chore: use IANA and Apache MIME sources only

Author: KSemenenko

.github/workflows/mime-sync.yml

@@ -28,7 +28,6 @@ jobs:
       - name: Run MIME sync tool
         run: >
           dotnet run --project ManagedCode.MimeTypes.Sync --configuration Release --
-          --prefer-remote
           --template-concurrency 8
 
       - name: Detect MIME database changes
@@ -72,14 +71,15 @@ jobs:
         run: dotnet test --configuration Release --no-restore --verbosity normal
 
       - name: Create Pull Request
+        if: steps.mime_changes.outputs.changed == 'true'
         uses: peter-evans/create-pull-request@v6
         with:
           token: ${{ secrets.GITHUB_TOKEN }}
           commit-message: "chore: sync MIME database"
           title: "chore: sync MIME database"
           body: |
             Automated update of the MIME database from the IANA media types registry,
-            mime-db, Apache mime.types, and curated ManagedCode overrides.
+            and Apache mime.types.
 
             When MIME data changes, this PR also bumps the package patch version.
             See the workflow summary for the IANA registry date, source counts, and version bump.

Directory.Build.props

@@ -17,8 +17,8 @@
     <RepositoryUrl>https://github.com/managedcode/MimeTypes</RepositoryUrl>
     <PackageProjectUrl>https://github.com/managedcode/MimeTypes</PackageProjectUrl>
     <Product>Managed Code - MimeTypes</Product>
-    <Version>10.0.2</Version>
-    <PackageVersion>10.0.2</PackageVersion>
+    <Version>10.0.4</Version>
+    <PackageVersion>10.0.4</PackageVersion>
 
 </PropertyGroup>
 <PropertyGroup Condition="'$(GITHUB_ACTIONS)' == 'true'">

ManagedCode.MimeTypes.Sync/Program.cs

@@ -22,7 +22,6 @@ internal static partial class MimeTypeSyncTool
 
     private static readonly string[] DefaultSupplementalSources =
     [
-        "https://raw.githubusercontent.com/jshttp/mime-db/master/db.json",
         "https://raw.githubusercontent.com/apache/httpd/trunk/docs/conf/mime.types"
     ];
 
@@ -58,9 +57,9 @@ public static async Task<int> RunAsync(string[] args)
                 sourceMappings.AddRange(ParseSupplementalSource(source, raw, kind));
             }
 
-            sourceMappings.AddRange(CustomMappings().Select(static kvp => new SourceMapping(kvp.Key, kvp.Value, MimeSourceKind.Curated)));
-
-            var existing = LoadExisting(options.OutputPath);
+            var existing = options.PreserveExisting
+                ? LoadExisting(options.OutputPath)
+                : new Dictionary<string, string>(StringComparer.OrdinalIgnoreCase);
             var merged = MergeMappings(sourceMappings, existing, options.PreferRemote);
             var mergedMetadata = BuildMergedMetadata(metadata, merged);
 
@@ -502,24 +501,6 @@ private static IReadOnlyDictionary<string, MimeMetadata> BuildMergedMetadata(Dic
             .ToDictionary(static kvp => kvp.Key, static kvp => kvp.Value.NormalizeForOutput(), StringComparer.OrdinalIgnoreCase);
     }
 
-    private static IEnumerable<KeyValuePair<string, string>> CustomMappings()
-    {
-        yield return new KeyValuePair<string, string>("tar.gz", "application/gzip");
-        yield return new KeyValuePair<string, string>("gz", "application/gzip");
-        yield return new KeyValuePair<string, string>("tar.bz2", "application/x-bzip2");
-        yield return new KeyValuePair<string, string>("tar.xz", "application/x-xz");
-        yield return new KeyValuePair<string, string>("tar.zst", "application/zstd");
-        yield return new KeyValuePair<string, string>("d.ts", "application/typescript");
-        yield return new KeyValuePair<string, string>("cjs", "application/node");
-        yield return new KeyValuePair<string, string>("mjs", "text/javascript");
-        yield return new KeyValuePair<string, string>("wasm", "application/wasm");
-        yield return new KeyValuePair<string, string>("heic", "image/heic");
-        yield return new KeyValuePair<string, string>("heif", "image/heif");
-        yield return new KeyValuePair<string, string>("ics", "text/calendar");
-        yield return new KeyValuePair<string, string>("ps1", "application/x-powershell");
-        yield return new KeyValuePair<string, string>("appx", "application/vnd.ms-appx");
-    }
-
     private static void WriteMimeMap(string outputPath, IReadOnlyDictionary<string, MergedMapping> data)
     {
         Directory.CreateDirectory(Path.GetDirectoryName(outputPath)!);
@@ -844,19 +825,13 @@ private static string FormatXref(XElement xref)
 
