Add option `--replace-in-metadata`

[yt-dlp.git] / yt_dlp / __init__.py

diff --git a/yt_dlp/__init__.py b/yt_dlp/__init__.py
index fb0f46a4a9c00139b541d12d2e1f279e9fb32845..73e3f9f78602fa3c4c334c1eb30990fe52ad6eb9 100644 (file)
--- a/yt_dlp/__init__.py
+++ b/yt_dlp/__init__.py
@@ -7,6 +7,7 @@
 
 import codecs
 import io
+import itertools
 import os
 import random
 import re
@@ -18,6 +19,7 @@
 )
 from .compat import (
     compat_getpass,
+    compat_shlex_quote,
     workaround_optparse_bug9161,
 )
 from .cookies import SUPPORTED_BROWSERS
@@ -46,14 +48,15 @@
 from .extractor import gen_extractors, list_extractors
 from .extractor.common import InfoExtractor
 from .extractor.adobepass import MSO_INFO
-from .postprocessor.ffmpeg import (
+from .postprocessor import (
     FFmpegExtractAudioPP,
     FFmpegSubtitlesConvertorPP,
     FFmpegThumbnailsConvertorPP,
     FFmpegVideoConvertorPP,
     FFmpegVideoRemuxerPP,
+    MetadataFromFieldPP,
+    MetadataParserPP,
 )
-from .postprocessor.metadatafromfield import MetadataFromFieldPP
 from .YoutubeDL import YoutubeDL
 
 
@@ -280,7 +283,7 @@ def parse_compat_opts():
         'filename', 'format-sort', 'abort-on-error', 'format-spec', 'no-playlist-metafiles',
         'multistreams', 'no-live-chat', 'playlist-index', 'list-formats', 'no-direct-merge',
         'no-youtube-channel-redirect', 'no-youtube-unavailable-videos', 'no-attach-info-json',
-        'embed-thumbnail-atomicparsley', 'seperate-video-versions', 'no-clean-infojson',
+        'embed-thumbnail-atomicparsley', 'seperate-video-versions', 'no-clean-infojson', 'no-keep-subs',
     ]
     compat_opts = parse_compat_opts()
 
@@ -344,13 +347,29 @@ def validate_outtmpl(tmpl, msg):
         if re.match(InfoExtractor.FormatSort.regex, f) is None:
             parser.error('invalid format sort string "%s" specified' % f)
 
-    if opts.metafromfield is None:
-        opts.metafromfield = []
+    def metadataparser_actions(f):
+        if isinstance(f, str):
+            cmd = '--parse-metadata %s' % compat_shlex_quote(f)
+            try:
+                actions = [MetadataFromFieldPP.to_action(f)]
+            except Exception as err:
+                parser.error(f'{cmd} is invalid; {err}')
+        else:
+            cmd = '--replace-in-metadata %s' % ' '.join(map(compat_shlex_quote, f))
+            actions = ((MetadataParserPP.Actions.REPLACE, x, *f[1:]) for x in f[0].split(','))
+
+        for action in actions:
+            try:
+                MetadataParserPP.validate_action(*action)
+            except Exception as err:
+                parser.error(f'{cmd} is invalid; {err}')
+            yield action
+
+    if opts.parse_metadata is None:
+        opts.parse_metadata = []
     if opts.metafromtitle is not None:
-        opts.metafromfield.append('title:%s' % opts.metafromtitle)
-    for f in opts.metafromfield:
-        if re.match(MetadataFromFieldPP.regex, f) is None:
-            parser.error('invalid format string "%s" specified for --parse-metadata' % f)
+        opts.parse_metadata.append('title:%s' % opts.metafromtitle)
+    opts.parse_metadata = list(itertools.chain(*map(metadataparser_actions, opts.parse_metadata)))
 
     any_getting = opts.forceprint or opts.geturl or opts.gettitle or opts.getid or opts.getthumbnail or opts.getdescription or opts.getfilename or opts.getformat or opts.getduration or opts.dumpjson or opts.dump_single_json
     any_printing = opts.print_json
@@ -402,10 +421,10 @@ def report_conflict(arg1, arg2):
 
     # PostProcessors
     postprocessors = []
-    if opts.metafromfield:
+    if opts.parse_metadata:
         postprocessors.append({
-            'key': 'MetadataFromField',
-            'formats': opts.metafromfield,
+            'key': 'MetadataParser',
+            'actions': opts.parse_metadata,
             # Run this immediately after extraction is complete
             'when': 'pre_process'
         })
@@ -458,13 +477,13 @@ def report_conflict(arg1, arg2):
     if opts.addmetadata:
         postprocessors.append({'key': 'FFmpegMetadata'})
     if opts.embedsubtitles:
-        already_have_subtitle = opts.writesubtitles
+        already_have_subtitle = opts.writesubtitles and 'no-keep-subs' not in compat_opts
         postprocessors.append({
             'key': 'FFmpegEmbedSubtitle',
             # already_have_subtitle = True prevents the file from being deleted after embedding
             'already_have_subtitle': already_have_subtitle
         })
-        if not already_have_subtitle:
+        if not opts.writeautomaticsub and 'no-keep-subs' not in compat_opts:
             opts.writesubtitles = True
     # --all-sub automatically sets --write-sub if --write-auto-sub is not given
     # this was the old behaviour if only --all-sub was given.
@@ -550,7 +569,7 @@ def report_args_compat(arg, name):
         'forcejson': opts.dumpjson or opts.print_json,
         'dump_single_json': opts.dump_single_json,
         'force_write_download_archive': opts.force_write_download_archive,
-        'simulate': opts.simulate or any_getting,
+        'simulate': (any_getting or None) if opts.simulate is None else opts.simulate,
         'skip_download': opts.skip_download,
         'format': opts.format,
         'allow_unplayable_formats': opts.allow_unplayable_formats,