encode_compat_str,
encodeFilename,
error_to_compat_str,
+ ExistingVideoReached,
expand_path,
ExtractorError,
format_bytes,
register_socks_protocols,
render_table,
replace_extension,
+ RejectedVideoReached,
SameFileError,
sanitize_filename,
sanitize_path,
download_archive: File name of a file where all downloads are recorded.
Videos already present in the file are not downloaded
again.
- break_on_existing: Stop the download process after attempting to download a file that's
- in the archive.
- cookiefile: File name where cookies should be read from and dumped to.
+ break_on_existing: Stop the download process after attempting to download a
+ file that is in the archive.
+ break_on_reject: Stop the download process when encountering a video that
+ has been filtered out.
+ cookiefile: File name where cookies should be read from and dumped to
nocheckcertificate:Do not verify SSL certificates
prefer_insecure: Use HTTP instead of HTTPS to retrieve information.
At the moment, this is only supported by YouTube.
_pps = []
_download_retcode = None
_num_downloads = None
+ _playlist_level = 0
+ _playlist_urls = set()
_screen_file = None
def __init__(self, params=None, auto_init=True):
def _match_entry(self, info_dict, incomplete):
""" Returns None if the file should be downloaded """
- video_title = info_dict.get('title', info_dict.get('id', 'video'))
- if 'title' in info_dict:
- # This can happen when we're just evaluating the playlist
- title = info_dict['title']
- matchtitle = self.params.get('matchtitle', False)
- if matchtitle:
- if not re.search(matchtitle, title, re.IGNORECASE):
- return '"' + title + '" title did not match pattern "' + matchtitle + '"'
- rejecttitle = self.params.get('rejecttitle', False)
- if rejecttitle:
- if re.search(rejecttitle, title, re.IGNORECASE):
- return '"' + title + '" title matched reject pattern "' + rejecttitle + '"'
- date = info_dict.get('upload_date')
- if date is not None:
- dateRange = self.params.get('daterange', DateRange())
- if date not in dateRange:
- return '%s upload date is not in range %s' % (date_from_str(date).isoformat(), dateRange)
- view_count = info_dict.get('view_count')
- if view_count is not None:
- min_views = self.params.get('min_views')
- if min_views is not None and view_count < min_views:
- return 'Skipping %s, because it has not reached minimum view count (%d/%d)' % (video_title, view_count, min_views)
- max_views = self.params.get('max_views')
- if max_views is not None and view_count > max_views:
- return 'Skipping %s, because it has exceeded the maximum view count (%d/%d)' % (video_title, view_count, max_views)
- if age_restricted(info_dict.get('age_limit'), self.params.get('age_limit')):
- return 'Skipping "%s" because it is age restricted' % video_title
- if self.in_download_archive(info_dict):
- return '%s has already been recorded in archive' % video_title
-
- if not incomplete:
- match_filter = self.params.get('match_filter')
- if match_filter is not None:
- ret = match_filter(info_dict)
- if ret is not None:
- return ret
-
- return None
+ def check_filter():
+ video_title = info_dict.get('title', info_dict.get('id', 'video'))
+ if 'title' in info_dict:
+ # This can happen when we're just evaluating the playlist
+ title = info_dict['title']
+ matchtitle = self.params.get('matchtitle', False)
+ if matchtitle:
+ if not re.search(matchtitle, title, re.IGNORECASE):
+ return '"' + title + '" title did not match pattern "' + matchtitle + '"'
+ rejecttitle = self.params.get('rejecttitle', False)
+ if rejecttitle:
+ if re.search(rejecttitle, title, re.IGNORECASE):
+ return '"' + title + '" title matched reject pattern "' + rejecttitle + '"'
+ date = info_dict.get('upload_date')
+ if date is not None:
