]> Raphaël G. Git Repositories - youtubedl/commitdiff
Imported Upstream version 2010.06.06
authorRogério Brito <rbrito@ime.usp.br>
Sat, 18 Jun 2011 05:16:54 +0000 (02:16 -0300)
committerRogério Brito <rbrito@ime.usp.br>
Sat, 18 Jun 2011 05:16:54 +0000 (02:16 -0300)
.hg_archival.txt
.hgtags
LATEST_VERSION
youtube-dl

index be1abdffba66563ea0315f1f4dea8bd36cde4a21..3cd68f2587ebe9b255d4df2a9cf4698ddf32aac6 100644 (file)
@@ -1,2 +1,2 @@
 repo: f87cf8f2527c4adda57b14bd79a354f815164a41
 repo: f87cf8f2527c4adda57b14bd79a354f815164a41
-node: 1d3bca5d5e5dd4dc014c3d2681b7a429be9edd2f
+node: ebc3c1a8ea9490f2406b3f9ef9669556182dc2ce
diff --git a/.hgtags b/.hgtags
index 4697dc59e0bc95581dcd050fdc8e4f6396c17de6..e774bdd9d3f8b765a3f653739133158e754fc9b7 100644 (file)
--- a/.hgtags
+++ b/.hgtags
@@ -35,3 +35,7 @@ af285beaf15be8c4800d7bfdd868dd6058b14d65 2010.03.13
 eadec6d49e8a8d266c2ac6674aef2b415ccd3424 2010.03.13
 a40f32f2978ab2c20450da2a708b8ece806fe147 2010.04.02
 d6f421afc953fcc2d7336e7e4d6248b4d015e360 2010.04.03
 eadec6d49e8a8d266c2ac6674aef2b415ccd3424 2010.03.13
 a40f32f2978ab2c20450da2a708b8ece806fe147 2010.04.02
 d6f421afc953fcc2d7336e7e4d6248b4d015e360 2010.04.03
+1d3bca5d5e5dd4dc014c3d2681b7a429be9edd2f 2010.04.04
+2bf8bad7237fbfee1bd292d8114588ae1f24b9b2 2010.06.06
+2bf8bad7237fbfee1bd292d8114588ae1f24b9b2 2010.06.06
+0000000000000000000000000000000000000000 2010.06.06
index 013f034f30e7e771a9e68de3dc6677ff9d75e007..47865d378fc6bbddc1c48f4399c15568cdb2411a 100644 (file)
@@ -1 +1 @@
-2010.04.04
+2010.06.06
index 43566b6153e880200267d0a4291f6984b526a1c4..01a61ba12015d03e90659aa107430eab2ecf9acd 100755 (executable)
@@ -193,6 +193,7 @@ class FileDownloader(object):
        ignoreerrors:   Do not stop on download errors.
        ratelimit:      Download speed limit, in bytes/sec.
        nooverwrites:   Prevent overwriting files.
        ignoreerrors:   Do not stop on download errors.
        ratelimit:      Download speed limit, in bytes/sec.
        nooverwrites:   Prevent overwriting files.
+       retries:        Number of times to retry for HTTP error 503
        continuedl:     Try to continue downloads if possible.
        noprogress:     Do not print the progress bar.
        """
        continuedl:     Try to continue downloads if possible.
        noprogress:     Do not print the progress bar.
        """
@@ -364,6 +365,10 @@ class FileDownloader(object):
                """Report attemtp to resume at given byte."""
                self.to_stdout(u'[download] Resuming download at byte %s' % resume_len)
        
                """Report attemtp to resume at given byte."""
                self.to_stdout(u'[download] Resuming download at byte %s' % resume_len)
        
+       def report_retry(self, count, retries):
+               """Report retry in case of HTTP error 503"""
+               self.to_stdout(u'[download] Got HTTP error 503. Retrying (attempt %d of %d)...' % (count, retries))
+       
        def report_file_already_downloaded(self, file_name):
                """Report file has already been fully downloaded."""
                try:
        def report_file_already_downloaded(self, file_name):
                """Report file has already been fully downloaded."""
                try:
@@ -398,6 +403,10 @@ class FileDownloader(object):
                                print info_dict['title'].encode(preferredencoding(), 'xmlcharrefreplace')
                        if self.params.get('forceurl', False):
                                print info_dict['url'].encode(preferredencoding(), 'xmlcharrefreplace')
                                print info_dict['title'].encode(preferredencoding(), 'xmlcharrefreplace')
                        if self.params.get('forceurl', False):
                                print info_dict['url'].encode(preferredencoding(), 'xmlcharrefreplace')
+                       if self.params.get('forcethumbnail', False) and 'thumbnail' in info_dict:
+                               print info_dict['thumbnail'].encode(preferredencoding(), 'xmlcharrefreplace')
+                       if self.params.get('forcedescription', False) and 'description' in info_dict:
+                               print info_dict['description'].encode(preferredencoding(), 'xmlcharrefreplace')
 
