X-Git-Url: http://git.cielonegro.org/gitweb.cgi?a=blobdiff_plain;f=youtube_dl%2Fextractor%2Fyoutube.py;h=35b95fdc2332f78288f3fdfc26fd271a5db8e353;hb=1c251cd9482bd0168ca844ad281317b5c19cd607;hp=d09128555f87f962339be4eb13f51d95c76e615f;hpb=c5e8d7af0ed867d70502491e3a80ee09b78ed2ce;p=youtube-dl.git

diff --git a/youtube_dl/extractor/youtube.py b/youtube_dl/extractor/youtube.py
index d09128555..35b95fdc2 100644
--- a/youtube_dl/extractor/youtube.py
+++ b/youtube_dl/extractor/youtube.py
@@ -1,12 +1,11 @@
 # coding: utf-8
-from __future__ import absolute_import
 
 import json
 import netrc
 import re
 import socket
 
-from .common import InfoExtractor
+from .common import InfoExtractor, SearchInfoExtractor
 from ..utils import (
     compat_http_client,
     compat_parse_qs,
@@ -59,7 +58,7 @@ class YoutubeIE(InfoExtractor):
         '18': 'mp4',
         '22': 'mp4',
         '37': 'mp4',
-        '38': 'video', # You actually don't know if this will be MOV, AVI or whatever
+        '38': 'mp4',
         '43': 'webm',
         '44': 'webm',
         '45': 'webm',
@@ -343,7 +342,7 @@ class YoutubeIE(InfoExtractor):
         request = compat_urllib_request.Request(self._AGE_URL, compat_urllib_parse.urlencode(age_form))
         try:
             self.report_age_confirmation()
-            age_results = compat_urllib_request.urlopen(request).read().decode('utf-8')
+            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 confirm age: %s' % compat_str(err))
 
@@ -471,7 +470,7 @@ class YoutubeIE(InfoExtractor):
                     self._downloader.report_warning(sub_error)
 
         if self._downloader.params.get('listsubtitles', False):
-            sub_lang_list = self._list_available_subtitles(video_id)
+            self._list_available_subtitles(video_id)
             return
 
         if 'length_seconds' not in video_info:
@@ -480,9 +479,6 @@ class YoutubeIE(InfoExtractor):
         else:
             video_duration = compat_urllib_parse.unquote_plus(video_info['length_seconds'][0])
 
-        # token
-        video_token = compat_urllib_parse.unquote_plus(video_info['token'][0])
-
         # Decide which formats to download
         req_format = self._downloader.params.get('format', None)
 
@@ -490,7 +486,7 @@ class YoutubeIE(InfoExtractor):
             mobj = re.search(r';ytplayer.config = ({.*?});', video_webpage)
             info = json.loads(mobj.group(1))
             args = info['args']
-            if args.get('ptk','') == 'vevo' or 'dashmpd':
+            if args.get('ptk','') == 'vevo' or 'dashmpd' in args:
                 # Vevo videos with encrypted signatures
                 self.to_screen(u'%s: Vevo video detected.' % video_id)
                 video_info['url_encoded_fmt_stream_map'] = [args['url_encoded_fmt_stream_map']]
@@ -755,3 +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)