+ dateRange = self.params.get('daterange', DateRange())
+ if date not in dateRange:
+ return '%s upload date is not in range %s' % (date_from_str(date).isoformat(), dateRange)
+ view_count = info_dict.get('view_count')
+ if view_count is not None:
+ min_views = self.params.get('min_views')
+ if min_views is not None and view_count < min_views:
+ return 'Skipping %s, because it has not reached minimum view count (%d/%d)' % (video_title, view_count, min_views)
+ max_views = self.params.get('max_views')
+ if max_views is not None and view_count > max_views:
+ return 'Skipping %s, because it has exceeded the maximum view count (%d/%d)' % (video_title, view_count, max_views)
+ if age_restricted(info_dict.get('age_limit'), self.params.get('age_limit')):
+ return 'Skipping "%s" because it is age restricted' % video_title
+ if self.in_download_archive(info_dict):
+ return '%s has already been recorded in archive' % video_title
+
+ if not incomplete:
+ match_filter = self.params.get('match_filter')
+ if match_filter is not None:
+ ret = match_filter(info_dict)
+ if ret is not None:
+ return ret
+ return None
+
+ reason = check_filter()
+ if reason is not None:
+ self.to_screen('[download] ' + reason)
+ if reason.endswith('has already been recorded in the archive') and self.params.get('break_on_existing', False):
+ raise ExistingVideoReached()
+ elif self.params.get('break_on_reject', False):
+ raise RejectedVideoReached()
+ return reason
@staticmethod
def add_extra_info(info_dict, extra_info):
self.report_error(msg)
except ExtractorError as e: # An error we somewhat expected
self.report_error(compat_str(e), e.format_traceback())
- except MaxDownloadsReached:
+ except (MaxDownloadsReached, ExistingVideoReached, RejectedVideoReached):
raise
except Exception as e:
if self.params.get('ignoreerrors', False):
return self.process_ie_result(
new_result, download=download, extra_info=extra_info)
elif result_type in ('playlist', 'multi_video'):
- # We process each entry in the playlist
- playlist = ie_result.get('title') or ie_result.get('id')
- self.to_screen('[download] Downloading playlist: %s' % playlist)
-
- playlist_results = []
-
- playliststart = self.params.get('playliststart', 1) - 1
- playlistend = self.params.get('playlistend')
- # For backwards compatibility, interpret -1 as whole list
- if playlistend == -1:
- playlistend = None
-
- playlistitems_str = self.params.get('playlist_items')
- playlistitems = None
- if playlistitems_str is not None:
- def iter_playlistitems(format):
- for string_segment in format.split(','):
- if '-' in string_segment:
- start, end = string_segment.split('-')
- for item in range(int(start), int(end) + 1):
- yield int(item)
- else:
- yield int(string_segment)
- playlistitems = orderedSet(iter_playlistitems(playlistitems_str))
-
- ie_entries = ie_result['entries']
-
- def make_playlistitems_entries(list_ie_entries):
- num_entries = len(list_ie_entries)
- return [
- list_ie_entries[i - 1] for i in playlistitems
- if -num_entries <= i - 1 < num_entries]
-
- def report_download(num_entries):
+ # Protect from infinite recursion due to recursively nested playlists
+ # (see https://github.com/ytdl-org/youtube-dl/issues/27833)
+ webpage_url = ie_result['webpage_url']
+ if webpage_url in self._playlist_urls:
self.to_screen(
- '[%s] playlist %s: Downloading %d videos' %
- (ie_result['extractor'], playlist, num_entries))
+ '[download] Skipping already downloaded playlist: %s'
+ % ie_result.get('title') or ie_result.get('id'))
+ return
- if isinstance(ie_entries, list):
- n_all_entries = len(ie_entries)
- if playlistitems:
- entries = make_playlistitems_entries(ie_entries)
- else:
- entries = ie_entries[playliststart:playlistend]
- n_entries = len(entries)