                        return
                        
 
                        return
                        
@@ -419,7 +428,7 @@ class FileDownloader(object):
                        return
 
                try:
                        return
 
                try:
-                       success = self._do_download(filename, info_dict['url'].encode('utf-8'))
+                       success = self._do_download(filename, info_dict['url'].encode('utf-8'), info_dict.get('player_url', None))
                except (OSError, IOError), err:
                        raise UnavailableFormatError
                except (urllib2.URLError, httplib.HTTPException, socket.error), err:
                except (OSError, IOError), err:
                        raise UnavailableFormatError
                except (urllib2.URLError, httplib.HTTPException, socket.error), err:
@@ -471,7 +480,7 @@ class FileDownloader(object):
                        if info is None:
                                break
        
                        if info is None:
                                break
        
-       def _download_with_rtmpdump(self, filename, url):
+       def _download_with_rtmpdump(self, filename, url, player_url):
                self.report_destination(filename)
 
                # Check for rtmpdump first
                self.report_destination(filename)
 
                # Check for rtmpdump first
@@ -484,12 +493,16 @@ class FileDownloader(object):
                # Download using rtmpdump. rtmpdump returns exit code 2 when
                # the connection was interrumpted and resuming appears to be
                # possible. This is part of rtmpdump's normal usage, AFAIK.
                # Download using rtmpdump. rtmpdump returns exit code 2 when
                # the connection was interrumpted and resuming appears to be
                # possible. This is part of rtmpdump's normal usage, AFAIK.
-               basic_args = ['rtmpdump', '-q''-r', url, '-o', filename]
+               basic_args = ['rtmpdump', '-q'] + [[], ['-W', player_url]][player_url is not None] + ['-r', url, '-o', filename]
                retval = subprocess.call(basic_args + [[], ['-e', '-k', '1']][self.params.get('continuedl', False)])
                while retval == 2 or retval == 1:
                retval = subprocess.call(basic_args + [[], ['-e', '-k', '1']][self.params.get('continuedl', False)])
                while retval == 2 or retval == 1:
-                       self.to_stdout(u'\r[rtmpdump] %s bytes' % os.path.getsize(filename), skip_eol=True)
-                       time.sleep(2.0) # This seems to be needed
+                       prevsize = os.path.getsize(filename)
+                       self.to_stdout(u'\r[rtmpdump] %s bytes' % prevsize, skip_eol=True)
+                       time.sleep(5.0) # This seems to be needed
                        retval = subprocess.call(basic_args + ['-e'] + [[], ['-k', '1']][retval == 1])
                        retval = subprocess.call(basic_args + ['-e'] + [[], ['-k', '1']][retval == 1])
+                       cursize = os.path.getsize(filename)
+                       if prevsize == cursize and retval == 1:
+                               break
                if retval == 0:
                        self.to_stdout(u'\r[rtmpdump] %s bytes' % os.path.getsize(filename))
                        return True
                if retval == 0:
                        self.to_stdout(u'\r[rtmpdump] %s bytes' % os.path.getsize(filename))
                        return True
@@ -497,10 +510,10 @@ class FileDownloader(object):
                        self.trouble('\nERROR: rtmpdump exited with code %d' % retval)
                        return False
 
                        self.trouble('\nERROR: rtmpdump exited with code %d' % retval)
                        return False
 
-       def _do_download(self, filename, url):
+       def _do_download(self, filename, url, player_url):
                # Attempt to download using rtmpdump
                if url.startswith('rtmp'):
                # Attempt to download using rtmpdump
                if url.startswith('rtmp'):
-                       return self._download_with_rtmpdump(filename, url)
+                       return self._download_with_rtmpdump(filename, url, player_url)
 
                stream = None
                open_mode = 'wb'
 
                stream = None
                open_mode = 'wb'
@@ -519,24 +532,35 @@ class FileDownloader(object):
                        request.add_header('Range','bytes=%d-' % resume_len)
                        open_mode = 'ab'
 
                        request.add_header('Range','bytes=%d-' % resume_len)
                        open_mode = 'ab'
 
-               # Establish connection
-               try:
-                       data = urllib2.urlopen(request)
-               except (urllib2.HTTPError, ), err:
-                       if err.code != 416: #  416 is 'Requested range not satisfiable'
-                               raise
-                       # Unable to resume
-                       data = urllib2.urlopen(basic_request)
-                       content_length = data.info()['Content-Length']
-
-                       if content_length is not None and long(content_length) == resume_len:
-                               # Because the file had already been fully downloaded
-                               self.report_file_already_downloaded(filename)
-                               return True
-                       else:
-                               # Because the server didn't let us
-                               self.report_unable_to_resume()
-                               open_mode = 'wb'
+               count = 0
+               retries = self.params.get('retries', 0)
+               while True:
+                       # Establish connection
+                       try:
+                               data = urllib2.urlopen(request)
+                               break
+                       except (urllib2.HTTPError, ), err:
+                               if err.code == 503:
+                                       # Retry in case of HTTP error 503
+                                       count += 1
+                                       if count <= retries:
+                                               self.report_retry(count, retries)
+                                               continue
+                               if err.code != 416: #  416 is 'Requested range not satisfiable'
+                                       raise
+                               # Unable to resume
+                               data = urllib2.urlopen(basic_request)
+                               content_length = data.info()['Content-Length']
+
+                               if content_length is not None and long(content_length) == resume_len:
+                                       # Because the file had already been fully downloaded
+                                       self.report_file_already_downloaded(filename)
+                                       self._num_downloads += 1
+                                       return True
+                               else:
+                                       # Because the server didn't let us
+                                       self.report_unable_to_resume()
+                                       open_mode = 'wb'
 
