adding support to tif images

src/UglyToad.PdfPig.Core/UglyToad.PdfPig.Core.csproj

@@ -1,4 +1,4 @@
-<Project Sdk="Microsoft.NET.Sdk">
+<Project Sdk="Microsoft.NET.Sdk">
   <PropertyGroup>
     <TargetFrameworks>netstandard2.0;netstandard2.1;net462;net471;net6.0;net8.0;net9.0</TargetFrameworks>
     <LangVersion>12</LangVersion>

src/UglyToad.PdfPig.DocumentLayoutAnalysis/UglyToad.PdfPig.DocumentLayoutAnalysis.csproj

@@ -1,4 +1,4 @@
-<Project Sdk="Microsoft.NET.Sdk">
+<Project Sdk="Microsoft.NET.Sdk">
   <PropertyGroup>
     <TargetFrameworks>netstandard2.0;netstandard2.1;net462;net471;net6.0;net8.0;net9.0</TargetFrameworks>
     <LangVersion>12</LangVersion>

src/UglyToad.PdfPig.Fonts/UglyToad.PdfPig.Fonts.csproj

@@ -1,4 +1,4 @@
-<Project Sdk="Microsoft.NET.Sdk">
+<Project Sdk="Microsoft.NET.Sdk">
   <PropertyGroup>
     <TargetFrameworks>netstandard2.0;netstandard2.1;net462;net471;net6.0;net8.0;net9.0</TargetFrameworks>
     <LangVersion>12</LangVersion>

src/UglyToad.PdfPig.Tests/Images/Files/Tif/TiffCcittG4.fixture.json

@@ -0,0 +1,9 @@
+{
+  "Width": 1800,
+  "Height": 3113,
+  "BlackIs1": false,
+  "Photometric": 0,
+  "FillOrder": 1,
+  "PayloadLength": 329546,
+  "SourceFile": "TiffCcittG4.tif"
+}

src/UglyToad.PdfPig.Tests/UglyToad.PdfPig.Tests.csproj

@@ -1,5 +1,4 @@
 <Project Sdk="Microsoft.NET.Sdk">
-
     <PropertyGroup>
         <TargetFrameworks>net471;net8.0;net9.0</TargetFrameworks>
         <IsTestProject>true</IsTestProject>
@@ -198,6 +197,12 @@
         <None Update="Images\Files\Jpx\fd42e6a0-5c7a-4eb2-b0e3-474cfde067a6.jp2">
             <CopyToOutputDirectory>Never</CopyToOutputDirectory>
         </None>
+        <Content Include="Images\Files\Tif\TiffCcittG4.fixture.json">
+            <CopyToOutputDirectory>PreserveNewest</CopyToOutputDirectory>
+        </Content>
+        <Content Include="Images\Files\Tif\TiffCcittG4.ccitt.base64">
+            <CopyToOutputDirectory>PreserveNewest</CopyToOutputDirectory>
+        </Content>
     </ItemGroup>
     <ItemGroup>
         <Using Include="Xunit" />

