Added transcription of audio and video files to local data sources (#1081)

Co-authored-by: Paul Koudelka <106623909+PaulKoudelka@users.noreply.github.com>

app/MindWork AI Studio/Tools/Databases/IndexStore/EmbeddingStateFile.cs

@@ -11,4 +11,5 @@ public sealed record EmbeddingStateFile(
     DateTimeOffset CreationUtc,
     DateTimeOffset LastWriteUtc,
     DateTimeOffset EmbeddedAtUtc,
-    int ChunkCount);
+    int ChunkCount,
+    string? TranscriptKey = null);

app/MindWork AI Studio/Tools/Databases/IndexStore/EmbeddingStateFileEntity.cs

@@ -26,6 +26,8 @@ internal sealed class EmbeddingStateFileEntity
 
     public int ChunkCount { get; set; }
 
+    public string? TranscriptKey { get; set; }
+
     public EmbeddingStateDataSourceEntity? DataSource { get; set; }
 
     public List<EmbeddingStateChunkEntity> Chunks { get; set; } = [];

app/MindWork AI Studio/Tools/Databases/IndexStore/IndexStoreDbContext.cs

@@ -80,6 +80,7 @@ internal sealed class IndexStoreDbContext(DbContextOptions<IndexStoreDbContext>
             entity.Property(file => file.LastWriteUtc).HasColumnName("last_write_utc").HasConversion(utcDateTimeOffsetConverter).IsRequired();
             entity.Property(file => file.EmbeddedAtUtc).HasColumnName("embedded_at_utc").HasConversion(utcDateTimeOffsetConverter).IsRequired();
             entity.Property(file => file.ChunkCount).HasColumnName("chunk_count");
+            entity.Property(file => file.TranscriptKey).HasColumnName("transcript_key");
 
             entity.HasIndex(file => file.DataSourceId).HasDatabaseName("idx_embedded_files_data_source");
             entity.HasIndex(file => file.AbsolutePath).HasDatabaseName("idx_embedded_files_absolute_path");

app/MindWork AI Studio/Tools/Databases/IndexStore/IndexStoreSchemaMigrator.cs

@@ -11,6 +11,7 @@ internal static class IndexStoreSchemaMigrator
     [DynamicDependency(DynamicallyAccessedMemberTypes.All, typeof(Migrations.DropFileConfidenceLevel))]
     [DynamicDependency(DynamicallyAccessedMemberTypes.All, typeof(Migrations.DropDataSourceName))]
     [DynamicDependency(DynamicallyAccessedMemberTypes.All, typeof(Migrations.MailboxIndex))]
+    [DynamicDependency(DynamicallyAccessedMemberTypes.All, typeof(Migrations.MediaTranscriptKey))]
     public static async Task MigrateAsync(IndexStoreDbContext context, CancellationToken token)
     {
         await context.Database.MigrateAsync(token);

app/MindWork AI Studio/Tools/Databases/IndexStore/Migrations/20261018000000_MediaTranscriptKey.cs

@@ -0,0 +1,35 @@
+#nullable disable
+
+using Microsoft.EntityFrameworkCore.Infrastructure;
+using Microsoft.EntityFrameworkCore.Migrations;
+
+namespace AIStudio.Tools.Databases.IndexStore.Migrations;
+
+/// <summary>
+/// Adds the key of the transcript to the indexed files, for audio and video files which a data source transcribed.
+/// </summary>
+/// <remarks>
+/// The transcripts lie below the data directory under the hash of the content of their media file.
+/// The index has to know which transcript belongs to which file, so that AI Studio can remove a
+/// transcript once its file is gone or changed. Every other file keeps NULL.
+/// </remarks>
+[DbContext(typeof(IndexStoreDbContext))]
+[Migration("20261018000000_MediaTranscriptKey")]
+public partial class MediaTranscriptKey : Migration
+{
+    /// <remarks>
+    /// Both directions go through raw SQL, like DropDataSourceName: the SQLite provider answers
+    /// DropColumn by rebuilding the table, and dropping the old embedded_files table would let the
+    /// cascade of the foreign key in embedding_chunks take every chunk with it. Adding a nullable
+    /// column is native in SQLite and leaves the rows alone.
+    /// </remarks>
+    protected override void Up(MigrationBuilder migrationBuilder)
+    {
+        migrationBuilder.Sql("ALTER TABLE embedded_files ADD COLUMN transcript_key TEXT NULL;");
+    }
+
+    protected override void Down(MigrationBuilder migrationBuilder)
+    {
+        migrationBuilder.Sql("ALTER TABLE embedded_files DROP COLUMN transcript_key;");
+    }
+}

app/MindWork AI Studio/Tools/Databases/IndexStore/Migrations/IndexStoreDbContextModelSnapshot.cs

@@ -119,6 +119,10 @@ partial class IndexStoreDbContextModelSnapshot : ModelSnapshot
                 .HasColumnType("TEXT")
                 .HasColumnName("relative_path");
 
+            entity.Property<string>("TranscriptKey")
+                .HasColumnType("TEXT")
+                .HasColumnName("transcript_key");
+
             entity.HasKey("ParentFileId");
 
             entity.HasIndex("AbsolutePath")

app/MindWork AI Studio/Tools/Databases/IndexStore/SqliteIndexStoreClientImplementation.cs

@@ -148,7 +148,8 @@ public sealed partial class SqliteIndexStoreClientImplementation(string name, st
                 file.FileSize,
                 file.LastWriteUtc,
                 file.EmbeddedAtUtc,
-                file.ChunkCount);
+                file.ChunkCount,
+                file.TranscriptKey);
         }
 
         var permanentFailures = await context.PermanentIndexingFailures
@@ -690,6 +691,7 @@ public sealed partial class SqliteIndexStoreClientImplementation(string name, st
         fileEntity.LastWriteUtc = file.LastWriteUtc;
         fileEntity.EmbeddedAtUtc = file.EmbeddedAtUtc;
         fileEntity.ChunkCount = file.ChunkCount;
+        fileEntity.TranscriptKey = file.TranscriptKey;
     }
 
     private static void ApplyPermanentFailure(IndexingFailureEntity failureEntity, string dataSourceId, PermanentIndexingFailure failure)