                data_len = data.info().get('Content-length', None)
                data_len_str = self.format_bytes(data_len)
 
                data_len = data.info().get('Content-length', None)
                data_len_str = self.format_bytes(data_len)
@@ -562,7 +586,10 @@ class FileDownloader(object):
                                except (OSError, IOError), err:
                                        self.trouble('ERROR: unable to open for writing: %s' % str(err))
                                        return False
                                except (OSError, IOError), err:
                                        self.trouble('ERROR: unable to open for writing: %s' % str(err))
                                        return False
-                       stream.write(data_block)
+                       try:
+                               stream.write(data_block)
+                       except (IOError, OSError), err:
+                               self.trouble('\nERROR: unable to write data: %s' % str(err))
                        block_size = self.best_block_size(after - before, data_block_len)
 
                        # Progress message
                        block_size = self.best_block_size(after - before, data_block_len)
 
                        # Progress message
@@ -598,6 +625,15 @@ class InfoExtractor(object):
        stitle:         Simplified title.
        ext:            Video filename extension.
        format:         Video format.
        stitle:         Simplified title.
        ext:            Video filename extension.
        format:         Video format.
+       player_url:     SWF Player URL (may be None).
+
+       The following fields are optional. Their primary purpose is to allow
+       youtube-dl to serve as the backend for a video search function, such
+       as the one in youtube2mp3.  They are only used when their respective
+       forced printing functions are called:
+
+       thumbnail:      Full URL to a video thumbnail image.
+       description:    One-line video description.
 
        Subclasses of this one should re-define the _real_initialize() and
        _real_extract() methods, as well as the suitable() static method.
 
        Subclasses of this one should re-define the _real_initialize() and
        _real_extract() methods, as well as the suitable() static method.
@@ -656,6 +692,8 @@ class YoutubeIE(InfoExtractor):
                '18': 'mp4',
                '22': 'mp4',
                '37': 'mp4',
                '18': 'mp4',
                '22': 'mp4',
                '37': 'mp4',
+               '43': 'webm',
+               '45': 'webm',
        }
 
        @staticmethod
        }
 
        @staticmethod
@@ -674,6 +712,10 @@ class YoutubeIE(InfoExtractor):
                """Report attempt to confirm age."""
                self._downloader.to_stdout(u'[youtube] Confirming age')
        
                """Report attempt to confirm age."""
                self._downloader.to_stdout(u'[youtube] Confirming age')
        
+       def report_video_webpage_download(self, video_id):
+               """Report attempt to download video webpage."""
+               self._downloader.to_stdout(u'[youtube] %s: Downloading video webpage' % video_id)
+       
        def report_video_info_webpage_download(self, video_id):
                """Report attempt to download video info webpage."""
                self._downloader.to_stdout(u'[youtube] %s: Downloading video info webpage' % video_id)
        def report_video_info_webpage_download(self, video_id):
                """Report attempt to download video info webpage."""
                self._downloader.to_stdout(u'[youtube] %s: Downloading video info webpage' % video_id)
@@ -786,10 +828,26 @@ class YoutubeIE(InfoExtractor):
                        # Extension
                        video_extension = self._video_extensions.get(format_param, 'flv')
 
                        # Extension
                        video_extension = self._video_extensions.get(format_param, 'flv')
 
+                       # Get video webpage
+                       self.report_video_webpage_download(video_id)
+                       request = urllib2.Request('http://www.youtube.com/watch?v=%s&gl=US&hl=en' % video_id, None, std_headers)
+                       try:
+                               video_webpage = urllib2.urlopen(request).read()
+                       except (urllib2.URLError, httplib.HTTPException, socket.error), err:
+                               self._downloader.trouble(u'ERROR: unable to download video webpage: %s' % str(err))
+                               return
+
+                       # Attempt to extract SWF player URL
+                       mobj = re.search(r'swfConfig.*"(http://.*?watch-.*?\.swf)"', video_webpage)
+                       if mobj is not None:
+                               player_url = mobj.group(1)
+                       else:
+                               player_url = None
+
                        # Get video info
                        self.report_video_info_webpage_download(video_id)
                        # Get video info
                        self.report_video_info_webpage_download(video_id)
-                       for el_type in ['embedded', 'detailpage', 'vevo']:
-                               video_info_url = ('http://www.youtube.com/get_video_info?&video_id=%s&el=%s&ps=default&eurl=&gl=US&hl=en'
+                       for el_type in ['&el=embedded', '&el=detailpage', '&el=vevo', '']:
+                               video_info_url = ('http://www.youtube.com/get_video_info?&video_id=%s%s&ps=default&eurl=&gl=US&hl=en'
                                                   % (video_id, el_type))
                                request = urllib2.Request(video_info_url, None, std_headers)
                                try:
                                                   % (video_id, el_type))
                                request = urllib2.Request(video_info_url, None, std_headers)
                                try:
@@ -842,6 +900,20 @@ class YoutubeIE(InfoExtractor):
                        simple_title = re.sub(ur'(?u)([^%s]+)' % simple_title_chars, ur'_', video_title)
                        simple_title = simple_title.strip(ur'_')
 
