Add option `--break-match-filters`

* Deprecates `--break-on-reject`

Closes #5962

README.md

@@ -114,7 +114,7 @@ yt-dlp is a [youtube-dl](https://github.com/ytdl-org/youtube-dl) fork based on t
 
 * **Output template improvements**: Output templates can now have date-time formatting, numeric offsets, object traversal etc. See [output template](#output-template) for details. Even more advanced operations can also be done with the help of `--parse-metadata` and `--replace-in-metadata`
 
-* **Other new options**: Many new options have been added such as `--alias`, `--print`, `--concat-playlist`, `--wait-for-video`, `--retry-sleep`, `--sleep-requests`, `--convert-thumbnails`, `--force-download-archive`, `--force-overwrites`, `--break-on-reject` etc
+* **Other new options**: Many new options have been added such as `--alias`, `--print`, `--concat-playlist`, `--wait-for-video`, `--retry-sleep`, `--sleep-requests`, `--convert-thumbnails`, `--force-download-archive`, `--force-overwrites`, `--break-match-filter` etc
 
 * **Improvements**: Regex and other operators in `--format`/`--match-filter`, multiple `--postprocessor-args` and `--downloader-args`, faster archive checking, more [format selection options](#format-selection), merge multi-video/audio, multiple `--config-locations`, `--exec` at different stages, etc
 
@@ -519,7 +519,10 @@ If you fork the project on GitHub, you can run your fork's [build workflow](.git
                                     dogs" (caseless). Use "--match-filter -" to
                                     interactively ask whether to download each
                                     video
-    --no-match-filter               Do not use generic video filter (default)
+    --no-match-filter               Do not use any --match-filter (default)
+    --break-match-filters FILTER    Same as "--match-filters" but stops the
+                                    download process when a video is rejected
+    --no-break-match-filters        Do not use any --break-match-filters (default)
     --no-playlist                   Download only the video, if the URL refers
                                     to a video and a playlist
     --yes-playlist                  Download the playlist, if the URL refers to
@@ -533,8 +536,6 @@ If you fork the project on GitHub, you can run your fork's [build workflow](.git
     --max-downloads NUMBER          Abort after downloading NUMBER files
     --break-on-existing             Stop the download process when encountering
                                     a file that is in the archive
-    --break-on-reject               Stop the download process when encountering
-                                    a file that has been filtered out
     --break-per-input               Alters --max-downloads, --break-on-existing,
                                     --break-on-reject, and autonumber to reset
                                     per input URL
@@ -2133,6 +2134,7 @@ While these options are redundant, they are still expected to be used due to the
     --reject-title REGEX             --match-filter "title !~= (?i)REGEX"
     --min-views COUNT                --match-filter "view_count >=? COUNT"
     --max-views COUNT                --match-filter "view_count <=? COUNT"
+    --break-on-reject                Use --break-match-filter
     --user-agent UA                  --add-header "User-Agent:UA"
     --referer URL                    --add-header "Referer:URL"
     --playlist-start NUMBER          -I NUMBER:

yt_dlp/YoutubeDL.py

@@ -300,8 +300,6 @@ class YoutubeDL:
                        Videos already present in the file are not downloaded again.
     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.
     break_per_url:     Whether break_on_reject and break_on_existing
                        should act on each input URL as opposed to for the entire queue
     cookiefile:        File name or text stream from where cookies should be read and dumped to
@@ -414,6 +412,8 @@ class YoutubeDL:
                        - If it returns None, the video is downloaded.
                        - If it returns utils.NO_DEFAULT, the user is interactively
                          asked whether to download the video.
+                       - Raise utils.DownloadCancelled(msg) to abort remaining
+                         downloads when a video is rejected.
                        match_filter_func in utils.py is one example for this.
     no_color:          Do not emit color codes in output.
     geo_bypass:        Bypass geographic restriction via faking X-Forwarded-For
@@ -483,6 +483,9 @@ class YoutubeDL:
 
     The following options are deprecated and may be removed in the future:
 
+    break_on_reject:   Stop the download process when encountering a video that
+                       has been filtered out.
+                       - `raise DownloadCancelled(msg)` in match_filter instead
     force_generic_extractor: Force downloader to use the generic extractor
                        - Use allowed_extractors = ['generic', 'default']
     playliststart:     - Use playlist_items
@@ -1407,31 +1410,44 @@ class YoutubeDL:
                 return 'Skipping "%s" because it is age restricted' % video_title
 
