]> gitweb @ CieloNegro.org - youtube-dl.git/blobdiff - youtube_dl/FileDownloader.py
Merge branch 'master' of github.com:rg3/youtube-dl
[youtube-dl.git] / youtube_dl / FileDownloader.py
index 724de17c7be0dd05a73bcff4ce55bd18ca5b9f57..870c8227235c5a423768be8b13e92f7cb9cb2a7c 100644 (file)
@@ -13,7 +13,7 @@ import urllib2
 
 if os.name == 'nt':
        import ctypes
-       
+
 from utils import *
 
 
@@ -44,39 +44,40 @@ class FileDownloader(object):
 
        Available options:
 
-       username:         Username for authentication purposes.
-       password:         Password for authentication purposes.
-       usenetrc:         Use netrc for authentication instead.
-       quiet:            Do not print messages to stdout.
-       forceurl:         Force printing final URL.
-       forcetitle:       Force printing title.
-       forcethumbnail:   Force printing thumbnail URL.
-       forcedescription: Force printing description.
-       forcefilename:    Force printing final filename.
-       simulate:         Do not download the video files.
-       format:           Video format code.
-       format_limit:     Highest quality format to try.
-       outtmpl:          Template for output names.
-       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 5xx
-       buffersize:       Size of download buffer in bytes.
-       noresizebuffer:   Do not automatically resize the download buffer.
-       continuedl:       Try to continue downloads if possible.
-       noprogress:       Do not print the progress bar.
-       playliststart:    Playlist item to start at.
-       playlistend:      Playlist item to end at.
-       matchtitle:       Download only matching titles.
-       rejecttitle:      Reject downloads for matching titles.
-       logtostderr:      Log messages to stderr instead of stdout.
-       consoletitle:     Display progress in console window's titlebar.
-       nopart:           Do not use temporary .part files.
-       updatetime:       Use the Last-modified header to set output file timestamps.
-       writedescription: Write the video description to a .description file
-       writeinfojson:    Write the video description to a .info.json file
-       writesubtitles:   Write the video subtitles to a .srt file
-       subtitleslang:    Language of the subtitles to download
+       username:          Username for authentication purposes.
+       password:          Password for authentication purposes.
+       usenetrc:          Use netrc for authentication instead.
+       quiet:             Do not print messages to stdout.
+       forceurl:          Force printing final URL.
+       forcetitle:        Force printing title.
+       forcethumbnail:    Force printing thumbnail URL.
+       forcedescription:  Force printing description.
+       forcefilename:     Force printing final filename.
+       simulate:          Do not download the video files.
+       format:            Video format code.
+       format_limit:      Highest quality format to try.
+       outtmpl:           Template for output names.
+       restrictfilenames: Do not allow "&" and spaces in file names
+       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 5xx
+       buffersize:        Size of download buffer in bytes.
+       noresizebuffer:    Do not automatically resize the download buffer.
+       continuedl:        Try to continue downloads if possible.
+       noprogress:        Do not print the progress bar.
+       playliststart:     Playlist item to start at.
+       playlistend:       Playlist item to end at.
+       matchtitle:        Download only matching titles.
+       rejecttitle:       Reject downloads for matching titles.
+       logtostderr:       Log messages to stderr instead of stdout.
+       consoletitle:      Display progress in console window's titlebar.
+       nopart:            Do not use temporary .part files.
+       updatetime:        Use the Last-modified header to set output file timestamps.
+       writedescription:  Write the video description to a .description file
+       writeinfojson:     Write the video description to a .info.json file
+       writesubtitles:    Write the video subtitles to a .srt file
+       subtitleslang:     Language of the subtitles to download
        """
 
        params = None
@@ -95,6 +96,9 @@ class FileDownloader(object):
                self._screen_file = [sys.stdout, sys.stderr][params.get('logtostderr', False)]
                self.params = params
 