                        simple_title = re.sub(ur'(?u)([^%s]+)' % simple_title_chars, ur'_', video_title)
                        simple_title = simple_title.strip(ur'_')
 
+                       # thumbnail image
+                       if 'thumbnail_url' not in video_info:
+                               self._downloader.trouble(u'WARNING: unable to extract video thumbnail')
+                               video_thumbnail = ''
+                       else:   # don't panic if we can't find it
+                               video_thumbnail = urllib.unquote_plus(video_info['thumbnail_url'][0])
+
+                       # description
+                       video_description = 'No description available.'
+                       if self._downloader.params.get('forcedescription', False):
+                               mobj = re.search(r'<meta name="description" content="(.*)"(?:\s*/)?>', video_webpage)
+                               if mobj is not None:
+                                       video_description = mobj.group(1)
+
                        try:
                                # Process video information
                                self._downloader.process_info({
                        try:
                                # Process video information
                                self._downloader.process_info({
@@ -852,34 +924,32 @@ class YoutubeIE(InfoExtractor):
                                        'stitle':       simple_title,
                                        'ext':          video_extension.decode('utf-8'),
                                        'format':       (format_param is None and u'NA' or format_param.decode('utf-8')),
                                        'stitle':       simple_title,
                                        'ext':          video_extension.decode('utf-8'),
                                        'format':       (format_param is None and u'NA' or format_param.decode('utf-8')),
+                                       'thumbnail':    video_thumbnail.decode('utf-8'),
+                                       'description':  video_description.decode('utf-8'),
+                                       'player_url':   player_url,
                                })
 
                                if all_formats:
                                })
 
                                if all_formats:
-                                       if quality_index == len(self._available_formats) - 1:
+                                       quality_index += 1
+                                       if quality_index == len(self._available_formats):
                                                # None left to get
                                                return
                                        else:
                                                # None left to get
                                                return
                                        else:
-                                               quality_index += 1
                                                format_param = self._available_formats[quality_index]
                                                format_param = self._available_formats[quality_index]
-                                               if format_param == None:
-                                                       return
                                                continue
                                                continue
-
                                return
 
                        except UnavailableFormatError, err:
                                if best_quality or all_formats:
                                return
 
                        except UnavailableFormatError, err:
                                if best_quality or all_formats:
-                                       if quality_index == len(self._available_formats) - 1:
+                                       quality_index += 1
+                                       if quality_index == len(self._available_formats):
                                                # I don't ever expect this to happen
                                                if not all_formats:
                                                        self._downloader.trouble(u'ERROR: no known formats available for video')
                                                return
                                        else:
                                                self.report_unavailable_format(video_id, format_param)
                                                # I don't ever expect this to happen
                                                if not all_formats:
                                                        self._downloader.trouble(u'ERROR: no known formats available for video')
                                                return
                                        else:
                                                self.report_unavailable_format(video_id, format_param)
-                                               quality_index += 1
                                                format_param = self._available_formats[quality_index]
                                                format_param = self._available_formats[quality_index]
-                                               if format_param == None:
-                                                       return
                                                continue
                                else: 
                                        self._downloader.trouble('ERROR: format not available for video')
                                                continue
                                else: 
                                        self._downloader.trouble('ERROR: format not available for video')
@@ -1009,6 +1079,7 @@ class MetacafeIE(InfoExtractor):
                                'stitle':       simple_title,
                                'ext':          video_extension.decode('utf-8'),
                                'format':       u'NA',
                                'stitle':       simple_title,
                                'ext':          video_extension.decode('utf-8'),
                                'format':       u'NA',
+                               'player_url':   None,
                        })
                except UnavailableFormatError:
                        self._downloader.trouble(u'ERROR: format not available for video')
                        })
                except UnavailableFormatError:
                        self._downloader.trouble(u'ERROR: format not available for video')
@@ -1080,6 +1151,32 @@ class GoogleIE(InfoExtractor):
                video_title = sanitize_title(video_title)
                simple_title = re.sub(ur'(?u)([^%s]+)' % simple_title_chars, ur'_', video_title)
 
                video_title = sanitize_title(video_title)
                simple_title = re.sub(ur'(?u)([^%s]+)' % simple_title_chars, ur'_', video_title)
 