     private static MimeSourceKind ClassifySupplementalSource(string source)
     {
-        if (source.Contains("jshttp/mime-db", StringComparison.OrdinalIgnoreCase) ||
-            source.EndsWith("db.json", StringComparison.OrdinalIgnoreCase))
-        {
-            return MimeSourceKind.MimeDb;
-        }
-
         if (source.Contains("apache", StringComparison.OrdinalIgnoreCase) ||
             source.EndsWith("mime.types", StringComparison.OrdinalIgnoreCase))
         {
             return MimeSourceKind.Apache;
         }
 
-        return MimeSourceKind.MimeDb;
+        return MimeSourceKind.Custom;
     }
 
     [GeneratedRegex(@"^\s*(?<name>[A-Za-z][A-Za-z0-9 /&().+\-]*(?:\(s\))?)\s*:\s*(?<value>.*)$", RegexOptions.Compiled)]
@@ -897,6 +872,7 @@ internal sealed record SyncOptions(
         bool PreferRemote,
         bool UseIana,
         bool SkipIanaTemplates,
+        bool PreserveExisting,
         int TemplateConcurrency)
     {
         public static SyncOptions Parse(string[] args)
@@ -908,6 +884,7 @@ public static SyncOptions Parse(string[] args)
             bool preferRemote = false;
             bool useIana = true;
             bool skipIanaTemplates = false;
+            bool preserveExisting = false;
             var templateConcurrency = 8;
             var customSupplementalSources = false;
 
@@ -952,13 +929,16 @@ public static SyncOptions Parse(string[] args)
                     case "--prefer-remote":
                         preferRemote = true;
                         break;
+                    case "--preserve-existing":
+                        preserveExisting = true;
+                        break;
                 }
             }
 
             output ??= Path.GetFullPath(Path.Combine(AppContext.BaseDirectory, "..", "..", "..", "..", "ManagedCode.MimeTypes", "mimeTypes.json"));
             metadataOutput ??= Path.Combine(Path.GetDirectoryName(output)!, "mimeTypes.metadata.json");
 
-            return new SyncOptions(ianaSource, supplementalSources, output, metadataOutput, preferRemote, useIana, skipIanaTemplates, templateConcurrency);
+            return new SyncOptions(ianaSource, supplementalSources, output, metadataOutput, preferRemote, useIana, skipIanaTemplates, preserveExisting, templateConcurrency);
         }
 
         private static void AddSources(List<string> sources, string value)
@@ -975,10 +955,9 @@ internal enum MimeSourceKind
 {
     Existing,
     Apache,
-    MimeDb,
+    Custom,
     Iana,
-    ExistingPreferred,
-    Curated
+    ExistingPreferred
 }
 
 internal static class MimeSourceKindExtensions
@@ -990,9 +969,8 @@ public static int Priority(this MimeSourceKind kind)
             MimeSourceKind.Existing => 0,
             MimeSourceKind.Apache => 10,
             MimeSourceKind.Iana => 15,
-            MimeSourceKind.MimeDb => 20,
+            MimeSourceKind.Custom => 20,
             MimeSourceKind.ExistingPreferred => 35,
-            MimeSourceKind.Curated => 40,
             _ => 0
         };
     }
@@ -1002,9 +980,8 @@ public static string SourceName(this MimeSourceKind kind)
         return kind switch
         {
             MimeSourceKind.Apache => "apache",
-            MimeSourceKind.MimeDb => "mime-db",
+            MimeSourceKind.Custom => "custom",
             MimeSourceKind.Iana => "iana",
-            MimeSourceKind.Curated => "curated",
             MimeSourceKind.ExistingPreferred => "existing",
             _ => "existing"
         };

ManagedCode.MimeTypes.Tests/GeneratorTests.cs

@@ -15,7 +15,6 @@ public void ExtensionsTest()
         MimeHelper.GetMimeType("C:\\\\users\\file.txt").ShouldBe(MimeHelper.TXT);
         MimeHelper.GetMimeType("https://cdn.example.com/assets/image.png?version=1").ShouldBe(MimeHelper.PNG);
         MimeHelper.GetMimeType("ARCHIVE.TAR.GZ").ShouldBe(MimeHelper.GZ);
-        MimeHelper.GetMimeType("module.d.ts").ShouldBe(MimeHelper.D_TS);
     }
 
