtractorError(u'[youtube] No video results') + + new_ids = list(video['id'] for video in api_response['items']) + video_ids += new_ids + + limit = min(n, api_response['totalItems']) + pagenum += 1 + + if len(video_ids) > n: + video_ids = video_ids[:n] + videos = [self.url_result('http://www.youtube.com/watch?v=%s' % id, 'Youtube') for id in video_ids] + return self.playlist_result(videos, query) |