- self.to_screen(
- '[%s] playlist %s: Collected %d video ids (downloading %d of them)' %
- (ie_result['extractor'], playlist, n_all_entries, n_entries))
- elif isinstance(ie_entries, PagedList):
- if playlistitems:
- entries = []
- for item in playlistitems:
- entries.extend(ie_entries.getslice(
- item - 1, item
- ))
- else:
- entries = ie_entries.getslice(
- playliststart, playlistend)
- n_entries = len(entries)
- report_download(n_entries)
- else: # iterable
- if playlistitems:
- entries = make_playlistitems_entries(list(itertools.islice(
- ie_entries, 0, max(playlistitems))))
- else:
- entries = list(itertools.islice(
- ie_entries, playliststart, playlistend))
- n_entries = len(entries)
- report_download(n_entries)
-
- if self.params.get('playlistreverse', False):
- entries = entries[::-1]
-
- if self.params.get('playlistrandom', False):
- random.shuffle(entries)
-
- x_forwarded_for = ie_result.get('__x_forwarded_for_ip')
-
- for i, entry in enumerate(entries, 1):
- self.to_screen('[download] Downloading video %s of %s' % (i, n_entries))
- # This __x_forwarded_for_ip thing is a bit ugly but requires
- # minimal changes
- if x_forwarded_for:
- entry['__x_forwarded_for_ip'] = x_forwarded_for
- extra = {
- 'n_entries': n_entries,
- 'playlist': playlist,
- 'playlist_id': ie_result.get('id'),
- 'playlist_title': ie_result.get('title'),
- 'playlist_uploader': ie_result.get('uploader'),
- 'playlist_uploader_id': ie_result.get('uploader_id'),
- 'playlist_index': playlistitems[i - 1] if playlistitems else i + playliststart,
- 'extractor': ie_result['extractor'],
- 'webpage_url': ie_result['webpage_url'],
- 'webpage_url_basename': url_basename(ie_result['webpage_url']),
- 'extractor_key': ie_result['extractor_key'],
- }
-
- reason = self._match_entry(entry, incomplete=True)
- if reason is not None:
- if reason.endswith('has already been recorded in the archive') and self.params.get('break_on_existing'):
- print('[download] tried downloading a file that\'s already in the archive, stopping since --break-on-existing is set.')
- break
- else:
- self.to_screen('[download] ' + reason)
- continue
-
- entry_result = self.__process_iterable_entry(entry, download, extra)
- # TODO: skip failed (empty) entries?
- playlist_results.append(entry_result)
- ie_result['entries'] = playlist_results
- self.to_screen('[download] Finished downloading playlist: %s' % playlist)
- return ie_result
+ self._playlist_level += 1
+ self._playlist_urls.add(webpage_url)
+ try:
+ return self.__process_playlist(ie_result, download)
+ finally:
+ self._playlist_level -= 1
+ if not self._playlist_level:
+ self._playlist_urls.clear()
elif result_type == 'compat_list':
self.report_warning(
'Extractor %s returned a compat_list result. '
else:
raise Exception('Invalid result type: %s' % result_type)
+ def __process_playlist(self, ie_result, download):
+ # We process each entry in the playlist
+ playlist = ie_result.get('title') or ie_result.get('id')
+ self.to_screen('[download] Downloading playlist: %s' % playlist)
+
+ playlist_results = []
+
+ playliststart = self.params.get('playliststart', 1) - 1
+ playlistend = self.params.get('playlistend')
+ # For backwards compatibility, interpret -1 as whole list
+ if playlistend == -1:
+ playlistend = None
+
+ playlistitems_str = self.params.get('playlist_items')
+ playlistitems = None
+ if playlistitems_str is not None:
+ def iter_playlistitems(format):
+ for string_segment in format.split(','):
+ if '-' in string_segment:
+ start, end = string_segment.split('-')
+ for item in range(int(start), int(end) + 1):
+ yield int(item)
+ else:
+ yield int(string_segment)
+ playlistitems = orderedSet(iter_playlistitems(playlistitems_str))
+
+ ie_entries = ie_result['entries']