+               # Extract video description
+               mobj = re.search(r'<span id=short-desc-content>([^<]*)</span>', webpage)
+               if mobj is None:
+                       self._downloader.trouble(u'ERROR: unable to extract video description')
+                       return
+               video_description = mobj.group(1).decode('utf-8')
+               if not video_description:
+                       video_description = 'No description available.'
+
+               # Extract video thumbnail
+               if self._downloader.params.get('forcethumbnail', False):
+                       request = urllib2.Request('http://video.google.com/videosearch?q=%s+site:video.google.com&hl=en' % abs(int(video_id)))
+                       try:
+                               webpage = urllib2.urlopen(request).read()
+                       except (urllib2.URLError, httplib.HTTPException, socket.error), err:
+                               self._downloader.trouble(u'ERROR: Unable to retrieve video webpage: %s' % str(err))
+                               return
+                       mobj = re.search(r'<img class=thumbnail-img (?:.* )?src=(http.*)>', webpage)
+                       if mobj is None:
+                               self._downloader.trouble(u'ERROR: unable to extract video thumbnail')
+                               return
+                       video_thumbnail = mobj.group(1)
+               else:   # we need something to pass to process_info
+                       video_thumbnail = ''
+
+
                try:
                        # Process video information
                        self._downloader.process_info({
                try:
                        # Process video information
                        self._downloader.process_info({
@@ -1090,6 +1187,7 @@ class GoogleIE(InfoExtractor):
                                'stitle':       simple_title,
                                'ext':          video_extension.decode('utf-8'),
                                'format':       u'NA',
                                'stitle':       simple_title,
                                'ext':          video_extension.decode('utf-8'),
                                'format':       u'NA',
+                               'player_url':   None,
                        })
                except UnavailableFormatError:
                        self._downloader.trouble(u'ERROR: format not available for video')
                        })
                except UnavailableFormatError:
                        self._downloader.trouble(u'ERROR: format not available for video')
@@ -1168,6 +1266,7 @@ class PhotobucketIE(InfoExtractor):
                                'stitle':       simple_title,
                                'ext':          video_extension.decode('utf-8'),
                                'format':       u'NA',
                                'stitle':       simple_title,
                                'ext':          video_extension.decode('utf-8'),
                                'format':       u'NA',
+                               'player_url':   None,
                        })
                except UnavailableFormatError:
                        self._downloader.trouble(u'ERROR: format not available for video')
                        })
                except UnavailableFormatError:
                        self._downloader.trouble(u'ERROR: format not available for video')
@@ -1258,6 +1357,21 @@ class YahooIE(InfoExtractor):
                        return
                video_uploader = mobj.group(1).decode('utf-8')
 
                        return
                video_uploader = mobj.group(1).decode('utf-8')
 
+               # Extract video thumbnail
+               mobj = re.search(r'<link rel="image_src" href="(.*)" />', webpage)
+               if mobj is None:
+                       self._downloader.trouble(u'ERROR: unable to extract video thumbnail')
+                       return
+               video_thumbnail = mobj.group(1).decode('utf-8')
+
+               # Extract video description
+               mobj = re.search(r'<meta name="description" content="(.*)" />', webpage)
+               if mobj is None:
+                       self._downloader.trouble(u'ERROR: unable to extract video description')
+                       return
+               video_description = mobj.group(1).decode('utf-8')
+               if not video_description: video_description = 'No description available.'
+
                # Extract video height and width
                mobj = re.search(r'<meta name="video_height" content="([0-9]+)" />', webpage)
                if mobj is None:
                # Extract video height and width
                mobj = re.search(r'<meta name="video_height" content="([0-9]+)" />', webpage)
                if mobj is None:
@@ -1303,6 +1417,11 @@ class YahooIE(InfoExtractor):
                                'title':        video_title,
                                'stitle':       simple_title,
                                'ext':          video_extension.decode('utf-8'),
                                'title':        video_title,
                                'stitle':       simple_title,
                                'ext':          video_extension.decode('utf-8'),
+                               'thumbnail':    video_thumbnail.decode('utf-8'),
+                               'description':  video_description,
+                               'thumbnail':    video_thumbnail,
+                               'description':  video_description,
+                               'player_url':   None,
                        })
                except UnavailableFormatError:
                        self._downloader.trouble(u'ERROR: format not available for video')
                        })
                except UnavailableFormatError:
                        self._downloader.trouble(u'ERROR: format not available for video')
@@ -1398,6 +1517,7 @@ class GenericIE(InfoExtractor):
                                'stitle':       simple_title,
                                'ext':          video_extension.decode('utf-8'),
                                'format':       u'NA',
                                'stitle':       simple_title,
                                'ext':          video_extension.decode('utf-8'),
                                'format':       u'NA',
+                               'player_url':   None,
                        })
                except UnavailableFormatError:
                        self._downloader.trouble(u'ERROR: format not available for video')
                        })
                except UnavailableFormatError:
                        self._downloader.trouble(u'ERROR: format not available for video')
@@ -1494,6 +1614,188 @@ class YoutubeSearchIE(InfoExtractor):
 
