updates for naming MuPDF.NET.PDF4LLM

.gitignore

@@ -18,6 +18,7 @@ mono_crash.*
 
 # Build results
 docs/build/
+TestDocuments/_Output/
 [Dd]ebug/
 [Dd]ebugPublic/
 [Rr]elease/
@@ -35,10 +36,9 @@ bld/
 
 # Test-generated output (PDFs, markdown, etc.)
 TestDocuments/MuPDF.NET.Test/_Output/
-TestDocuments/PDF4LLM.Test/_Output/
+TestDocuments/MuPDF.NET.PDF4LLM.Test/_Output/
 TestDocuments/Demo/_Output/
 
-
 # Visual Studio 2015/2017 cache/options directory
 .vs/
 # Uncomment if you have tasks that create the project's static files in wwwroot
@@ -47,9 +47,6 @@ TestDocuments/Demo/_Output/
 # Visual Studio 2017 auto generated files
 Generated\ Files/
 
-# MuPDF.NativeAssets generation files
-MuPDF/packages
-
 # MSTest test Results
 [Tt]est[Rr]esult*/
 [Bb]uild[Ll]og.*

Demo/Demo.csproj

@@ -1,16 +1,20 @@
 <Project Sdk="Microsoft.NET.Sdk">
-  <!-- Demo: Support/, Samples/&lt;area&gt;/; PDF4LLM fixtures in Samples/Llm/Program.Llm.*.Fixtures.cs. -->
+  <!-- Demo: MuPDF.NET and MuPDF.NET.PDF4LLM samples under Samples/. -->
 
   <PropertyGroup>
     <OutputType>Exe</OutputType>
     <TargetFramework>net8.0</TargetFramework>
     <Platforms>AnyCPU;x64;x86</Platforms>
-    <NoWarn>$(NoWarn);0618;CA1416</NoWarn>
+    <ImplicitUsings>enable</ImplicitUsings>
+    <Nullable>annotations</Nullable>
+    <NoWarn>$(NoWarn);0618;CA1416;NU1603</NoWarn>
   </PropertyGroup>
 
   <ItemGroup>
-    <ProjectReference Include="..\PDF4LLM\PDF4LLM.csproj" />
     <ProjectReference Include="..\MuPDF.NET\MuPDF.NET.csproj" />
+    <ProjectReference Include="..\MuPDF.NET.PDF4LLM\MuPDF.NET.PDF4LLM.csproj" />
   </ItemGroup>
 
+  <Import Project="..\MuPDF.NET.PDF4LLM\build\MuPDF.NET.PDF4LLM.targets" Condition="Exists('..\MuPDF.NET.PDF4LLM\build\MuPDF.NET.PDF4LLM.targets')" />
+
 </Project>

Demo/GlobalUsings.cs

@@ -7,16 +7,15 @@
 global using System.Threading.Tasks;
 global using mupdf;
 global using MuPDF.NET;
-global using PDF4LLM;
-global using static global::PDF4LLM.PdfExtractor;
-global using PDF4LLM.Helpers;
-global using PDF4LLM.Llama;
-global using PDF4LLM.Ocr;
 global using SkiaSharp;
 global using Box = MuPDF.NET.Box;
 global using Encoding = System.Text.Encoding;
 global using File = System.IO.File;
 global using Font = MuPDF.NET.Font;
 global using Morph = MuPDF.NET.Morph;
 global using TextWriter = MuPDF.NET.TextWriter;
+global using MuPDF.NET.PDF4LLM;
+global using MuPDF.NET.PDF4LLM.Helpers;
+global using MuPDF.NET.PDF4LLM.Llama;
+global using MuPDF.NET.PDF4LLM.Ocr;
 global using Utils = MuPDF.NET.Utils;

Demo/Program.cs

@@ -1,4 +1,10 @@
-namespace Demo
+using System;
+using System.IO;
+using System.Linq;
+using MuPDF.NET;
+using mupdf;
+
+namespace Demo
 {
     /// <summary>
     /// Demo entry point. With no arguments, every sample in <see cref="SampleMenu"/> runs (including <c>[diag]</c>).

