Language tag matching expands to extended form

PlexCleaner/Language.cs

@@ -152,11 +152,9 @@ public static bool IsUndefined(string language)
         return string.IsNullOrEmpty(language) || language.Equals(Undefined, StringComparison.OrdinalIgnoreCase);
     }
 
-    public static bool IsMatch(string prefix, string language)
+    public bool IsMatch(string prefix, string language)
     {
-        // TODO: Is there an easy to use C# BCP 47 matcher?
         // https://r12a.github.io/app-subtags/
-        // https://www.loc.gov/standards/iso639-2/php/langcodes-search.php
 
         // zh match: zh: zh, zh-Hant, zh-Hans, zh-cmn-Hant
         // zho not: zh
@@ -176,11 +174,25 @@ public static bool IsMatch(string prefix, string language)
             return true;
         }
 
+        // Get the extended format of the language
+        // E.g. cmn-Hant should be expanded to zh-cmn-Hant else zh will not match
+
+        // Find a matching RFC 5646 record
+        var rfc5646 = Rfc5646.Find(language, false);
+        if (rfc5646 != null)
+        {
+            // If the lookup is different then rematch
+            if (!string.Equals(language, rfc5646.TagAny, StringComparison.OrdinalIgnoreCase))
+            {
+                return IsMatch(prefix, rfc5646.TagAny);
+            }
+        }
+
         // No match
         return false;
     }
 
-    public static bool IsMatch(string language, IEnumerable<string> prefixList)
+    public bool IsMatch(string language, IEnumerable<string> prefixList)
     {
         // Match language with any of the prefixes
         return prefixList.Any(prefix => IsMatch(prefix, language));

PlexCleaner/ProcessFile.cs

@@ -1910,7 +1910,7 @@ public SelectMediaInfo FindUnwantedLanguageTracks()
         // Select tracks with wanted languages, or the original language if set to keep
         // Selected is Keep
         // NotSelected is Remove
-        SelectMediaInfo selectMediaInfo = new(MkvMergeInfo, item => Language.IsMatch(item.LanguageIetf, Program.Config.ProcessOptions.KeepLanguages) ||
+        SelectMediaInfo selectMediaInfo = new(MkvMergeInfo, item => Language.Singleton.IsMatch(item.LanguageIetf, Program.Config.ProcessOptions.KeepLanguages) ||
                                                                     (Program.Config.ProcessOptions.KeepOriginalLanguage && item.Flags.HasFlag(TrackInfo.FlagsType.Original)));
 
         // Keep at least one video track if any

PlexCleanerTests/LanguageTests.cs

@@ -27,11 +27,12 @@ public void Convert_Iso_To_Ietf(string tag, string ietf)
     [InlineData("en", "en-US")]
     [InlineData("en", "en-GB")]
     [InlineData("en-GB", "en-GB")]
-    [InlineData("zh", "zh-Hant")]
+    [InlineData("zh", "zh-cmn-Hant")]
+    [InlineData("zh", "cmn-Hant")]
     [InlineData("sr-Latn", "sr-Latn-RS")]
     public void Match_Language_Tags(string prefix, string tag)
     {
-        Assert.True(Language.IsMatch(prefix, tag));
+        Assert.True(Language.Singleton.IsMatch(prefix, tag));
     }
 
     [Theory]
@@ -40,7 +41,7 @@ public void Match_Language_Tags(string prefix, string tag)
     [InlineData("zh-Hant", "zh-Hans")]
     public void NotMatch_Language_Tags(string prefix, string tag)
     {
-        Assert.False(Language.IsMatch(prefix, tag));
+        Assert.False(Language.Singleton.IsMatch(prefix, tag));
     }
 
     [Theory]

README.md

@@ -27,7 +27,7 @@ Docker images are published on [Docker Hub](https://hub.docker.com/u/ptr727/plex
 
 - Version 3.0:
   - Switched docker base image from `ubuntu:latest` to `archlinux:latest`.
-    - The up to date FFmpeg and HandBrake PPA installations provided by Rob Savoury are [no longer freely available](https://launchpad.net/~savoury1), a big historic thank you to Rob.
+    - The always up to date FFmpeg and HandBrake PPA installations provided by Rob Savoury are [no longer generally available](https://launchpad.net/~savoury1), a big historic thank you to Rob.
     - Switched to [Arch Linux](https://archlinux.org/) with up to date media tools found in the [Arch Linux Package Repository](https://archlinux.org/packages/).
   - Switched from .NET 6 to .NET 7.
     - Utilizing some new capabilities, e.g. `GeneratedRegex` and `LibraryImport`.
@@ -51,7 +51,6 @@ Docker images are published on [Docker Hub](https://hub.docker.com/u/ptr727/plex
       - If you care and can, please do communicate the need for IETF language support to the FFmpeg and HandBrake development teams.
     - Added warnings and attempt to repair when the Language and LanguageIetf are set and are invalid or do not match.
     - `MkvMerge --identify` added the `--normalize-language-ietf extlang` option to reported e.g. `zh-cmn-Hant` vs. the normalized `cmn-Hant`.
-      - Old MkvMerge sidecar information will be out of data, recommend recreating sidecar files using `createsidecar` option.
   - Added `ProcessOptions:KeepOriginalLanguage` to keep tracks marked as [original language](https://www.ietf.org/archive/id/draft-ietf-cellar-matroska-15.html#name-original-flag).
   - Added `ProcessOptions:RemoveClosedCaptions` to conditionally vs. always remove closed captions.
   - Added `ProcessOptions:SetTrackFlags` to set track flags based on track title keywords, e.g. `SDH` -> `HearingImpaired`.