                        pagenum = pagenum + 1
 
 
                        pagenum = pagenum + 1
 
+class GoogleSearchIE(InfoExtractor):
+       """Information Extractor for Google Video search queries."""
+       _VALID_QUERY = r'gvsearch(\d+|all)?:[\s\S]+'
+       _TEMPLATE_URL = 'http://video.google.com/videosearch?q=%s+site:video.google.com&start=%s&hl=en'
+       _VIDEO_INDICATOR = r'videoplay\?docid=([^\&>]+)\&'
+       _MORE_PAGES_INDICATOR = r'<span>Next</span>'
+       _google_ie = None
+       _max_google_results = 1000
+
+       def __init__(self, google_ie, downloader=None):
+               InfoExtractor.__init__(self, downloader)
+               self._google_ie = google_ie
+       
+       @staticmethod
+       def suitable(url):
+               return (re.match(GoogleSearchIE._VALID_QUERY, url) is not None)
+
+       def report_download_page(self, query, pagenum):
+               """Report attempt to download playlist page with given number."""
+               query = query.decode(preferredencoding())
+               self._downloader.to_stdout(u'[video.google] query "%s": Downloading page %s' % (query, pagenum))
+
+       def _real_initialize(self):
+               self._google_ie.initialize()
+       
+       def _real_extract(self, query):
+               mobj = re.match(self._VALID_QUERY, query)
+               if mobj is None:
+                       self._downloader.trouble(u'ERROR: invalid search query "%s"' % query)
+                       return
+
+               prefix, query = query.split(':')
+               prefix = prefix[8:]
+               query  = query.encode('utf-8')
+               if prefix == '':
+                       self._download_n_results(query, 1)
+                       return
+               elif prefix == 'all':
+                       self._download_n_results(query, self._max_google_results)
+                       return
+               else:
+                       try:
+                               n = long(prefix)
+                               if n <= 0:
+                                       self._downloader.trouble(u'ERROR: invalid download number %s for query "%s"' % (n, query))
+                                       return
+                               elif n > self._max_google_results:
+                                       self._downloader.to_stderr(u'WARNING: gvsearch returns max %i results (you requested %i)'  % (self._max_google_results, n))
+                                       n = self._max_google_results
+                               self._download_n_results(query, n)
+                               return
+                       except ValueError: # parsing prefix as integer fails
+                               self._download_n_results(query, 1)
+                               return
+
+       def _download_n_results(self, query, n):
+               """Downloads a specified number of results for a query"""
+
+               video_ids = []
+               already_seen = set()
+               pagenum = 1
+
+               while True:
+                       self.report_download_page(query, pagenum)
+                       result_url = self._TEMPLATE_URL % (urllib.quote_plus(query), pagenum)
+                       request = urllib2.Request(result_url, None, std_headers)
+                       try:
+                               page = urllib2.urlopen(request).read()
+                       except (urllib2.URLError, httplib.HTTPException, socket.error), err:
+                               self._downloader.trouble(u'ERROR: unable to download webpage: %s' % str(err))
+                               return
+
+                       # Extract video identifiers
+                       for mobj in re.finditer(self._VIDEO_INDICATOR, page):
+                               video_id = mobj.group(1)
+                               if video_id not in already_seen:
+                                       video_ids.append(video_id)
+                                       already_seen.add(video_id)
+                                       if len(video_ids) == n:
+                                               # Specified n videos reached
+                                               for id in video_ids:
+                                                       self._google_ie.extract('http://video.google.com/videoplay?docid=%s' % id)
+                                               return
+
+                       if re.search(self._MORE_PAGES_INDICATOR, page) is None:
+                               for id in video_ids:
+                                       self._google_ie.extract('http://video.google.com/videoplay?docid=%s' % id)
+                               return
+
+                       pagenum = pagenum + 1
+
+class YahooSearchIE(InfoExtractor):
+       """Information Extractor for Yahoo! Video search queries."""
+       _VALID_QUERY = r'yvsearch(\d+|all)?:[\s\S]+'
+       _TEMPLATE_URL = 'http://video.yahoo.com/search/?p=%s&o=%s'
+       _VIDEO_INDICATOR = r'href="http://video\.yahoo\.com/watch/([0-9]+/[0-9]+)"'
+       _MORE_PAGES_INDICATOR = r'\s*Next'
+       _yahoo_ie = None
+       _max_yahoo_results = 1000
+
+       def __init__(self, yahoo_ie, downloader=None):
+               InfoExtractor.__init__(self, downloader)
+               self._yahoo_ie = yahoo_ie
+       
+       @staticmethod
+       def suitable(url):
+               return (re.match(YahooSearchIE._VALID_QUERY, url) is not None)
+
+       def report_download_page(self, query, pagenum):
+               """Report attempt to download playlist page with given number."""
+               query = query.decode(preferredencoding())
+               self._downloader.to_stdout(u'[video.yahoo] query "%s": Downloading page %s' % (query, pagenum))
+
+       def _real_initialize(self):
+               self._yahoo_ie.initialize()
+       
+       def _real_extract(self, query):
+               mobj = re.match(self._VALID_QUERY, query)
+               if mobj is None:
+                       self._downloader.trouble(u'ERROR: invalid search query "%s"' % query)
+                       return
+
+               prefix, query = query.split(':')
+               prefix = prefix[8:]
+               query  = query.encode('utf-8')
+               if prefix == '':
+                       self._download_n_results(query, 1)
+                       return
+               elif prefix == 'all':
+                       self._download_n_results(query, self._max_yahoo_results)
+                       return
+               else:
+                       try:
+                               n = long(prefix)
+                               if n <= 0:
+                                       self._downloader.trouble(u'ERROR: invalid download number %s for query "%s"' % (n, query))
+                                       return
+                               elif n > self._max_yahoo_results:
+                                       self._downloader.to_stderr(u'WARNING: yvsearch returns max %i results (you requested %i)'  % (self._max_yahoo_results, n))
+                                       n = self._max_yahoo_results
+                               self._download_n_results(query, n)
+                               return
+                       except ValueError: # parsing prefix as integer fails
+                               self._download_n_results(query, 1)
+                               return
+
+       def _download_n_results(self, query, n):
+               """Downloads a specified number of results for a query"""
+
+               video_ids = []
+               already_seen = set()
+               pagenum = 1
+
+               while True:
+                       self.report_download_page(query, pagenum)
+                       result_url = self._TEMPLATE_URL % (urllib.quote_plus(query), pagenum)
+                       request = urllib2.Request(result_url, None, std_headers)
+                       try:
+                               page = urllib2.urlopen(request).read()
+                       except (urllib2.URLError, httplib.HTTPException, socket.error), err:
+                               self._downloader.trouble(u'ERROR: unable to download webpage: %s' % str(err))
+                               return
+
+                       # Extract video identifiers
+                       for mobj in re.finditer(self._VIDEO_INDICATOR, page):
+                               video_id = mobj.group(1)
+                               if video_id not in already_seen:
+                                       video_ids.append(video_id)
+                                       already_seen.add(video_id)
+                                       if len(video_ids) == n:
+                                               # Specified n videos reached
+                                               for id in video_ids:
+                                                       self._yahoo_ie.extract('http://video.yahoo.com/watch/%s' % id)
+                                               return
+
+                       if re.search(self._MORE_PAGES_INDICATOR, page) is None:
+                               for id in video_ids:
+                                       self._yahoo_ie.extract('http://video.yahoo.com/watch/%s' % id)
+                               return
+
+                       pagenum = pagenum + 1
+
 class YoutubePlaylistIE(InfoExtractor):
        """Information Extractor for YouTube playlists."""
 
