Pass the playlist info_dict to process_info
[youtube-dl] / youtube_dl / FileDownloader.py
index 7ad9d9a76ff6325f91f4f58215303c0390d1b0cb..d2b9be9ef88cb93e3e54fb0d3650880e35c36b43 100644 (file)
@@ -78,7 +78,11 @@ class FileDownloader(object):
     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
+    writesubtitles:    Write the video subtitles to a file
+    onlysubtitles:     Downloads only the subtitles of the video
+    allsubtitles:      Downloads all the subtitles of the video
+    listsubtitles:     Lists all available subtitles for the video
+    subtitlesformat:   Subtitle format [sbv/srt] (default=srt)
     subtitleslang:     Language of the subtitles to download
     test:              Download only first bytes to test the downloader.
     keepvideo:         Keep the video file after post-processing
@@ -104,7 +108,7 @@ class FileDownloader(object):
         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.')
+            self.report_warning(u'%(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):
@@ -234,6 +238,18 @@ class FileDownloader(object):
             raise DownloadError(message)
         self._download_retcode = 1
 
+    def report_warning(self, message):
+        '''
+        Print the message to stderr, it will be prefixed with 'WARNING:'
+        If stderr is a tty file the 'WARNING:' will be colored
+        '''
+        if sys.stderr.isatty():
+            _msg_header=u'\033[0;33mWARNING:\033[0m'
+        else:
+            _msg_header=u'WARNING:'
+        warning_message=u'%s %s' % (_msg_header,message)
+        self.to_stderr(warning_message)
+
     def slow_down(self, start_time, byte_counter):
         """Sleep if the download speed is over the rate limit."""
         rate_limit = self.params.get('ratelimit', None)
@@ -289,9 +305,9 @@ class FileDownloader(object):
         """ Report that the description file is being written """
         self.to_screen(u'[info] Writing video description to: ' + descfn)
 
-    def report_writesubtitles(self, srtfn):
+    def report_writesubtitles(self, sub_filename):
         """ Report that the subtitles file is being written """
-        self.to_screen(u'[info] Writing video subtitles to: ' + srtfn)
+        self.to_screen(u'[info] Writing video subtitles to: ' + sub_filename)
 
     def report_writeinfojson(self, infofn):
         """ Report that the metadata file has been written """
@@ -305,10 +321,11 @@ class FileDownloader(object):
         """Report download progress."""
         if self.params.get('noprogress', False):
             return
-        if self.params.get('newline', True):
+        if self.params.get('progress_with_newline', False):
             self.to_screen(u'[download] %s of %s at %s ETA %s' %
                 (percent_str, data_len_str, speed_str, eta_str))
-        else: self.to_screen(u'\r[download] %s of %s at %s ETA %s' %
+        else:
+            self.to_screen(u'\r[download] %s of %s at %s ETA %s' %
                 (percent_str, data_len_str, speed_str, eta_str), skip_eol=True)
         self.to_cons_title(u'youtube-dl - %s of %s at %s ETA %s' %
                 (percent_str.strip(), data_len_str.strip(), speed_str.strip(), eta_str.strip()))
@@ -359,8 +376,11 @@ class FileDownloader(object):
 
             filename = self.params['outtmpl'] % template_dict
             return filename
-        except (ValueError, KeyError) as err:
-            self.trouble(u'ERROR: invalid system charset or erroneous output template')
+        except KeyError as err:
+            self.trouble(u'ERROR: Erroneous output template')
+            return None
+        except ValueError as err:
+            self.trouble(u'ERROR: Insufficient system charset ' + repr(preferredencoding()))
             return None
 
     def _match_entry(self, info_dict):
@@ -369,19 +389,92 @@ class FileDownloader(object):
         title = info_dict['title']
         matchtitle = self.params.get('matchtitle', False)
         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:
-            rejecttitle = rejecttitle.decode('utf8')
             if re.search(rejecttitle, title, re.IGNORECASE):
                 return u'"' + title + '" title matched reject pattern "' + rejecttitle + '"'
         return None
+        
+    def extract_info(self, url):
+        '''
+        Returns a list with a dictionary for each video we find.
+         '''
+        suitable_found = False
+        for ie in self._ies:
+            # Go to next InfoExtractor if not suitable
+            if not ie.suitable(url):
+                continue
+
+            # Warn if the _WORKING attribute is False
+            if not ie.working():
+                self.to_stderr(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
+            try:
+                ie_results = ie.extract(url)
+                results = self.process_ie_results(ie_results, ie)
+                return results
+            except ExtractorError as de: # An error we somewhat expected
+                self.trouble(u'ERROR: ' + compat_str(de), de.format_traceback())
+                break
+            except Exception as e:
+                if self.params.get('ignoreerrors', False):
+                    self.trouble(u'ERROR: ' + compat_str(e), tb=compat_str(traceback.format_exc()))
+                    break
+                else:
+                    raise
+        if not suitable_found:
+                self.trouble(u'ERROR: no suitable InfoExtractor: %s' % url)
+    def extract_info_iterable(self, urls):
+        '''
+            Return the videos founded for the urls
+        '''
+        results = []
+        for url in urls:
+            results.extend(self.extract_info(url))
+        return results
+        
+    def process_ie_results(self, ie_results, ie):
+        """
+        Take the results of the ie and return a list of videos.
+        For url elements it will seartch the suitable ie and get the videos
+        For playlist elements it will process each of the elements of the 'entries' key
+        """
+        results = [] 
+        for result in ie_results or []:
+            result_type = result.get('_type', 'video') #If not given we suppose it's a video, support the dafault old system
+            if result_type == 'video':
+                if not 'extractor' in result:
+                    #The extractor has already been set somewhere else
+                    result['extractor'] = ie.IE_NAME
+                results.append(result)
+            elif result_type == 'url':
+                #We get the videos pointed by the url
+                results.extend(self.extract_info(result['url']))
+            elif result_type == 'playlist':
+                #We process each entry in the playlist
+                entries_result = self.process_ie_results(result['entries'], ie)
+                result['entries'] = entries_result
+                results.extend([result])
+        return results
 
     def process_info(self, info_dict):
         """Process a single dictionary returned by an InfoExtractor."""
 