Demo/README.md

@@ -1,6 +1,6 @@
 # MuPDF.NET Demo
 
-Runnable samples for **MuPDF.NET** and **PDF4LLM**, grouped by API area under `Samples/`.
+Runnable samples for **MuPDF.NET** and **MuPDF.NET.PDF4LLM**, grouped by API area under `Samples/`.
 
 ## Run
 
@@ -11,7 +11,8 @@ dotnet run                    # all samples in SampleMenu.cs (default, includes
 dotnet run -- user            # user-facing samples only (skips [diag])
 dotnet run -- diagnostics     # [diag] samples only
 dotnet run -- help            # list all samples
-dotnet run -- hello-new-pdf   # one sample by name
+dotnet run -- hello-new-pdf   # one MuPDF.NET sample by name
+dotnet run -- rag-markdown    # one MuPDF.NET.PDF4LLM sample by name
 ```
 
 ## Layout
@@ -24,11 +25,7 @@ dotnet run -- hello-new-pdf   # one sample by name
 | `Samples/PageContent/` | Images, recolor, widgets, OCR |
 | `Samples/ImageFilters/` | Skia image filters |
 | `Samples/Barcodes/` | Barcode read/write |
-| `Samples/Llm/` | PDF4LLM markdown, tables, AI connector |
 | `Samples/Regression/` | Issue repros (diagnostics only) |
+| `Samples/Llm/` | MuPDF.NET.PDF4LLM: Markdown, tables, OCR, RAG |
 
-Input PDFs and images live in `TestDocuments/Demo/`. Generated PDFs are written to `TestDocuments/Demo/_Output/` (gitignored).
-
-## PDF4LLM AI sample
-
-The `ai-connector` sample uses `PdfExtractor.LoadAiAsync`. Set `AZURE_OPENAI_ENDPOINT` and related variables for Azure OpenAI, or run without them to use the in-memory demo pipeline.
+Input PDFs and images live in `TestDocuments/Demo/` (and `TestDocuments/Demo/Llm/`). Generated PDFs are written to `TestDocuments/Demo/_Output/` (gitignored).

Demo/SampleMenu.cs

@@ -1,7 +1,7 @@
 namespace Demo
 {
     /// <summary>
-    /// Demo samples grouped by MuPDF.NET / PDF4LLM feature areas.
+    /// Demo samples grouped by MuPDF.NET feature areas.
     /// Default run executes every entry in <see cref="Samples"/> (including <c>[diag]</c> samples).
     /// Use <c>user</c> to skip diagnostics, or <c>diagnostics</c> to run only <c>[diag]</c> samples.
     /// </summary>
@@ -61,30 +61,30 @@ private sealed record Sample(
             new("Barcodes", "write-barcode", "Write many barcode types to PDF and PNG", a => Program.TestWriteBarcode(a)),
             new("Barcodes", "write-barcode1", "Write CODE39/CODE128/DM with Units rects", _ => Program.TestWriteBarcode1()),
 
-            // —— PDF4LLM —— Samples/Llm
-            new("PDF4LLM", "rag-markdown", "PDF to Markdown (Magazine.pdf)", _ => Program.TestMuPdfRagToMarkdown()),
-            new("PDF4LLM", "table", "Detect tables and export markdown", _ => Program.TestTable()),
-            new("PDF4LLM", "table-extract-1", "Dump detected tables by page to console", _ => Program.TestTableExtract1()),
-            new("PDF4LLM", "table-extract-2", "Export detected tables to tables.csv", _ => Program.TestTableExtract2()),
-            new("PDF4LLM", "table-extract-3", "Merge continued table pages by column count", _ => Program.TestTableExtract3()),
-            new("PDF4LLM", "table-ocr", "Extract OCR text from Ocr.pdf", _ => Program.TestOcr()),
-            new("PDF4LLM", "llm-reader-save-pages", "Load markdown chunks and save per-page .md files", _ => Program.TestLLM2()),