 class YoutubePlaylistIE(InfoExtractor):
        """Information Extractor for YouTube playlists."""
 
@@ -1686,7 +1988,7 @@ if __name__ == '__main__':
                # Parse command line
                parser = optparse.OptionParser(
                        usage='Usage: %prog [options] url...',
                # Parse command line
                parser = optparse.OptionParser(
                        usage='Usage: %prog [options] url...',
-                       version='2010.04.04',
+                       version='2010.06.06',
                        conflict_handler='resolve',
                )
 
                        conflict_handler='resolve',
                )
 
@@ -1700,6 +2002,8 @@ if __name__ == '__main__':
                                action='store_true', dest='ignoreerrors', help='continue on download errors', default=False)
                parser.add_option('-r', '--rate-limit',
                                dest='ratelimit', metavar='L', help='download rate limit (e.g. 50k or 44.6m)')
                                action='store_true', dest='ignoreerrors', help='continue on download errors', default=False)
                parser.add_option('-r', '--rate-limit',
                                dest='ratelimit', metavar='L', help='download rate limit (e.g. 50k or 44.6m)')
+               parser.add_option('-R', '--retries',
+                               dest='retries', metavar='T', help='number of retries (default is 10)', default=10)
 
                authentication = optparse.OptionGroup(parser, 'Authentication Options')
                authentication.add_option('-u', '--username',
 
                authentication = optparse.OptionGroup(parser, 'Authentication Options')
                authentication.add_option('-u', '--username',
@@ -1732,6 +2036,10 @@ if __name__ == '__main__':
                                action='store_true', dest='geturl', help='simulate, quiet but print URL', default=False)
                verbosity.add_option('-e', '--get-title',
                                action='store_true', dest='gettitle', help='simulate, quiet but print title', default=False)
                                action='store_true', dest='geturl', help='simulate, quiet but print URL', default=False)
                verbosity.add_option('-e', '--get-title',
                                action='store_true', dest='gettitle', help='simulate, quiet but print title', default=False)
+               verbosity.add_option('--get-thumbnail',
+                               action='store_true', dest='getthumbnail', help='simulate, quiet but print thumbnail URL', default=False)
+               verbosity.add_option('--get-description',
+                               action='store_true', dest='getdescription', help='simulate, quiet but print video description', default=False)
                verbosity.add_option('--no-progress',
                                action='store_true', dest='noprogress', help='do not print progress bar', default=False)
                parser.add_option_group(verbosity)
                verbosity.add_option('--no-progress',
                                action='store_true', dest='noprogress', help='do not print progress bar', default=False)
                parser.add_option_group(verbosity)
@@ -1744,7 +2052,7 @@ if __name__ == '__main__':
                filesystem.add_option('-o', '--output',
                                dest='outtmpl', metavar='TPL', help='output filename template')
                filesystem.add_option('-a', '--batch-file',
                filesystem.add_option('-o', '--output',
                                dest='outtmpl', metavar='TPL', help='output filename template')
                filesystem.add_option('-a', '--batch-file',
-                               dest='batchfile', metavar='F', help='file containing URLs to download')
+                               dest='batchfile', metavar='F', help='file containing URLs to download (\'-\' for stdin)')
                filesystem.add_option('-w', '--no-overwrites',
                                action='store_true', dest='nooverwrites', help='do not overwrite files', default=False)
                filesystem.add_option('-c', '--continue',
                filesystem.add_option('-w', '--no-overwrites',
                                action='store_true', dest='nooverwrites', help='do not overwrite files', default=False)
                filesystem.add_option('-c', '--continue',
@@ -1752,12 +2060,16 @@ if __name__ == '__main__':
                parser.add_option_group(filesystem)
 
                (opts, args) = parser.parse_args()
                parser.add_option_group(filesystem)
 