src/UglyToad.PdfPig.Tests/Writer/PdfPageBuilderCcittG4Tests.cs

@@ -0,0 +1,110 @@
+namespace UglyToad.PdfPig.Tests.Writer
+{
+    using System.IO;
+    using UglyToad.PdfPig.Content;
+    using System.Linq;
+    using UglyToad.PdfPig;
+    using UglyToad.PdfPig.Tests.Writer.TestImages;
+    using UglyToad.PdfPig.Tokens;
+    using UglyToad.PdfPig.Writer;
+    using Xunit;
+
+    public class PdfPageBuilderCcittG4Tests
+    {
+        [Fact]
+        public void FixtureProvidesRawCcittData()
+        {
+            var fixture = CcittG4TestImage.Load();
+
+            Assert.True(fixture.Width > 0);
+            Assert.True(fixture.Height > 0);
+            Assert.NotEmpty(fixture.RawCcittData);
+            Assert.False(fixture.BlackIs1);
+        }
+
+        [Fact]
+        public void AddCcittG4AddsFaxImage()
+        {
+            var fixture = CcittG4TestImage.Load();
+
+            byte[] pdfBytes;
+            using (var documentBuilder = new PdfDocumentBuilder())
+            {
+                var pageBuilder = documentBuilder.AddPage(fixture.Width, fixture.Height);
+
+                pageBuilder.AddCcittG4(fixture.RawCcittData, fixture.Width, fixture.Height, blackIs1: fixture.BlackIs1);
+
+                pdfBytes = documentBuilder.Build();
+            }
+
+            File.WriteAllBytes(@"PdfPageBuilderTests_CanAddTifCcittG4.pdf", pdfBytes);
+
+            using (var document = PdfDocument.Open(pdfBytes))
+            {
+                var page = document.GetPage(1);
+                var image = Assert.Single(page.GetImages());
+
+                Assert.Equal(fixture.Width, image.WidthInSamples);
+                Assert.Equal(fixture.Height, image.HeightInSamples);
+                Assert.Equal(1, image.BitsPerComponent);
+            }
+        }
+
+        [Fact]
+        public void FixtureWritesDecodeZeroOne()
+        {
+            var fixture = CcittG4TestImage.Load();
+            var image = BuildFaxImage(fixture.BlackIs1);
+
+            var decode = Assert.IsType<ArrayToken>(image.ImageDictionary.Data[NameToken.Decode]);
+
+            // The generated fixture comes from a TIFF with Photometric.MINISWHITE, which must map to
+            // blackIs1 = false for AddCcittG4. Reversing this to match TIFF naming reintroduces negative PDFs.
+            Assert.Equal([0, 1], decode.Data.OfType<NumericToken>().Select(x => x.Data).ToArray());
+        }
+
+        [Fact]
+        public void AddCcittG4WithBlackIs1TrueWritesDecodeOneZero()
+        {
+            var image = BuildFaxImage(blackIs1: true);
+
+            var decode = Assert.IsType<ArrayToken>(image.ImageDictionary.Data[NameToken.Decode]);
+
+            Assert.Equal([1, 0], decode.Data.OfType<NumericToken>().Select(x => x.Data).ToArray());
+            Assert.True(image.ImageDictionary.TryGet(NameToken.DecodeParms, out DictionaryToken? decodeParms));
+            Assert.False(decodeParms!.Data.ContainsKey(NameToken.BlackIs1));
+        }
+
+        [Fact]
+        public void AddCcittG4WithBlackIs1FalseWritesDecodeZeroOne()
+        {
+            var image = BuildFaxImage(blackIs1: false);
+
+            var decode = Assert.IsType<ArrayToken>(image.ImageDictionary.Data[NameToken.Decode]);
+
+            Assert.Equal([0, 1], decode.Data.OfType<NumericToken>().Select(x => x.Data).ToArray());
+            Assert.True(image.ImageDictionary.TryGet(NameToken.DecodeParms, out DictionaryToken? decodeParms));
+            Assert.False(decodeParms!.Data.ContainsKey(NameToken.BlackIs1));
+        }
+
+        private static IPdfImage BuildFaxImage(bool blackIs1)
+        {
+            var fixture = CcittG4TestImage.Load();
+
+            byte[] pdfBytes;
+            using (var documentBuilder = new PdfDocumentBuilder())
+            {
+                var pageBuilder = documentBuilder.AddPage(fixture.Width, fixture.Height);
+
+                pageBuilder.AddCcittG4(fixture.RawCcittData, fixture.Width, fixture.Height, blackIs1: blackIs1);
+
+                pdfBytes = documentBuilder.Build();
+            }
+
+            using var document = PdfDocument.Open(pdfBytes);
+            var page = document.GetPage(1);
+
+            return Assert.Single(page.GetImages());
+        }
+    }
+}