-            new("PDF4LLM", "markdown-reader", "LlamaIndex PDFMarkdownReader", _ => Program.TestMarkdownReader()),
-            new("PDF4LLM", "ai-connector", "LoadAiAsync: Ask / Summarize / Search", a => Program.TestMicrosoftAiConnector(a).GetAwaiter().GetResult()),
-            new("PDF4LLM", "llm-to-markdown-fixture-370", "[diag] ToMarkdown vs test_370 expected output", a => Program.Test4LlmToMarkdownCompareExpected370(a), Diagnostic: true),
-            new("PDF4LLM", "llm-to-markdown-ocr-1", "[diag] ToMarkdown OCR fixture (FFFD)", a => Program.Test4LlmToMarkdownOcrFixture1(a), Diagnostic: true),
-            new("PDF4LLM", "llm-to-markdown-ocr-2", "[diag] ToMarkdown without OCR on FFFD fixture", a => Program.Test4LlmToMarkdownOcrFixture2(a), Diagnostic: true),
-            new("PDF4LLM", "llm-to-markdown-ocr-3", "[diag] ToMarkdown OCR on/off SVG fixture", a => Program.Test4LlmToMarkdownOcrFixture3(a), Diagnostic: true),
-            new("PDF4LLM", "llm-pdf-reader-empty", "[diag] PDFMarkdownReader empty page", a => Program.Test4LlmPdfMarkdownReaderEmptyPage(a), Diagnostic: true),
-            new("PDF4LLM", "llm-pdf-reader-missing-file", "[diag] PDFMarkdownReader missing file", a => Program.Test4LlmPdfMarkdownReaderMissingFile(a), Diagnostic: true),
-
             // —— Regression & diagnostics —— Samples/Regression
             new("Regression & diagnostics", "issue-213", "[diag] Drawing paths / line width", _ => Program.TestIssue213(), Diagnostic: true),
             new("Regression & diagnostics", "issue-1880", "[diag] Read Data Matrix barcodes", _ => Program.TestIssue1880(), Diagnostic: true),
             new("Regression & diagnostics", "issue-234", "[diag] Pixmap scale + insert image", _ => Program.TestIssue234(), Diagnostic: true),
             new("Regression & diagnostics", "pixmap-parallel", "[diag] Parallel Pixmap.ToBytes", _ => Program.TestPixmapParallel(), Diagnostic: true),
             new("Regression & diagnostics", "gettables-parallel", "[diag] Parallel Utils.GetTables", _ => Program.TestGetTablesParallel(), Diagnostic: true),
             new("Regression & diagnostics", "jbig2", "[diag] JBIG2 image recompression", _ => Program.TestRecompressJBIG2(), Diagnostic: true),
+
+            // —— MuPDF.NET.PDF4LLM —— Samples/Llm
+            new("MuPDF.NET.PDF4LLM", "rag-markdown", "PDF to Markdown (Magazine.pdf)", _ => Program.TestMuPdfRagToMarkdown()),
+            new("MuPDF.NET.PDF4LLM", "table", "Detect tables and export markdown", _ => Program.TestTable()),
+            new("MuPDF.NET.PDF4LLM", "table-extract-1", "Dump detected tables by page to console", _ => Program.TestTableExtract1()),
+            new("MuPDF.NET.PDF4LLM", "table-extract-2", "Export detected tables to tables.csv", _ => Program.TestTableExtract2()),
+            new("MuPDF.NET.PDF4LLM", "table-extract-3", "Merge continued table pages by column count", _ => Program.TestTableExtract3()),
+            new("MuPDF.NET.PDF4LLM", "table-ocr", "Extract OCR text from Ocr.pdf", _ => Program.TestOcr()),
+            new("MuPDF.NET.PDF4LLM", "llm-reader-save-pages", "Load markdown chunks and save per-page .md files", _ => Program.TestLLM2()),