     [Fact]
@@ -30,13 +29,11 @@ public void EmptyExtensionsTest()
     public void GeneratedPropertiesTest()
     {
         // Test static properties generated from mimeTypes.json
-        const string eventStreamMime = "text/event-stream";
         MimeHelper.PDF.ShouldBe(MimeHelper.GetMimeType(".pdf"));
         MimeHelper.DOCX.ShouldBe(MimeHelper.GetMimeType(".docx"));
         MimeHelper.PNG.ShouldBe(MimeHelper.GetMimeType(".png"));
         MimeHelper.MP4.ShouldBe(MimeHelper.GetMimeType(".mp4"));
         MimeHelper._7Z.ShouldBe(MimeHelper.GetMimeType(".7z"));
-        MimeHelper.EVENT_STREAM.ShouldBe(eventStreamMime);
     }
 
     [Fact]

ManagedCode.MimeTypes.Tests/MimeMetadataTests.cs

@@ -34,7 +34,7 @@ public void GetKnownMimeTypes_ShouldIncludeIanaAndSupplementalMetadata()
         var known = MimeHelper.GetKnownMimeTypes();
 
         known.ShouldContain(static info => info.Mime == "application/pdf" && info.IsIanaRegistered);
-        known.ShouldContain(static info => info.Mime == "application/x-powershell" && !info.IsIanaRegistered && info.Source == "curated");
+        known.ShouldContain(static info => info.Mime == "application/x-7z-compressed" && !info.IsIanaRegistered && info.Source == "apache");
     }
 
     [Fact]

ManagedCode.MimeTypes.Tests/StreamProbeTests.cs

@@ -7,7 +7,7 @@
 namespace ManagedCode.MimeTypes.Tests;
 
 /// <summary>
-/// Exercises stream-based MIME detection against the curated sample set.
+/// Exercises stream-based MIME detection against the local sample set.
 /// </summary>
 public class StreamProbeTests
 {

ManagedCode.MimeTypes.Tests/SyncToolTests.cs

@@ -70,21 +70,17 @@ public void MergeMappings_ShouldApplySourcePrecedence()
         {
             new SourceMapping("conflict", "application/apache", MimeSourceKind.Apache),
             new SourceMapping("conflict", "application/iana", MimeSourceKind.Iana),
-            new SourceMapping("conflict", "application/mime-db", MimeSourceKind.MimeDb),
-            new SourceMapping("iana-only", "application/iana", MimeSourceKind.Iana),
-            new SourceMapping("mjs", "application/x-old", MimeSourceKind.Iana),
-            new SourceMapping("mjs", "text/javascript", MimeSourceKind.Curated)
+            new SourceMapping("conflict", "application/custom", MimeSourceKind.Custom),
+            new SourceMapping("iana-only", "application/iana", MimeSourceKind.Iana)
         };
 
         var preferRemote = MimeTypeSyncTool.MergeMappings(mappings, existing, preferRemote: true);
-        preferRemote["conflict"].Mime.ShouldBe("application/mime-db");
+        preferRemote["conflict"].Mime.ShouldBe("application/custom");
         preferRemote["iana-only"].Mime.ShouldBe("application/iana");
         preferRemote["local"].Mime.ShouldBe("application/existing-local");
-        preferRemote["mjs"].Mime.ShouldBe("text/javascript");
 
         var preferExisting = MimeTypeSyncTool.MergeMappings(mappings, existing, preferRemote: false);
         preferExisting["conflict"].Mime.ShouldBe("application/existing");
-        preferExisting["mjs"].Mime.ShouldBe("text/javascript");
     }
 
     private const string IanaXml = """

ManagedCode.MimeTypes/ManagedCode.MimeTypes.csproj

@@ -21,8 +21,8 @@
     <PropertyGroup>
         <Title>ManagedCode.MimeTypes</Title>
         <PackageId>ManagedCode.MimeTypes</PackageId>
-        <Description>MimeTypes for .NET</Description>
-        <PackageTags>managedcode, mimetypes, mime</PackageTags>
+        <Description>Official IANA and Apache MIME/media type lookup, metadata, categorisation, and content detection for .NET</Description>
+        <PackageTags>managedcode, mimetypes, mime, media-types, iana, apache, content-detection, file-signatures, magic-numbers, mime-sniffing</PackageTags>
     </PropertyGroup>
 
     <ItemGroup>