             match_filter = self.params.get('match_filter')
-            if match_filter is not None:
+            if match_filter is None:
+                return None
+
+            cancelled = None
+            try:
                 try:
                     ret = match_filter(info_dict, incomplete=incomplete)
                 except TypeError:
                     # For backward compatibility
                     ret = None if incomplete else match_filter(info_dict)
-                if ret is NO_DEFAULT:
-                    while True:
-                        filename = self._format_screen(self.prepare_filename(info_dict), self.Styles.FILENAME)
-                        reply = input(self._format_screen(
-                            f'Download "{filename}"? (Y/n): ', self.Styles.EMPHASIS)).lower().strip()
-                        if reply in {'y', ''}:
-                            return None
-                        elif reply == 'n':
-                            return f'Skipping {video_title}'
-                elif ret is not None:
-                    return ret
-            return None
+            except DownloadCancelled as err:
+                if err.msg is not NO_DEFAULT:
+                    raise
+                ret, cancelled = err.msg, err
+
+            if ret is NO_DEFAULT:
+                while True:
+                    filename = self._format_screen(self.prepare_filename(info_dict), self.Styles.FILENAME)
+                    reply = input(self._format_screen(
+                        f'Download "{filename}"? (Y/n): ', self.Styles.EMPHASIS)).lower().strip()
+                    if reply in {'y', ''}:
+                        return None
+                    elif reply == 'n':
+                        if cancelled:
+                            raise type(cancelled)(f'Skipping {video_title}')
+                        return f'Skipping {video_title}'
+            return ret
 
         if self.in_download_archive(info_dict):
             reason = '%s has already been recorded in the archive' % video_title
             break_opt, break_err = 'break_on_existing', ExistingVideoReached
         else:
-            reason = check_filter()
-            break_opt, break_err = 'break_on_reject', RejectedVideoReached
+            try:
+                reason = check_filter()
+            except DownloadCancelled as e:
+                reason, break_opt, break_err = e.msg, 'match_filter', type(e)
+            else:
+                break_opt, break_err = 'break_on_reject', RejectedVideoReached
         if reason is not None:
             if not silent:
                 self.to_screen('[download] ' + reason)

yt_dlp/__init__.py

@@ -403,7 +403,7 @@ def validate_options(opts):
         except Exception:
             raise ValueError('unsupported geo-bypass country or ip-block')
 
-    opts.match_filter = match_filter_func(opts.match_filter)
+    opts.match_filter = match_filter_func(opts.match_filter, opts.breaking_match_filter)
 
     if opts.download_archive is not None:
         opts.download_archive = expand_path(opts.download_archive)

yt_dlp/options.py

@@ -613,8 +613,16 @@ def create_parser():
             'Use "--match-filter -" to interactively ask whether to download each video'))
     selection.add_option(
         '--no-match-filter',
-        metavar='FILTER', dest='match_filter', action='store_const', const=None,
-        help='Do not use generic video filter (default)')
+        dest='match_filter', action='store_const', const=None,
+        help='Do not use any --match-filter (default)')
+    selection.add_option(
+        '--break-match-filters',
+        metavar='FILTER', dest='breaking_match_filter', action='append',
+        help='Same as "--match-filters" but stops the download process when a video is rejected')
+    selection.add_option(
+        '--no-break-match-filters',
+        dest='breaking_match_filter', action='store_const', const=None,
+        help='Do not use any --break-match-filters (default)')
     selection.add_option(
         '--no-playlist',
         action='store_true', dest='noplaylist', default=False,
@@ -646,11 +654,11 @@ def create_parser():
     selection.add_option(
         '--break-on-reject',
         action='store_true', dest='break_on_reject', default=False,
-        help='Stop the download process when encountering a file that has been filtered out')
+        help=optparse.SUPPRESS_HELP)
     selection.add_option(
         '--break-per-input',
         action='store_true', dest='break_per_url', default=False,
-        help='Alters --max-downloads, --break-on-existing, --break-on-reject, and autonumber to reset per input URL')
+        help='Alters --max-downloads, --break-on-existing, --break-match-filter, and autonumber to reset per input URL')
     selection.add_option(
         '--no-break-per-input',
         action='store_false', dest='break_per_url',

yt_dlp/utils.py

@@ -1230,8 +1230,8 @@ class ExistingVideoReached(DownloadCancelled):
 
 
 class RejectedVideoReached(DownloadCancelled):
-    """ --break-on-reject triggered """
-    msg = 'Encountered a video that did not match filter, stopping due to --break-on-reject'
+    """ --break-match-filter triggered """
+    msg = 'Encountered a video that did not match filter, stopping due to --break-match-filter'
 
 
 class MaxDownloadsReached(DownloadCancelled):
@@ -3911,16 +3911,21 @@ def match_str(filter_str, dct, incomplete=False):
         for filter_part in re.split(r'(?<!\\)&', filter_str))
 
 
-def match_filter_func(filters):
-    if not filters:
+def match_filter_func(filters, breaking_filters=None):
+    if not filters and not breaking_filters:
         return None
-    filters = set(variadic(filters))
+    breaking_filters = match_filter_func(breaking_filters) or (lambda _, __: None)
+    filters = set(variadic(filters or []))
 
     interactive = '-' in filters
     if interactive:
         filters.remove('-')
 
     def _match_func(info_dict, incomplete=False):
+        ret = breaking_filters(info_dict, incomplete)
+        if ret is not None:
+            raise RejectedVideoReached(ret)
+
         if not filters or any(match_str(f, info_dict, incomplete) for f in filters):
             return NO_DEFAULT if interactive and not incomplete else None
         else: