Merge branch 'master' into extract_info_rewrite
[youtube-dl] / youtube_dl / FileDownloader.py
index 5087b4cc824db2d7857407f3215cd0baaede9838..2237d355df893621a735f72f55cba38f9f7366ac 100644 (file)
@@ -231,11 +231,21 @@ class FileDownloader(object):
             self.to_stderr(message)
         if self.params.get('verbose'):
             if tb is None:
-                tb_data = traceback.format_list(traceback.extract_stack())
-                tb = u''.join(tb_data)
+                if sys.exc_info()[0]:  # if .trouble has been called from an except block
+                    tb = u''
+                    if hasattr(sys.exc_info()[1], 'exc_info') and sys.exc_info()[1].exc_info[0]:
+                        tb += u''.join(traceback.format_exception(*sys.exc_info()[1].exc_info))
+                    tb += compat_str(traceback.format_exc())
+                else:
+                    tb_data = traceback.format_list(traceback.extract_stack())
+                    tb = u''.join(tb_data)
             self.to_stderr(tb)
         if not self.params.get('ignoreerrors', False):
-            raise DownloadError(message)
+            if sys.exc_info()[0] and hasattr(sys.exc_info()[1], 'exc_info') and sys.exc_info()[1].exc_info[0]:
+                exc_info = sys.exc_info()[1].exc_info
+            else:
+                exc_info = sys.exc_info()
+            raise DownloadError(message, exc_info)
         self._download_retcode = 1
 
     def report_warning(self, message):
@@ -408,10 +418,85 @@ class FileDownloader(object):
             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']
 
@@ -475,14 +560,17 @@ class FileDownloader(object):
             subtitle = info_dict['subtitles'][0]
             (sub_error, sub_lang, sub) = subtitle
             sub_format = self.params.get('subtitlesformat')
-            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.report_error(u'Cannot write subtitles file ' + descfn)
-                return
+            if sub_error:
+                self.report_warning("Some error while getting the subtitles")
+            else:
+                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.report_error(u'Cannot write subtitles file ' + descfn)
+                    return
             if self.params.get('onlysubtitles', False):
                 return 
 
@@ -491,14 +579,17 @@ class FileDownloader(object):
             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 sub_error:
+                    self.report_warning("Some error while getting the subtitles")
+                else:
+                    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 
 
@@ -540,53 +631,17 @@ 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.report_warning(u'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.')
+            videos = self.extract_info(url)
 
-                # Suitable InfoExtractor found
-                suitable_found = True
-
-                # 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
+                    self.increment_downloads()
+                    self.process_info(video)
+                except UnavailableVideoError:
+                    self.trouble(u'\nERROR: unable to download video')
                 except MaxDownloadsReached:
                     self.to_screen(u'[info] Maximum number of downloaded files reached.')
                     raise
-                except Exception as e:
-                    if self.params.get('ignoreerrors', False):
-                        self.report_error(u'' + 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.to_stderr(u"\n")
-                        self.report_error(u'unable to download video')
-
-                # Suitable InfoExtractor had been found; go to next URL
-                break
-
-            if not suitable_found:
-                self.report_error(u'no suitable InfoExtractor: %s' % url)
 
         return self._download_retcode