Fix report block schema export in FileContext 1.0.20

CHANGELOG.md

@@ -2,6 +2,10 @@
 
 All notable changes to ManagedCode.FileContext are documented here.
 
+## 1.0.20
+
+- Keep report block content optional when exporting AI tool schemas and deserializing report inputs.
+
 ## 1.0.19
 
 - Apply file, result, page, concurrency, and timeout limits only when explicitly configured or requested. Default text and document reads return the remaining content.

Directory.Build.props

@@ -12,7 +12,7 @@
     <AnalysisMode>Recommended</AnalysisMode>
     <TreatWarningsAsErrors>true</TreatWarningsAsErrors>
     <NoWarn>$(NoWarn);CS1591;MAAI001</NoWarn>
-    <Version>1.0.19</Version>
+    <Version>1.0.20</Version>
     <PackageVersion>$(Version)</PackageVersion>
   </PropertyGroup>
 

src/ManagedCode.FileContext/Documents/Reports/FileContextReportBlock.cs

@@ -5,4 +5,12 @@ namespace ManagedCode.FileContext;
 
 /// <summary>A report block. Content follows the block JSON schema.</summary>
 public sealed record FileContextReportBlock(string Type,
-    [property: JsonIgnore(Condition = JsonIgnoreCondition.WhenWritingDefault)] JsonElement Content = default);
+    [property: JsonIgnore(Condition = JsonIgnoreCondition.WhenWritingDefault)] JsonElement Content)
+{
+    /// <summary>Creates a report block with no content.</summary>
+    // Bind Content through its init property so schema export can represent omitted content.
+    [JsonConstructor]
+    public FileContextReportBlock(string Type) : this(Type, default)
+    {
+    }
+}

tests/ManagedCode.FileContext.Tests/ReportInputValidationTests.cs

@@ -1,4 +1,5 @@
 using System.Text.Json;
+using Microsoft.Extensions.AI;
 using UglyToad.PdfPig;
 
 namespace ManagedCode.FileContext.Tests;
@@ -14,9 +15,13 @@ public async Task Page_break_can_omit_content()
         var options = new FileContextOptions { EnableWriteTools = true };
         var store = new ManagedCodeStorageFileStore(storage.Storage, options);
         var context = new FileContextService(store, options);
-        var pageBreak = JsonSerializer.Deserialize<FileContextReportBlock>("""{"Type":"page-break"}""", StrictJson)!;
-        var report = new FileContextReport("Report", "pages.pdf",
-            [pageBreak, new("text", JsonSerializer.SerializeToElement(new { text = "Second page" }))], ["pdf"]);
+        var function = AIFunctionFactory.Create(context.Documents.CreateReportAsync);
+        function.JsonSchema.GetRawText().ShouldContain("Content", Case.Insensitive);
+        var blocks = JsonSerializer.Deserialize<FileContextReportBlock[]>(
+            """[{"Type":"page-break"},{"Type":"text","Content":{"text":"Second page"}}]""", StrictJson)!;
+        blocks[0].Content.ValueKind.ShouldBe(JsonValueKind.Undefined);
+        blocks[1].Content.GetProperty("text").GetString().ShouldBe("Second page");
+        var report = new FileContextReport("Report", "pages.pdf", blocks, ["pdf"]);
 
         var result = await context.Documents.CreateReportAsync(report);