+
+ def make_playlistitems_entries(list_ie_entries):
+ num_entries = len(list_ie_entries)
+ return [
+ list_ie_entries[i - 1] for i in playlistitems
+ if -num_entries <= i - 1 < num_entries]
+
+ def report_download(num_entries):
+ self.to_screen(
+ '[%s] playlist %s: Downloading %d videos' %
+ (ie_result['extractor'], playlist, num_entries))
+
+ if isinstance(ie_entries, list):
+ n_all_entries = len(ie_entries)
+ if playlistitems:
+ entries = make_playlistitems_entries(ie_entries)
+ else:
+ entries = ie_entries[playliststart:playlistend]
+ n_entries = len(entries)
+ self.to_screen(
+ '[%s] playlist %s: Collected %d video ids (downloading %d of them)' %
+ (ie_result['extractor'], playlist, n_all_entries, n_entries))
+ elif isinstance(ie_entries, PagedList):
+ if playlistitems:
+ entries = []
+ for item in playlistitems:
+ entries.extend(ie_entries.getslice(
+ item - 1, item
+ ))
+ else:
+ entries = ie_entries.getslice(
+ playliststart, playlistend)
+ n_entries = len(entries)
+ report_download(n_entries)
+ else: # iterable
+ if playlistitems:
+ entries = make_playlistitems_entries(list(itertools.islice(
+ ie_entries, 0, max(playlistitems))))
+ else:
+ entries = list(itertools.islice(
+ ie_entries, playliststart, playlistend))
+ n_entries = len(entries)
+ report_download(n_entries)
+
+ if self.params.get('playlistreverse', False):
+ entries = entries[::-1]
+
+ if self.params.get('playlistrandom', False):
+ random.shuffle(entries)
+
+ x_forwarded_for = ie_result.get('__x_forwarded_for_ip')
+
+ for i, entry in enumerate(entries, 1):
+ self.to_screen('[download] Downloading video %s of %s' % (i, n_entries))
+ # This __x_forwarded_for_ip thing is a bit ugly but requires
+ # minimal changes
+ if x_forwarded_for:
+ entry['__x_forwarded_for_ip'] = x_forwarded_for
+ extra = {
+ 'n_entries': n_entries,
+ 'playlist': playlist,
+ 'playlist_id': ie_result.get('id'),
+ 'playlist_title': ie_result.get('title'),
+ 'playlist_uploader': ie_result.get('uploader'),
+ 'playlist_uploader_id': ie_result.get('uploader_id'),
+ 'playlist_index': playlistitems[i - 1] if playlistitems else i + playliststart,
+ 'extractor': ie_result['extractor'],
+ 'webpage_url': ie_result['webpage_url'],
+ 'webpage_url_basename': url_basename(ie_result['webpage_url']),
+ 'extractor_key': ie_result['extractor_key'],
+ }
+
+ if self._match_entry(entry, incomplete=True) is not None:
+ continue
+
+ entry_result = self.__process_iterable_entry(entry, download, extra)
+ # TODO: skip failed (empty) entries?
+ playlist_results.append(entry_result)
+ ie_result['entries'] = playlist_results
+ self.to_screen('[download] Finished downloading playlist: %s' % playlist)
+ return ie_result
+
@__handle_extraction_exceptions
def __process_iterable_entry(self, entry, download, extra_info):
return self.process_ie_result(
if 'format' not in info_dict:
info_dict['format'] = info_dict['ext']
- reason = self._match_entry(info_dict, incomplete=False)
- if reason is not None:
- self.to_screen('[download] ' + reason)
+ if self._match_entry(info_dict, incomplete=False) is not None:
return
self._num_downloads += 1
except UnavailableVideoError:
self.report_error('unable to download video')
except MaxDownloadsReached:
- self.to_screen('[info] Maximum number of downloaded files reached.')
+ self.to_screen('[info] Maximum number of downloaded files reached')
+ raise
+ except ExistingVideoReached:
+ self.to_screen('[info] Encountered a file that is already in the archive, stopping due to --break-on-existing')
+ raise
+ except RejectedVideoReached:
+ self.to_screen('[info] Encountered a file that did not match filter, stopping due to --break-on-reject')
raise
else:
if self.params.get('dump_single_json', False):
projects / yt-dlp.git / blobdiff