src/UglyToad.PdfPig.Tests/Writer/TestImages/CcittG4TestImage.cs

@@ -0,0 +1,92 @@
+namespace UglyToad.PdfPig.Tests.Writer.TestImages
+{
+    using System;
+    using System.IO;
+    using System.Text;
+    using System.Text.RegularExpressions;
+
+    /// <summary>
+    /// Helper for loading the CCITT Group 4 fixture used by PDF page builder tests.
+    /// </summary>
+    internal sealed class CcittG4TestImage
+    {
+        private CcittG4TestImage(int width, int height, byte[] rawCcittData, bool blackIs1)
+        {
+            Width = width;
+            Height = height;
+            RawCcittData = rawCcittData;
+            BlackIs1 = blackIs1;
+        }
+
+        public int Width { get; }
+
+        public int Height { get; }
+
+        public byte[] RawCcittData { get; }
+
+        public bool BlackIs1 { get; }
+
+        /// <summary>
+        /// Loads the CCITT Group 4 sample payload relative to the test output directory.
+        /// The payload is already raw CCITT data; width, height and polarity come from the generated metadata
+        /// derived from the original TIFF using the same extraction logic as the external converter.
+        /// </summary>
+        public static CcittG4TestImage Load()
+        {
+            var metadataPath = Path.GetFullPath(Path.Combine(AppDomain.CurrentDomain.BaseDirectory, "Images", "Files", "Tif",
+                "TiffCcittG4.fixture.json"));
+            var base64Path = Path.GetFullPath(Path.Combine(AppDomain.CurrentDomain.BaseDirectory, "Images", "Files", "Tif",
+                "TiffCcittG4.ccitt.base64"));
+            var metadataJson = File.ReadAllText(metadataPath, Encoding.UTF8);
+            var metadata = CcittG4FixtureMetadata.Parse(metadataJson);
+            var base64 = File.ReadAllText(base64Path, Encoding.ASCII)
+                .Replace("\r", string.Empty)
+                .Replace("\n", string.Empty)
+                .Trim();
+            var rawCcittData = Convert.FromBase64String(base64);
+
+            return new CcittG4TestImage(metadata.Width, metadata.Height, rawCcittData, metadata.BlackIs1);
+        }
+
+        private sealed class CcittG4FixtureMetadata
+        {
+            public int Width { get; private set; }
+
+            public int Height { get; private set; }
+
+            public bool BlackIs1 { get; private set; }
+
+            public static CcittG4FixtureMetadata Parse(string json)
+            {
+                return new CcittG4FixtureMetadata
+                {
+                    Width = ReadInt(json, "Width"),
+                    Height = ReadInt(json, "Height"),
+                    BlackIs1 = ReadBool(json, "BlackIs1")
+                };
+            }
+
+            private static int ReadInt(string json, string propertyName)
+            {
+                var match = Regex.Match(json, $"\"{propertyName}\"\\s*:\\s*(\\d+)");
+                if (!match.Success)
+                {
+                    throw new InvalidOperationException($"Missing integer property '{propertyName}' in CCITT fixture metadata.");
+                }
+
+                return int.Parse(match.Groups[1].Value);
+            }
+
+            private static bool ReadBool(string json, string propertyName)
+            {
+                var match = Regex.Match(json, $"\"{propertyName}\"\\s*:\\s*(true|false)", RegexOptions.IgnoreCase);
+                if (!match.Success)
+                {
+                    throw new InvalidOperationException($"Missing boolean property '{propertyName}' in CCITT fixture metadata.");
+                }
+
+                return bool.Parse(match.Groups[1].Value);
+            }
+        }
+    }
+}

src/UglyToad.PdfPig.Tokenization/UglyToad.PdfPig.Tokenization.csproj

@@ -1,4 +1,4 @@
-<Project Sdk="Microsoft.NET.Sdk">
+<Project Sdk="Microsoft.NET.Sdk">
   <PropertyGroup>
     <TargetFrameworks>netstandard2.0;netstandard2.1;net462;net471;net6.0;net8.0;net9.0</TargetFrameworks>
     <LangVersion>12</LangVersion>

src/UglyToad.PdfPig.Tokens/UglyToad.PdfPig.Tokens.csproj

@@ -1,4 +1,4 @@
-<Project Sdk="Microsoft.NET.Sdk">
+<Project Sdk="Microsoft.NET.Sdk">
   <PropertyGroup>
     <TargetFrameworks>netstandard2.0;netstandard2.1;net462;net471;net6.0;net8.0;net9.0</TargetFrameworks>
     <LangVersion>12</LangVersion>

src/UglyToad.PdfPig/Writer/PdfPageBuilder.cs

@@ -735,7 +735,7 @@ public AddedImage AddPng(Stream pngStream, PdfRectangle placementRectangle = def
 
             var data = new byte[png.Width * png.Height * 3];
             int pixelIndex = 0;
-          
+
             for (var rowIndex = 0; rowIndex < png.Height; rowIndex++)
             {
                 for (var colIndex = 0; colIndex < png.Width; colIndex++)
@@ -828,6 +828,88 @@ public AddedImage AddPng(Stream pngStream, PdfRectangle placementRectangle = def
             return new AddedImage(reference.Data, png.Width, png.Height);
         }
 