+            new("MuPDF.NET.PDF4LLM", "markdown-reader", "LlamaIndex PDFMarkdownReader", _ => Program.TestMarkdownReader()),
+            new("MuPDF.NET.PDF4LLM", "ai-connector", "LoadAiAsync: Ask / Summarize / Search", a => Program.TestMicrosoftAiConnector(a).GetAwaiter().GetResult()),
+            new("MuPDF.NET.PDF4LLM", "llm-to-markdown-fixture-370", "[diag] ToMarkdown vs test_370 expected output", a => Program.Test4LlmToMarkdownCompareExpected370(a), Diagnostic: true),
+            new("MuPDF.NET.PDF4LLM", "llm-to-markdown-ocr-1", "[diag] ToMarkdown OCR fixture (FFFD)", a => Program.Test4LlmToMarkdownOcrFixture1(a), Diagnostic: true),
+            new("MuPDF.NET.PDF4LLM", "llm-to-markdown-ocr-2", "[diag] ToMarkdown without OCR on FFFD fixture", a => Program.Test4LlmToMarkdownOcrFixture2(a), Diagnostic: true),
+            new("MuPDF.NET.PDF4LLM", "llm-to-markdown-ocr-3", "[diag] ToMarkdown OCR on/off SVG fixture", a => Program.Test4LlmToMarkdownOcrFixture3(a), Diagnostic: true),
+            new("MuPDF.NET.PDF4LLM", "llm-pdf-reader-empty", "[diag] PDFMarkdownReader empty page", a => Program.Test4LlmPdfMarkdownReaderEmptyPage(a), Diagnostic: true),
+            new("MuPDF.NET.PDF4LLM", "llm-pdf-reader-missing-file", "[diag] PDFMarkdownReader missing file", a => Program.Test4LlmPdfMarkdownReaderMissingFile(a), Diagnostic: true),
         };
 
         private static readonly Dictionary<string, Sample> ByName = BuildIndex();

Demo/Samples/Llm/Program.Llm.AiConnector.cs

@@ -1,19 +1,20 @@
 using Microsoft.Extensions.AI;
-using PDF4LLM.AI;
-using PDF4LLM.AI.Models;
-using PDF4LLM.AI.Options;
+using MuPDF.NET.PDF4LLM;
+using MuPDF.NET.PDF4LLM.AI;
+using MuPDF.NET.PDF4LLM.AI.Models;
+using MuPDF.NET.PDF4LLM.AI.Options;
 
 namespace Demo
 {
     /// <summary>
-    /// <see cref="PdfExtractor.LoadAiAsync"/> demo: index two PDFs, then Ask, Summarize, and Search.
+    /// <see cref="MuPDF.NET.PDF4LLM.MuPDF4LLM.LoadAiAsync"/> demo: index two PDFs, then Ask, Summarize, and Search.
     /// Uses Azure OpenAI when <c>AZURE_OPENAI_*</c> env vars are set; otherwise an offline demo client.
     /// </summary>
     internal partial class Program
     {
         internal static async Task TestMicrosoftAiConnector(string[] args)
         {
-            Console.WriteLine("\n=== TestMicrosoftAiConnector (PDF4LLM.AI) =======================");
+            Console.WriteLine("\n=== TestMicrosoftAiConnector (MuPDF.NET.PDF4LLM.AI) =======================");
 
             string capitals = DemoPaths.Input("Llm/national-capitals.pdf");
             string nato = DemoPaths.Input("Llm/nato-members.pdf");
@@ -22,7 +23,7 @@ internal static async Task TestMicrosoftAiConnector(string[] args)
             bool azure = !string.IsNullOrWhiteSpace(Environment.GetEnvironmentVariable("AZURE_OPENAI_ENDPOINT"));
             Console.WriteLine($"Pipeline: {(azure ? "Azure OpenAI" : "development (in-memory index + demo chat)")}");
 
-            AiDocumentCollection aiDocs = await PdfExtractor.LoadAiAsync(
+            AiDocumentCollection aiDocs = await MuPDF4LLM.LoadAiAsync(
                 new[] { capitals, nato },
                 options);
 

Demo/Samples/Llm/Program.Llm.PdfMarkdownReader.Fixtures.cs

@@ -1,7 +1,7 @@
 namespace Demo
 {
     /// <summary>
-    /// <see cref="PDFMarkdownReader"/> demos aligned with PDF4LLM / repository reader tests.
+    /// <see cref="PDFMarkdownReader"/> demos aligned with MuPDF.NET.PDF4LLM / repository reader tests.
     /// </summary>
     internal partial class Program
     {

Demo/Samples/Llm/Program.Llm.TableExtract.cs

@@ -8,8 +8,8 @@ internal partial class Program
 
         internal static void TestTableExtract1()
         {
-            PdfExtractor.UseLayout = true;
-            JArray pages = GetPagesFromJson(PdfExtractor.ToJson(NationalCapitalsPdf));
+            MuPDF4LLM.UseLayout = true;
+            JArray pages = GetPagesFromJson(MuPDF4LLM.ToJson(NationalCapitalsPdf));
 
             foreach (JObject page in pages)
             {
@@ -35,8 +35,8 @@ internal static void TestTableExtract1()
 
         internal static void TestTableExtract2()
         {
-            PdfExtractor.UseLayout = true;
-            JArray pages = GetPagesFromJson(PdfExtractor.ToJson(NationalCapitalsPdf));
+            MuPDF4LLM.UseLayout = true;
+            JArray pages = GetPagesFromJson(MuPDF4LLM.ToJson(NationalCapitalsPdf));
             var csvLines = new List<string>();
 
             foreach (JObject page in pages)
@@ -64,8 +64,8 @@ internal static void TestTableExtract2()
 
         internal static void TestTableExtract3()
         {
-            PdfExtractor.UseLayout = true;
-            JArray pages = GetPagesFromJson(PdfExtractor.ToJson(NationalCapitalsPdf));
+            MuPDF4LLM.UseLayout = true;
+            JArray pages = GetPagesFromJson(MuPDF4LLM.ToJson(NationalCapitalsPdf));
             var mergedRows = new List<List<string>>();
             int? prevColCount = null;
 
@@ -101,25 +101,27 @@ internal static void TestTableExtract3()
 
         internal static void TestOcr()
         {
-            PdfExtractor.UseLayout = true;
-            string md = PdfExtractor.ToMarkdown(@"..\..\..\..\TestDocuments\Demo\Ocr.pdf", useOcr: true, writeImages: false, embedImages: false);
+            MuPDF4LLM.UseLayout = true;
+            string ocrPdf = DemoPaths.Input("Ocr.pdf");
+            string md = MuPDF4LLM.ToMarkdown(ocrPdf, useOcr: true, writeImages: false, embedImages: false);
             Console.WriteLine(md);
-            string text = PdfExtractor.ToText(@"..\..\..\..\TestDocuments\Demo\Ocr.pdf", useOcr: true);
+            string text = MuPDF4LLM.ToText(ocrPdf, useOcr: true);
             Console.WriteLine(text);
         }
 
         internal static void TestLLM2()
         {
-            PdfExtractor.UseLayout = true;
-            var reader = PdfExtractor.LlamaMarkdownReader();
-            var chunks = reader.LoadData(@"..\..\..\..\TestDocuments\Demo\magazine.pdf");
+            MuPDF4LLM.UseLayout = true;
+            var reader = MuPDF4LLM.LlamaMarkdownReader();
+            var chunks = reader.LoadData(DemoPaths.Input("Magazine.pdf"));
 
-            Directory.CreateDirectory("Output");
+            string outDir = DemoPaths.Output("pages");
+            Directory.CreateDirectory(outDir);
             foreach (var chunk in chunks)
             {
                 int pageNum = (int)chunk.ExtraInfo["page"];
                 Console.WriteLine(pageNum);
-                string filePath = $"output/page-{pageNum}.md";
+                string filePath = Path.Combine(outDir, $"page-{pageNum}.md");
                 File.WriteAllText(filePath, chunk.Text, Encoding.UTF8);
             }
         }

Demo/Samples/Llm/Program.Llm.ToMarkdown.Fixtures.cs

@@ -1,7 +1,7 @@
 namespace Demo
 {
     /// <summary>
-    /// PDF4LLM <see cref="PDF4LLM.PdfExtractor.ToMarkdown"/> demos aligned with PDF4LLM test fixtures
+    /// MuPDF.NET.PDF4LLM <see cref="MuPDF.NET.PDF4LLM.MuPDF4LLM.ToMarkdown"/> demos aligned with MuPDF.NET.PDF4LLM test fixtures
     /// (golden markdown, OCR behavior). PDFs live under <c>TestDocuments/Demo/Llm/</c>; samples skip if missing.
     /// </summary>
     internal partial class Program
@@ -13,7 +13,7 @@ private static bool LlmOcrEnvironmentLikelyAvailable() =>
         internal static void Test4LlmToMarkdownCompareExpected370(string[] args)
         {
             _ = args;
-            Console.WriteLine("\n=== Test4LlmToMarkdownCompareExpected370 (PDF4LLM) =======================");
+            Console.WriteLine("\n=== Test4LlmToMarkdownCompareExpected370 (MuPDF.NET.PDF4LLM) =======================");
 
             string pdfPath = DemoPaths.Input("Llm/test_370.pdf");
             string expectedPath = DemoPaths.Input("Llm/test_370_expected.md");
@@ -27,7 +27,7 @@ internal static void Test4LlmToMarkdownCompareExpected370(string[] args)
             Document document = new Document(pdfPath);
             try
             {
-                string actual = ToMarkdown(
+                string actual = MuPDF4LLM.ToMarkdown(
                     document,
                     header: false,
                     footer: false,
@@ -75,7 +75,7 @@ internal static void Test4LlmToMarkdownOcrFixture1(string[] args)
             string md;
             try
             {
-                md = ToMarkdown(doc);
+                md = MuPDF4LLM.ToMarkdown(doc);
             }
             finally
             {
@@ -104,7 +104,7 @@ internal static void Test4LlmToMarkdownOcrFixture1(string[] args)
             }
         }
 
-        /// <summary><c>ToMarkdown(..., useOcr: false)</c> on FFFD fixture.</summary>
+        /// <summary><c>MuPDF4LLM.ToMarkdown(..., useOcr: false)</c> on FFFD fixture.</summary>
         internal static void Test4LlmToMarkdownOcrFixture2(string[] args)
         {
             _ = args;
@@ -121,7 +121,7 @@ internal static void Test4LlmToMarkdownOcrFixture2(string[] args)
             string md;
             try
             {
-                md = ToMarkdown(doc, useOcr: false);
+                md = MuPDF4LLM.ToMarkdown(doc, useOcr: false);
             }
             finally
             {
@@ -151,8 +151,8 @@ internal static void Test4LlmToMarkdownOcrFixture3(string[] args)
             string mdNoOcr;
             try
             {
-                md = ToMarkdown(doc);
-                mdNoOcr = ToMarkdown(doc, useOcr: false);
+                md = MuPDF4LLM.ToMarkdown(doc);
+                mdNoOcr = MuPDF4LLM.ToMarkdown(doc, useOcr: false);
             }
             finally
             {