From eee4f7163d52165dcfb2e4bc6db200b74f3e3176 Mon Sep 17 00:00:00 2001 From: Luc Ritchie Date: Tue, 13 Dec 2022 07:17:44 -0500 Subject: [PATCH] Handle `--flat-playlist` when recording `multi_video` in archive --- yt_dlp/YoutubeDL.py | 23 ++++++++++++++++++----- 1 file changed, 18 insertions(+), 5 deletions(-) diff --git a/yt_dlp/YoutubeDL.py b/yt_dlp/YoutubeDL.py index f4f07c1ad..9c5dd37fe 100644 --- a/yt_dlp/YoutubeDL.py +++ b/yt_dlp/YoutubeDL.py @@ -1413,7 +1413,8 @@ class YoutubeDL: return self.get_output_path(dir_type, filename) def _match_entry(self, info_dict, incomplete=False, silent=False): - """Returns None if the file should be downloaded""" + """Returns None if the file should be downloaded, False if the file is already present in + the download archive, or a string describing another reason to skip the file""" _type = 'video' if 'playlist-match-filter' in self.params['compat_opts'] else info_dict.get('_type', 'video') assert incomplete or _type == 'video', 'Only video result can be considered complete' @@ -1488,6 +1489,7 @@ class YoutubeDL: format_field(info_dict, 'id', f'{self._format_screen("%s", self.Styles.ID)}: '), format_field(info_dict, 'title', f'{self._format_screen("%s", self.Styles.EMPHASIS)} '), 'has already been recorded in the archive')) + ret = False break_opt, break_err = 'break_on_existing', ExistingVideoReached else: try: @@ -1496,12 +1498,13 @@ class YoutubeDL: reason, break_opt, break_err = e.msg, 'match_filter', type(e) else: break_opt, break_err = 'break_on_reject', RejectedVideoReached + ret = reason if reason is not None: if not silent: self.to_screen('[download] ' + reason) if self.params.get(break_opt, False): raise break_err() - return reason + return ret @staticmethod def add_extra_info(info_dict, extra_info): @@ -1775,6 +1778,7 @@ class YoutubeDL: self._raise_pending_errors(info_copy) if self.params.get('force_write_download_archive', False): self.record_download_archive(info_copy) + ie_result['__write_download_archive'] = self.params.get('force_write_download_archive', False) return ie_result if result_type == 'video': @@ -1912,7 +1916,9 @@ class YoutubeDL: common_info = self._playlist_infodict(ie_result, strict=True) title = common_info.get('playlist') or '' - if self._match_entry(common_info, incomplete=True) is not None: + skip_reason = self._match_entry(common_info, incomplete=True) + if skip_reason is not None: + ie_result['__write_download_archive'] = skip_reason is False return self.to_screen(f'[download] Downloading {ie_result["_type"]}: {title}') @@ -1967,6 +1973,7 @@ class YoutubeDL: self.write_debug('The information of all playlist entries will be held in memory') failures = 0 + all_write_download_archive = True max_failures = self.params.get('skip_playlist_after_errors') or float('inf') for i, (playlist_index, entry) in enumerate(entries): if lazy: @@ -1999,6 +2006,8 @@ class YoutubeDL: }, extra)) if not entry_result: failures += 1 + elif not entry_result.get('__write_download_archive', False): + all_write_download_archive = False if failures >= max_failures: self.report_error( f'Skipping the remaining entries in playlist "{title}" since {failures} items failed extraction') @@ -2015,7 +2024,8 @@ class YoutubeDL: if ie_result['_type'] == 'multi_video' and not failures: if self.params.get('force_write_download_archive') or ( - not self.params.get('simulate') and not self.params.get('skip_download')): + all_write_download_archive and not self.params.get('simulate') + and not self.params.get('skip_download')): self.record_download_archive(ie_result) # Write the updated info to json @@ -2830,7 +2840,9 @@ class YoutubeDL: info_dict, _ = self.pre_process(info_dict) - if self._match_entry(info_dict, incomplete=self._format_fields) is not None: + skip_reason = self._match_entry(info_dict, incomplete=self._format_fields) + if skip_reason is not None: + info_dict['__write_download_archive'] = skip_reason is False return info_dict self.post_extract(info_dict) @@ -2938,6 +2950,7 @@ class YoutubeDL: assert write_archive.issubset({True, False, 'ignore'}) if True in write_archive and False not in write_archive: self.record_download_archive(info_dict) + info_dict['__write_download_archive'] = True info_dict['requested_downloads'] = downloaded_formats info_dict = self.run_all_pps('after_video', info_dict)