+               if '%(stitle)s' in self.params['outtmpl']:
+                       self.to_stderr(u'WARNING: %(stitle)s is deprecated. Use the %(title)s and the --restrict-filenames flag(which also secures %(uploader)s et al) instead.')
+
        @staticmethod
        def format_bytes(bytes):
                if bytes is None:
@@ -104,7 +108,7 @@ class FileDownloader(object):
                if bytes == 0.0:
                        exponent = 0
                else:
-                       exponent = long(math.log(bytes, 1024.0))
+                       exponent = int(math.log(bytes, 1024.0))
                suffix = 'bkMGTPEZY'[exponent]
                converted = float(bytes) / float(1024 ** exponent)
                return '%.2f%s' % (converted, suffix)
@@ -123,7 +127,7 @@ class FileDownloader(object):
                if current == 0 or dif < 0.001: # One millisecond
                        return '--:--'
                rate = float(current) / dif
-               eta = long((float(total) - float(current)) / rate)
+               eta = int((float(total) - float(current)) / rate)
                (eta_mins, eta_secs) = divmod(eta, 60)
                if eta_mins > 99:
                        return '--:--'
@@ -141,23 +145,23 @@ class FileDownloader(object):
                new_min = max(bytes / 2.0, 1.0)
                new_max = min(max(bytes * 2.0, 1.0), 4194304) # Do not surpass 4 MB
                if elapsed_time < 0.001:
-                       return long(new_max)
+                       return int(new_max)
                rate = bytes / elapsed_time
                if rate > new_max:
-                       return long(new_max)
+                       return int(new_max)
                if rate < new_min:
-                       return long(new_min)
-               return long(rate)
+                       return int(new_min)
+               return int(rate)
 
        @staticmethod
        def parse_bytes(bytestr):
-               """Parse a string indicating a byte quantity into a long integer."""
+               """Parse a string indicating a byte quantity into an integer."""
                matchobj = re.match(r'(?i)^(\d+(?:\.\d+)?)([kMGTPEZY]?)$', bytestr)
                if matchobj is None:
                        return None
                number = float(matchobj.group(1))
                multiplier = 1024.0 ** 'bkmgtpezy'.index(matchobj.group(2).lower())
-               return long(round(number * multiplier))
+               return int(round(number * multiplier))
 
        def add_info_extractor(self, ie):
                """Add an InfoExtractor object to the end of the list."""
@@ -175,15 +179,15 @@ class FileDownloader(object):
                if not self.params.get('quiet', False):
                        terminator = [u'\n', u''][skip_eol]
                        output = message + terminator
-
-                       if 'b' not in self._screen_file.mode or sys.version_info[0] < 3: # Python 2 lies about the mode of sys.stdout/sys.stderr
+                       if 'b' in getattr(self._screen_file, 'mode', '') or sys.version_info[0] < 3: # Python 2 lies about the mode of sys.stdout/sys.stderr
                                output = output.encode(preferredencoding(), 'ignore')
                        self._screen_file.write(output)
                        self._screen_file.flush()
 
        def to_stderr(self, message):
                """Print message to stderr."""
-               print >>sys.stderr, message.encode(preferredencoding())
+               assert type(message) == type(u'')
+               sys.stderr.write((message + u'\n').encode(preferredencoding()))
 
        def to_cons_title(self, message):
                """Set console/terminal window title to message."""
@@ -323,8 +327,13 @@ class FileDownloader(object):
                """Generate the output filename."""
                try:
                        template_dict = dict(info_dict)
-                       template_dict['epoch'] = unicode(long(time.time()))
-                       template_dict['autonumber'] = unicode('%05d' % self._num_downloads)
+
+                       template_dict['epoch'] = int(time.time())
+                       template_dict['autonumber'] = u'%05d' % self._num_downloads
+
+                       template_dict = dict((key, u'NA' if val is None else val) for key, val in template_dict.items())
+                       template_dict = dict((k, sanitize_filename(u(v), self.params.get('restrictfilenames'))) for k,v in template_dict.items())
+
                        filename = self.params['outtmpl'] % template_dict
                        return filename
                except (ValueError, KeyError), err:
@@ -336,17 +345,25 @@ class FileDownloader(object):
 
                title = info_dict['title']
                matchtitle = self.params.get('matchtitle', False)
-               if matchtitle and not re.search(matchtitle, title, re.IGNORECASE):
-                       return u'[download] "' + title + '" title did not match pattern "' + matchtitle + '"'
+               if matchtitle:
+                       matchtitle = matchtitle.decode('utf8')
+                       if not re.search(matchtitle, title, re.IGNORECASE):
+                               return u'[download] "' + title + '" title did not match pattern "' + matchtitle + '"'
                rejecttitle = self.params.get('rejecttitle', False)
-               if rejecttitle and re.search(rejecttitle, title, re.IGNORECASE):
-                       return u'"' + title + '" title matched reject pattern "' + rejecttitle + '"'
+               if rejecttitle:
+                       rejecttitle = rejecttitle.decode('utf8')
+                       if re.search(rejecttitle, title, re.IGNORECASE):
+                               return u'"' + title + '" title matched reject pattern "' + rejecttitle + '"'
                return None
 
        def process_info(self, info_dict):
                """Process a single dictionary returned by an InfoExtractor."""
 