+        if info_dict.get('_type','video') == 'playlist':
+            playlist = info_dict.get('title', None) or info_dict.get('id', None)
+            self.to_screen(u'[download] Downloading playlist: %s'  % playlist)
+            for video in info_dict['entries']:
+                video['playlist'] = playlist
+                self.process_info(video)
+            return
+        
         # Keep for backwards compatibility
         info_dict['stitle'] = info_dict['title']
 
@@ -442,14 +535,35 @@ class FileDownloader(object):
         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
+            subtitle = info_dict['subtitles'][0]
+            (sub_error, sub_lang, sub) = subtitle
+            sub_format = self.params.get('subtitlesformat')
             try:
-                srtfn = filename.rsplit('.', 1)[0] + u'.srt'
-                self.report_writesubtitles(srtfn)
-                with io.open(encodeFilename(srtfn), 'w', encoding='utf-8') as srtfile:
-                    srtfile.write(info_dict['subtitles'])
+                sub_filename = filename.rsplit('.', 1)[0] + u'.' + sub_lang + u'.' + sub_format
+                self.report_writesubtitles(sub_filename)
+                with io.open(encodeFilename(sub_filename), 'w', encoding='utf-8') as subfile:
+                    subfile.write(sub)
             except (OSError, IOError):
                 self.trouble(u'ERROR: Cannot write subtitles file ' + descfn)
                 return
+            if self.params.get('onlysubtitles', False):
+                return 
+
+        if self.params.get('allsubtitles', False) and 'subtitles' in info_dict and info_dict['subtitles']:
+            subtitles = info_dict['subtitles']
+            sub_format = self.params.get('subtitlesformat')
+            for subtitle in subtitles:
+                (sub_error, sub_lang, sub) = subtitle
+                try:
+                    sub_filename = filename.rsplit('.', 1)[0] + u'.' + sub_lang + u'.' + sub_format
+                    self.report_writesubtitles(sub_filename)
+                    with io.open(encodeFilename(sub_filename), 'w', encoding='utf-8') as subfile:
+                            subfile.write(sub)
+                except (OSError, IOError):
+                    self.trouble(u'ERROR: Cannot write subtitles file ' + descfn)
+                    return
+            if self.params.get('onlysubtitles', False):
+                return 
 
         if self.params.get('writeinfojson', False):
             infofn = filename + u'.info.json'
@@ -489,49 +603,14 @@ class FileDownloader(object):
             raise SameFileError(self.params['outtmpl'])
 
         for url in url_list:
-            suitable_found = False
-            for ie in self._ies:
-                # Go to next InfoExtractor if not suitable
-                if not ie.suitable(url):
-                    continue
-
-                # Warn if the _WORKING attribute is False
-                if not ie.working():
-                    self.to_stderr(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
+            videos = self.extract_info(url)
 
-                # Extract information from URL and process it
+            for video in videos or []:
                 try:
-                    videos = ie.extract(url)
-                except ExtractorError as de: # An error we somewhat expected
-                    self.trouble(u'ERROR: ' + compat_str(de), de.format_traceback())
-                    break
-                except Exception as e:
-                    if self.params.get('ignoreerrors', False):
-                        self.trouble(u'ERROR: ' + compat_str(e), tb=compat_str(traceback.format_exc()))
-                        break
-                    else:
-                        raise
-
-                if len(videos or []) > 1 and self.fixed_template():
-                    raise SameFileError(self.params['outtmpl'])
-
-                for video in videos or []:
-                    video['extractor'] = ie.IE_NAME
-                    try:
-                        self.increment_downloads()
-                        self.process_info(video)
-                    except UnavailableVideoError:
-                        self.trouble(u'\nERROR: unable to download video')
-
-                # Suitable InfoExtractor had been found; go to next URL
-                break
-
-            if not suitable_found:
-                self.trouble(u'ERROR: no suitable InfoExtractor: %s' % url)
+                    self.increment_downloads()
+                    self.process_info(video)
+                except UnavailableVideoError:
+                    self.trouble(u'\nERROR: unable to download video')
 
         return self._download_retcode
 
@@ -553,10 +632,10 @@ class FileDownloader(object):
                 self.to_stderr(u'ERROR: ' + e.msg)
         if keep_video is False and not self.params.get('keepvideo', False):
             try:
-                self.to_stderr(u'Deleting original file %s (pass -k to keep)' % filename)
+                self.to_screen(u'Deleting original file %s (pass -k to keep)' % filename)
                 os.remove(encodeFilename(filename))
             except (IOError, OSError):
-                self.to_stderr(u'WARNING: Unable to remove downloaded video file')
+                self.report_warning(u'Unable to remove downloaded video file')
 
     def _download_with_rtmpdump(self, filename, url, player_url, page_url):
         self.report_destination(filename)
@@ -564,7 +643,7 @@ class FileDownloader(object):
 
         # Check for rtmpdump first
         try:
-            subprocess.call(['rtmpdump', '-h'], stdout=(file(os.path.devnull, 'w')), stderr=subprocess.STDOUT)
+            subprocess.call(['rtmpdump', '-h'], stdout=(open(os.path.devnull, 'w')), stderr=subprocess.STDOUT)
         except (OSError, IOError):
             self.trouble(u'ERROR: RTMP download detected but "rtmpdump" could not be run')
             return False