+
+        /// <summary>
+        /// Adds a CCITT Group 4 (fax) encoded 1bpp image stream as a PDF image XObject (CCITTFaxDecode).
+        /// This is ideal for monochrome scanned TIFF pages that are already CCITT G4 compressed.
+        /// </summary>
+        /// <param name="ccittG4Data">
+        /// Raw CCITT Group 4 (T.6) compressed bytes (NOT decoded bitmap pixels, NOT wrapped in TIFF).
+        /// </param>
+        /// <param name="width">Image width in pixels (Columns).</param>
+        /// <param name="height">Image height in pixels (Rows).</param>
+        /// <param name="placementRectangle">
+        /// Placement rectangle in user space. If default, uses (0,0,width,height).
+        /// </param>
+        /// <param name="blackIs1">
+        /// Set to true if 1 bits represent black (common for bilevel scans).
+        /// If the result looks inverted, pass false.
+        /// </param>
+        public AddedImage AddCcittG4(byte[] ccittG4Data, int width, int height, PdfRectangle placementRectangle = default, bool blackIs1 = true)
+        {
+            if (ccittG4Data is null) throw new ArgumentNullException(nameof(ccittG4Data));
+            if (width <= 0) throw new ArgumentOutOfRangeException(nameof(width));
+            if (height <= 0) throw new ArgumentOutOfRangeException(nameof(height));
+
+            if (placementRectangle.Equals(default(PdfRectangle)))
+            {
+                placementRectangle = new PdfRectangle(0, 0, width, height);
+            }
+
+            // DecodeParms for CCITTFaxDecode:
+            // K = -1 => Group 4 (T.6)
+            // Columns/Rows = image dimensions
+            // BlackIs1 controls polarity via /Decode.
+            var decodeParms = new Dictionary<NameToken, IToken>
+            {
+                { NameToken.Create("K"), new NumericToken(-1) },
+                { NameToken.Create("Columns"), new NumericToken(width) },
+                { NameToken.Create("Rows"), new NumericToken(height) }
+            };
+
+            var imgDictionary = new Dictionary<NameToken, IToken>
+            {
+                { NameToken.Type, NameToken.Xobject },
+                { NameToken.Subtype, NameToken.Image },
+                { NameToken.Width, new NumericToken(width) },
+                { NameToken.Height, new NumericToken(height) },
+                { NameToken.ColorSpace, NameToken.Devicegray },
+                { NameToken.BitsPerComponent, new NumericToken(1) },
+                { NameToken.Filter, NameToken.Create("CCITTFaxDecode") },
+                { NameToken.DecodeParms, new DictionaryToken(decodeParms) },
+                { NameToken.Length, new NumericToken(ccittG4Data.Length) }
+            };
+
+            imgDictionary[NameToken.Decode] = new ArrayToken(
+            [
+                blackIs1 ? new NumericToken(1) : new NumericToken(0),
+                blackIs1 ? new NumericToken(0) : new NumericToken(1)
+            ]);
+
+
+            // IMPORTANT: Do NOT recompress. ccittG4Data is already compressed with CCITT Group 4.
+            var reference = documentBuilder.AddImage(new DictionaryToken(imgDictionary), ccittG4Data);
+
+            var resources = pageDictionary.GetOrCreateDict(NameToken.Resources);
+            var xObjects = resources.GetOrCreateDict(NameToken.Xobject);
+
+            var key = NameToken.Create(xobjectsNames.NewName());
+            xObjects[key] = reference;
+
+            currentStream.Add(Push.Value);
+            currentStream.Add(new ModifyCurrentTransformationMatrix(
+            [
+                placementRectangle.Width, 0,
+                0, placementRectangle.Height,
+                placementRectangle.BottomLeft.X, placementRectangle.BottomLeft.Y
+            ]));
+            currentStream.Add(new InvokeNamedXObject(key));
+            currentStream.Add(Pop.Value);
+
+            return new AddedImage(reference.Data, width, height);
+        }
+
+
         /// <summary>
         /// Adds a URL link annotation to the page at the specified rectangle area.
         /// </summary>
@@ -1034,7 +1116,7 @@ public PdfPageBuilder CopyFrom(Page srcPage)
 
                         gstateName = newName;
                     }
-                    
+
                     // According to PDF spec 32000-1:2008, section 8.4.5, ExtGState can contain both direct values and indirect references
                     if (gstate.Value is IndirectReferenceToken fontReferenceToken)
                     {

tools/CcittFixtureGenerator/CcittFixtureGenerator.csproj

@@ -0,0 +1,12 @@
+<Project Sdk="Microsoft.NET.Sdk">
+  <PropertyGroup>
+    <OutputType>Exe</OutputType>
+    <TargetFramework>net8.0</TargetFramework>
+    <ImplicitUsings>enable</ImplicitUsings>
+    <Nullable>enable</Nullable>
+  </PropertyGroup>
+
+  <ItemGroup>
+    <PackageReference Include="BitMiracle.LibTiff.NET" Version="2.4.660" />
+  </ItemGroup>
+</Project>