]> jfr.im git - yt-dlp.git/blobdiff - yt_dlp/YoutubeDL.py
[cleanup] Misc
[yt-dlp.git] / yt_dlp / YoutubeDL.py
index 4b652d1725ee2033c08849cd335e342c18302115..f701738c9650da4cc15d0f192bfa61eac562e40f 100644 (file)
     write_json_file,
     write_string,
 )
-from .version import RELEASE_GIT_HEAD, VARIANT, __version__
+from .version import CHANNEL, RELEASE_GIT_HEAD, VARIANT, __version__
 
 if compat_os_name == 'nt':
     import ctypes
@@ -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
@@ -614,7 +617,7 @@ def __init__(self, params=None, auto_init=True):
                    '\n                    You will no longer receive updates on this version')
             if current_version < MIN_SUPPORTED:
                 msg = 'Python version %d.%d is no longer supported'
-            self.deprecation_warning(
+            self.deprecated_feature(
                 f'{msg}! Please update to Python %d.%d or above' % (*current_version, *MIN_RECOMMENDED))
 
         if self.params.get('allow_unplayable_formats'):
@@ -1407,31 +1410,44 @@ def check_filter():
                 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)
@@ -2466,15 +2482,8 @@ def _raise_pending_errors(self, info):
 
     def sort_formats(self, info_dict):
         formats = self._get_formats(info_dict)
-        if not formats:
-            return
-        # Backward compatibility with InfoExtractor._sort_formats
-        field_preference = formats[0].pop('__sort_fields', None)
-        if field_preference:
-            info_dict['_format_sort_fields'] = field_preference
-
         formats.sort(key=FormatSorter(
-            self, info_dict.get('_format_sort_fields', [])).calculate_preference)
+            self, info_dict.get('_format_sort_fields') or []).calculate_preference)
 
     def process_video_result(self, info_dict, download=True):
         assert info_dict.get('_type', 'video') == 'video'
@@ -2563,6 +2572,11 @@ def sanitize_numeric_fields(info):
 
         formats = self._get_formats(info_dict)
 
+        # Backward compatibility with InfoExtractor._sort_formats
+        field_preference = (formats or [{}])[0].pop('__sort_fields', None)
+        if field_preference:
+            info_dict['_format_sort_fields'] = field_preference
+
         # or None ensures --clean-infojson removes it
         info_dict['_has_drm'] = any(f.get('has_drm') for f in formats) or None
         if not self.params.get('allow_unplayable_formats'):
@@ -2623,7 +2637,10 @@ def is_wellformed(f):
         if '__x_forwarded_for_ip' in info_dict:
             del info_dict['__x_forwarded_for_ip']
 
-        self.sort_formats({'formats': formats})
+        self.sort_formats({
+            'formats': formats,
+            '_format_sort_fields': info_dict.get('_format_sort_fields')
+        })
 
         # Sanitize and group by format_id
         formats_dict = {}
@@ -2809,10 +2826,14 @@ def process_subtitles(self, video_id, normal_subtitles, automatic_captions):
                     self.params.get('subtitleslangs'), {'all': all_sub_langs}, use_regex=True)
             except re.error as e:
                 raise ValueError(f'Wrong regex for subtitlelangs: {e.pattern}')
-        elif normal_sub_langs:
-            requested_langs = ['en'] if 'en' in normal_sub_langs else normal_sub_langs[:1]
         else:
-            requested_langs = ['en'] if 'en' in all_sub_langs else all_sub_langs[:1]
+            requested_langs = LazyList(itertools.chain(
+                ['en'] if 'en' in normal_sub_langs else [],
+                filter(lambda f: f.startswith('en'), normal_sub_langs),
+                ['en'] if 'en' in all_sub_langs else [],
+                filter(lambda f: f.startswith('en'), all_sub_langs),
+                normal_sub_langs, all_sub_langs,
+            ))[:1]
         if requested_langs:
             self.to_screen(f'[info] {video_id}: Downloading subtitles: {", ".join(requested_langs)}')
 
@@ -3660,6 +3681,7 @@ def simplified_codec(f, field):
                 format_field(f, 'asr', '\t%s', func=format_decimal_suffix),
                 join_nonempty(
                     self._format_out('UNSUPPORTED', 'light red') if f.get('ext') in ('f4f', 'f4m') else None,
+                    self._format_out('DRM', 'light red') if f.get('has_drm') else None,
                     format_field(f, 'language', '[%s]'),
                     join_nonempty(format_field(f, 'format_note'),
                                   format_field(f, 'container', ignore=(None, f.get('ext'))),
@@ -3759,12 +3781,13 @@ def get_encoding(stream):
         source = detect_variant()
         if VARIANT not in (None, 'pip'):
             source += '*'
+        klass = type(self)
         write_debug(join_nonempty(
             f'{"yt-dlp" if REPOSITORY == "yt-dlp/yt-dlp" else REPOSITORY} version',
-            __version__,
-            f'[{RELEASE_GIT_HEAD}]' if RELEASE_GIT_HEAD else '',
+            f'{CHANNEL}@{__version__}',
+            f'[{RELEASE_GIT_HEAD[:9]}]' if RELEASE_GIT_HEAD else '',
             '' if source == 'unknown' else f'({source})',
-            '' if _IN_CLI else 'API',
+            '' if _IN_CLI else 'API' if klass == YoutubeDL else f'API:{self.__module__}.{klass.__qualname__}',
             delim=' '))
 
         if not _IN_CLI: