Fix PDF spec compliance, git line-ending corruption, and add xref validation test

.gitattributes

@@ -16,7 +16,7 @@
 *.DOCX diff=astextplain
 *.dot  diff=astextplain
 *.DOT  diff=astextplain
-*.pdf  diff=astextplain
-*.PDF	 diff=astextplain
+*.pdf  binary
+*.PDF  binary
 *.rtf	 diff=astextplain
 *.RTF	 diff=astextplain

QRCoder/PdfByteQRCode.cs

@@ -72,7 +72,7 @@ public byte[] GetGraphic(int pixelsPerModule, string darkColorHtmlHex, string li
 
             // Binary comment - ensures PDF is treated as binary file (prevents text mode corruption)
             stream.Write(_pdfBinaryComment, 0, _pdfBinaryComment.Length);
-            writer.WriteLine();
+            writer.Write("\r\n");
 
             writer.Flush();
             xrefs.Add(stream.Position);
@@ -92,18 +92,12 @@ public byte[] GetGraphic(int pixelsPerModule, string darkColorHtmlHex, string li
 
             // Object 2: Pages - defines page tree structure
             writer.Write(
-                ToStr(xrefs.Count) + " 0 obj\r\n" +                                     // Object number and generation number (0)
-                "<<\r\n" +                                                              // Begin dictionary
-                "/Count 1\r\n" +                                                        // Number of pages in document
-                "/Kids [ <<\r\n" +                                                      // Array of page objects - begin inline page dictionary
-                    "/Type /Page\r\n" +                                                 // Declares this as a page
-                    "/Parent 2 0 R\r\n" +                                               // References parent Pages object
-                    "/MediaBox [0 0 " + pdfMediaSize + " " + pdfMediaSize + "]\r\n" +   // Page dimensions [x1 y1 x2 y2]
-                    "/Resources << /ProcSet [ /PDF ] >>\r\n" +                          // Required resources: PDF operations only (no images)
-                    "/Contents 3 0 R\r\n" +                                             // References content stream (object 3)
-                    ">> ]\r\n" +                                                        // End inline page dictionary and Kids array
-                ">>\r\n" +                                                              // End dictionary
-                "endobj\r\n"                                                            // End object
+                ToStr(xrefs.Count) + " 0 obj\r\n" +   // Object number and generation number (0)
+                "<<\r\n" +                             // Begin dictionary
+                "/Count 1\r\n" +                       // Number of pages in document
+                "/Kids [ 3 0 R ]\r\n" +                // Kids must contain indirect references to Page objects
+                ">>\r\n" +                             // End dictionary
+                "endobj\r\n"                           // End object
             );
 
             // Content stream - PDF drawing instructions
@@ -122,13 +116,29 @@ public byte[] GetGraphic(int pixelsPerModule, string darkColorHtmlHex, string li
             writer.Flush();
             xrefs.Add(stream.Position);
 
-            // Object 3: Content stream - contains the drawing instructions
+            // Object 3: Page - indirect page object (Kids array must reference pages indirectly)
+            writer.Write(
+                ToStr(xrefs.Count) + " 0 obj\r\n" +                                     // Object number and generation number (0)
+                "<<\r\n" +                                                              // Begin dictionary
+                "/Type /Page\r\n" +                                                     // Declares this as a page
+                "/Parent 2 0 R\r\n" +                                                   // References parent Pages object
+                "/MediaBox [0 0 " + pdfMediaSize + " " + pdfMediaSize + "]\r\n" +       // Page dimensions [x1 y1 x2 y2]
+                "/Resources << /ProcSet [ /PDF ] >>\r\n" +                              // Required resources: PDF operations only (no images)
+                "/Contents 4 0 R\r\n" +                                                 // References content stream (object 4)
+                ">>\r\n" +                                                              // End dictionary
+                "endobj\r\n"                                                            // End object
+            );
+
+            writer.Flush();
+            xrefs.Add(stream.Position);
+
+            // Object 4: Content stream - contains the drawing instructions
             writer.Write(
-                ToStr(xrefs.Count) + " 0 obj\r\n" +                  // Object number and generation number (0)
-                "<< /Length " + ToStr(content.Length) + " >>\r\n" +  // Dictionary with stream length in bytes
-                "stream\r\n" +                                       // Begin stream data
-                content + "endstream\r\n" +                          // Stream content followed by end stream marker
-                "endobj\r\n"                                         // End object
+                ToStr(xrefs.Count) + " 0 obj\r\n" +                                                   // Object number and generation number (0)
+                "<< /Length " + ToStr(System.Text.Encoding.ASCII.GetByteCount(content)) + " >>\r\n" + // Dictionary with stream length in bytes
+                "stream\r\n" +                                                                        // Begin stream data
+                content + "endstream\r\n" +                                                           // Stream content followed by end stream marker
+                "endobj\r\n"                                                                          // End object
             );
 
             writer.Flush();

QRCoderTests/PdfByteQRCodeRendererTests.cs

@@ -1,6 +1,4 @@
-using QRCoder;
-using Shouldly;
-using Xunit;
+using System.IO;
 
 namespace QRCoderTests;
 
@@ -54,4 +52,97 @@ public void can_render_pdfbyte_qrcode_from_helper_2()
         var pdfCodeGfx = PdfByteQRCodeHelper.GetQRCode("This is a quick test! 123#?", 5, "#FF0000", "#0000FF", QRCodeGenerator.ECCLevel.L);
         pdfCodeGfx.ShouldMatchApproved("pdf");
     }
+
+    private static readonly char[] _lineEndChars = { '\r', '\n' };
+
+    [Fact]
+    public void pdf_xref_table_is_valid()
+    {
+        var gen = new QRCodeGenerator();
+        var data = gen.CreateQrCode("This is a quick test! 123#?", QRCodeGenerator.ECCLevel.L);
+        var pdfBytes = new PdfByteQRCode(data).GetGraphic(5);
+
+        // Parse from the end to find startxref
+        var pdfText = Encoding.ASCII.GetString(pdfBytes);
+
+        // Verify no \n line breaks; only \r\n should be used (this test file has no binary image data)
+        pdfText.Replace("\r\n", "CRLF").ShouldNotContain('\n', "PDF should not contain LF line breaks; only CRLF should be used");
+        pdfText.Replace("\r\n", "CRLF").ShouldNotContain('\r', "PDF should not contain CR line breaks; only CRLF should be used");
+
+        // Find %%EOF at the end, then work backward to find startxref
+        var eofIndex = pdfText.LastIndexOf("%%EOF", StringComparison.Ordinal);
+        eofIndex.ShouldBeGreaterThan(0, "%%EOF not found");
+
+        var startxrefIndex = pdfText.LastIndexOf("startxref\r\n", eofIndex, StringComparison.Ordinal);
+        startxrefIndex.ShouldBeGreaterThan(0, "startxref not found");
+
+        // Read the xref byte offset (the number on the line after "startxref")
+        var afterStartxref = startxrefIndex + "startxref\r\n".Length;
+        var endOfOffset = pdfText.IndexOf("\r\n", afterStartxref, StringComparison.Ordinal);
+        var xrefOffsetStr = pdfText.Substring(afterStartxref, endOfOffset - afterStartxref);
+        var xrefOffset = int.Parse(xrefOffsetStr, NumberStyles.None, CultureInfo.InvariantCulture);
+        xrefOffset.ShouldBeGreaterThan(0, "xref byte offset should be positive");
+
+        // Seek to xref table and parse it
+        using var stream = new MemoryStream(pdfBytes);
+        stream.Position = xrefOffset;
+        var reader = new StreamReader(stream, Encoding.ASCII, detectEncodingFromByteOrderMarks: false, bufferSize: 1024, leaveOpen: true);
+
+        // First line must be "xref"
+        var xrefLine = reader.ReadLine();
+        xrefLine.ShouldBe("xref", "xref keyword not found at expected offset");
+
+        // Parse subsections: "firstObjNum count"
+        var objectOffsets = new Dictionary<int, long>();
+        string? subsectionLine;
+        while ((subsectionLine = reader.ReadLine()) != null && subsectionLine != "trailer")
+        {
+            var parts = subsectionLine.Split(' ');
+            parts.Length.ShouldBe(2, $"Expected 'firstObj count' but got: {subsectionLine}");
+            var firstObj = int.Parse(parts[0], NumberStyles.None, CultureInfo.InvariantCulture);
+            firstObj.ShouldBe(0);
+            var count = int.Parse(parts[1], NumberStyles.None, CultureInfo.InvariantCulture);
+
+            for (int i = 0; i < count; i++)
+            {
+                // Each entry: "NNNNNNNNNN GGGGG f\r\n" or "NNNNNNNNNN GGGGG n\r\n"
+                var entry = reader.ReadLine();
+                entry.ShouldNotBeNull();
+                entry.Length.ShouldBe(18);
+                var entryParts = entry.Split(' ');
+                entryParts.Length.ShouldBe(3, $"Expected 'offset gen type' but got: {entry}");
+                var offset = long.Parse(entryParts[0], NumberStyles.None, CultureInfo.InvariantCulture);
+                var generation = int.Parse(entryParts[1], NumberStyles.None, CultureInfo.InvariantCulture);
+                var type = entryParts[2];
+                type.ShouldBeOneOf("n", "f");
+
+                if (type == "n")
+                {
+                    generation.ShouldBe(0, $"Expected generation 0 for in-use object but got {generation}");
+                    objectOffsets[i] = offset;
+                }
+                else
+                {
+                    // Free objects should only be listed for the first object in the subsection
+                    i.ShouldBe(0);
+                    offset.ShouldBe(0);
+                    generation.ShouldBe(65535, $"Expected generation 65535 for free object but got {generation}");
+                }
+            }
+        }
+
+        objectOffsets.Count.ShouldBeGreaterThan(0, "No in-use objects found in xref table");
+
+        // Verify each object: seek to its offset and confirm "N 0 obj" is present
+        foreach (var kvp in objectOffsets)
+        {
+            stream.Position = kvp.Value;
+            var objNum = kvp.Key;
+            var offset = kvp.Value;
+            var objReader = new StreamReader(stream, Encoding.ASCII, detectEncodingFromByteOrderMarks: false, bufferSize: 1024, leaveOpen: true);
+            var objLine = objReader.ReadLine();
+            objLine.ShouldNotBeNull($"No content at offset {offset} for object {objNum}");
+            objLine.ShouldBe($"{objNum} 0 obj", $"Object {objNum} at offset {offset} did not start with '{objNum} 0 obj'");
+        }
+    }
 }