-               info_dict['stitle'] = sanitize_filename(info_dict['title'])
+               # Keep for backwards compatibility
+               info_dict['stitle'] = info_dict['title']
+
+               if not 'format' in info_dict:
+                       info_dict['format'] = info_dict['ext']
 
                reason = self._match_entry(info_dict)
                if reason is not None:
@@ -359,20 +376,20 @@ class FileDownloader(object):
                                raise MaxDownloadsReached()
 
                filename = self.prepare_filename(info_dict)
-               
+
                # Forced printings
                if self.params.get('forcetitle', False):
-                       print info_dict['title'].encode(preferredencoding(), 'xmlcharrefreplace')
+                       print(info_dict['title'].encode(preferredencoding(), 'xmlcharrefreplace'))
                if self.params.get('forceurl', False):
-                       print info_dict['url'].encode(preferredencoding(), 'xmlcharrefreplace')
+                       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')
+                       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')
+                       print(info_dict['description'].encode(preferredencoding(), 'xmlcharrefreplace'))
                if self.params.get('forcefilename', False) and filename is not None:
-                       print filename.encode(preferredencoding(), 'xmlcharrefreplace')
+                       print(filename.encode(preferredencoding(), 'xmlcharrefreplace'))
                if self.params.get('forceformat', False):
-                       print info_dict['format'].encode(preferredencoding(), 'xmlcharrefreplace')
+                       print(info_dict['format'].encode(preferredencoding(), 'xmlcharrefreplace'))
 
                # Do nothing else if in simulate mode
                if self.params.get('simulate', False):
@@ -386,7 +403,7 @@ class FileDownloader(object):
                        if dn != '' and not os.path.exists(dn): # dn is already encoded
                                os.makedirs(dn)
                except (OSError, IOError), err:
-                       self.trouble(u'ERROR: unable to create directory ' + unicode(err))
+                       self.trouble(u'ERROR: unable to create directory ' + u(err))
                        return
 
                if self.params.get('writedescription', False):
@@ -401,10 +418,10 @@ class FileDownloader(object):
                        except (OSError, IOError):
                                self.trouble(u'ERROR: Cannot write description file ' + descfn)
                                return
-                               
+
                if self.params.get('writesubtitles', False) and 'subtitles' in info_dict and info_dict['subtitles']:
                        # subtitles download errors are already managed as troubles in relevant IE
-                       # that way it will silently go on when used with unsupporting IE 
+                       # that way it will silently go on when used with unsupporting IE
                        try:
                                srtfn = filename.rsplit('.', 1)[0] + u'.srt'
                                self.report_writesubtitles(srtfn)
@@ -450,7 +467,7 @@ class FileDownloader(object):
                                except (ContentTooShortError, ), err:
                                        self.trouble(u'ERROR: content too short (expected %s bytes and served %s)' % (err.expected, err.downloaded))
                                        return
-       
+
                        if success:
                                try:
                                        self.post_process(filename, info_dict)
@@ -470,12 +487,18 @@ class FileDownloader(object):
                                if not ie.suitable(url):
                                        continue
 
+                               # Warn if the _WORKING attribute is False
+                               if not ie.working():
+                                       self.trouble(u'WARNING: the program functionality for this site has been marked as broken, '
+                                                        u'and will probably not work. If you want to go on, use the -i option.')
+
                                # Suitable InfoExtractor found
                                suitable_found = True
 
                                # Extract information from URL and process it
                                videos = ie.extract(url)
                                for video in videos or []:
+                                       video['extractor'] = ie.IE_NAME
                                        try:
                                                self.increment_downloads()
                                                self.process_info(video)
@@ -605,7 +628,7 @@ class FileDownloader(object):
                                        else:
                                                # Examine the reported length
                                                if (content_length is not None and
-                                                               (resume_len - 100 < long(content_length) < resume_len + 100)):
+                                                               (resume_len - 100 < int(content_length) < resume_len + 100)):
                                                        # The file had already been fully downloaded.
                                                        # Explanation to the above condition: in issue #175 it was revealed that
                                                        # YouTube sometimes adds or removes a few bytes from the end of the file,
@@ -632,7 +655,7 @@ class FileDownloader(object):
 
                data_len = data.info().get('Content-length', None)
                if data_len is not None:
-                       data_len = long(data_len) + resume_len
+                       data_len = int(data_len) + resume_len
                data_len_str = self.format_bytes(data_len)
                byte_counter = 0 + resume_len
                block_size = self.params.get('buffersize', 1024)
@@ -682,7 +705,7 @@ class FileDownloader(object):
                stream.close()
                self.report_finish()
                if data_len is not None and byte_counter != data_len:
-                       raise ContentTooShortError(byte_counter, long(data_len))
+                       raise ContentTooShortError(byte_counter, int(data_len))
                self.try_rename(tmpfilename, filename)
 
                # Update file modification time