                (opts, args) = parser.parse_args()
-        
+
                # Batch file verification
                batchurls = []
                if opts.batchfile is not None:
                        try:
                # Batch file verification
                batchurls = []
                if opts.batchfile is not None:
                        try:
-                               batchurls = open(opts.batchfile, 'r').readlines()
+                               if opts.batchfile == '-':
+                                       batchfd = sys.stdin
+                               else:
+                                       batchfd = open(opts.batchfile, 'r')
+                               batchurls = batchfd.readlines()
                                batchurls = [x.strip() for x in batchurls]
                                batchurls = [x for x in batchurls if len(x) > 0]
                        except IOError:
                                batchurls = [x.strip() for x in batchurls]
                                batchurls = [x for x in batchurls if len(x) > 0]
                        except IOError:
@@ -1780,6 +2092,11 @@ if __name__ == '__main__':
                        if numeric_limit is None:
                                parser.error(u'invalid rate limit specified')
                        opts.ratelimit = numeric_limit
                        if numeric_limit is None:
                                parser.error(u'invalid rate limit specified')
                        opts.ratelimit = numeric_limit
+               if opts.retries is not None:
+                       try:
+                               opts.retries = long(opts.retries)
+                       except (TypeError, ValueError), err:
+                               parser.error(u'invalid retry count specified')
 
                # Information extractors
                youtube_ie = YoutubeIE()
 
                # Information extractors
                youtube_ie = YoutubeIE()
@@ -1788,8 +2105,10 @@ if __name__ == '__main__':
                youtube_user_ie = YoutubeUserIE(youtube_ie)
                youtube_search_ie = YoutubeSearchIE(youtube_ie)
                google_ie = GoogleIE()
                youtube_user_ie = YoutubeUserIE(youtube_ie)
                youtube_search_ie = YoutubeSearchIE(youtube_ie)
                google_ie = GoogleIE()
+               google_search_ie = GoogleSearchIE(google_ie)
                photobucket_ie = PhotobucketIE()
                yahoo_ie = YahooIE()
                photobucket_ie = PhotobucketIE()
                yahoo_ie = YahooIE()
+               yahoo_search_ie = YahooSearchIE(yahoo_ie)
                generic_ie = GenericIE()
 
                # File downloader
                generic_ie = GenericIE()
 
                # File downloader
@@ -1797,10 +2116,12 @@ if __name__ == '__main__':
                        'usenetrc': opts.usenetrc,
                        'username': opts.username,
                        'password': opts.password,
                        'usenetrc': opts.usenetrc,
                        'username': opts.username,
                        'password': opts.password,
-                       'quiet': (opts.quiet or opts.geturl or opts.gettitle),
+                       'quiet': (opts.quiet or opts.geturl or opts.gettitle or opts.getthumbnail or opts.getdescription),
                        'forceurl': opts.geturl,
                        'forcetitle': opts.gettitle,
                        'forceurl': opts.geturl,
                        'forcetitle': opts.gettitle,
-                       'simulate': (opts.simulate or opts.geturl or opts.gettitle),
+                       'forcethumbnail': opts.getthumbnail,
+                       'forcedescription': opts.getdescription,
+                       'simulate': (opts.simulate or opts.geturl or opts.gettitle or opts.getthumbnail or opts.getdescription),
                        'format': opts.format,
                        'outtmpl': ((opts.outtmpl is not None and opts.outtmpl.decode(preferredencoding()))
                                or (opts.format == '-1' and opts.usetitle and u'%(stitle)s-%(id)s-%(format)s.%(ext)s')
                        'format': opts.format,
                        'outtmpl': ((opts.outtmpl is not None and opts.outtmpl.decode(preferredencoding()))
                                or (opts.format == '-1' and opts.usetitle and u'%(stitle)s-%(id)s-%(format)s.%(ext)s')
@@ -1812,6 +2133,7 @@ if __name__ == '__main__':
                        'ignoreerrors': opts.ignoreerrors,
                        'ratelimit': opts.ratelimit,
                        'nooverwrites': opts.nooverwrites,
                        'ignoreerrors': opts.ignoreerrors,
                        'ratelimit': opts.ratelimit,
                        'nooverwrites': opts.nooverwrites,
+                       'retries': opts.retries,
                        'continuedl': opts.continue_dl,
                        'noprogress': opts.noprogress,
                        })
                        'continuedl': opts.continue_dl,
                        'noprogress': opts.noprogress,
                        })
@@ -1821,8 +2143,10 @@ if __name__ == '__main__':
                fd.add_info_extractor(metacafe_ie)
                fd.add_info_extractor(youtube_ie)
                fd.add_info_extractor(google_ie)
                fd.add_info_extractor(metacafe_ie)
                fd.add_info_extractor(youtube_ie)
                fd.add_info_extractor(google_ie)
+               fd.add_info_extractor(google_search_ie)
                fd.add_info_extractor(photobucket_ie)
                fd.add_info_extractor(yahoo_ie)
                fd.add_info_extractor(photobucket_ie)
                fd.add_info_extractor(yahoo_ie)
+               fd.add_info_extractor(yahoo_search_ie)
 
                # This must come last since it's the
                # fallback if none of the others work
 
                # This must come last since it's the
                # fallback if none of the others work