fix bug with computing text positions

src/UglyToad.Pdf.Tests/Integration/SinglePageNonLatinAcrobatDistillerTests.cs

@@ -1,6 +1,7 @@
 namespace UglyToad.Pdf.Tests.Integration
 {
     using System;
+    using System.Collections.Generic;
     using System.IO;
     using System.Linq;
     using Content;
@@ -51,5 +52,66 @@
                 Assert.Equal("Hello ﺪﻤﺤﻣ World. ", text);
             }
         }
+
+        [Fact]
+        public void LetterPositionsAreCorrectPdfBox()
+        {
+            using (var document = PdfDocument.Open(GetFilename()))
+            {
+                var page = document.GetPage(1);
+
+                var pdfBoxData = GetPdfBoxPositionData();
+
+                var index = 0;
+                foreach (var pageLetter in page.Letters)
+                {
+                    if (index >= pdfBoxData.Count)
+                    {
+                        break;
+                    }
+
+                    var myX = pageLetter.Location.X;
+                    var theirX = pdfBoxData[index].X;
+
+                    var myLetter = pageLetter.Value;
+                    var theirLetter = pdfBoxData[index].Text;
+
+                    if (myLetter == " " && theirLetter != " ")
+                    {
+                        continue;
+                    }
+
+                    Assert.Equal(theirLetter, myLetter);
+                    Assert.Equal(theirX, myX, 2);
+
+                    index++;
+                }
+            }
+        }
+
+        private static IReadOnlyList<AssertablePositionData> GetPdfBoxPositionData()
+        {
+            const string data = @"90	90.65997	14.42556	H	19	FFJICI+TimesNewRomanPSMT
+104.4395	90.65997	8.871117	e	19	FFJICI+TimesNewRomanPSMT
+113.3247	90.65997	5.554443	l	19	FFJICI+TimesNewRomanPSMT
+118.8931	90.65997	5.554443	l	19	FFJICI+TimesNewRomanPSMT
+124.4615	90.65997	9.989998	o	19	FFJICI+TimesNewRomanPSMT
+139.4505	90.65997	6.733261	ﺪ	19	FFJIAH+TimesNewRomanPSMT
+146.1778	90.65997	7.872116	ﻤ	19	FFJIAH+TimesNewRomanPSMT
+154.0439	90.65997	10.5894	ﺤ	19	FFJIAH+TimesNewRomanPSMT
+164.6273	90.65997	7.872116	ﻣ	19	FFJIAH+TimesNewRomanPSMT
+177.4964	90.65997	18.86111	W	19	FFJICI+TimesNewRomanPSMT
+196.3575	90.65997	9.990005	o	19	FFJICI+TimesNewRomanPSMT
+206.4275	90.65997	6.653336	r	19	FFJICI+TimesNewRomanPSMT
+213.0808	90.65997	5.554443	l	19	FFJICI+TimesNewRomanPSMT
+218.6352	90.65997	9.990005	d	19	FFJICI+TimesNewRomanPSMT
+228.6252	90.65997	4.994995	.	19	FFJICI+TimesNewRomanPSMT";
+
+            var result = data.Split(new[] {"\r", "\n", "\r\n"}, StringSplitOptions.RemoveEmptyEntries)
+                .Select(AssertablePositionData.Parse)
+                .ToList();
+
+            return result;
+        }
     }
 }

src/UglyToad.Pdf.Tests/Integration/hex mappings tounicode.txt

@@ -1,15 +0,0 @@
-
-128	64	32	16		8	4	2	1
-0	0	0	0	|	0	0	0	0
-03
-0	0	0	0	|	0	0	1	1
-20
-0	0	1	0	|	0	0	0	0	
-
-37
-0	0	1	1	|	0	1	1	1
-54
-0	1	0	1	|	0	1	0	0
-
-41
-0	1	0	0	|	0	0	0	1

src/UglyToad.Pdf/Content/PageFactory.cs

@@ -8,6 +8,7 @@
     using Graphics;
     using IO;
     using Parser;
+    using Util;
 
     internal class PageFactory : IPageFactory
     {
@@ -61,6 +62,8 @@
 
                 var contents = contentStream.Decode(filterProvider);
 
+                var texty = OtherEncodings.BytesAsLatin1String(contents);
+
                 var operations = pageContentParser.Parse(new ByteArrayInputBytes(contents));
 
                 var context = new ContentStreamProcessor(cropBox.Bounds, resourceStore, userSpaceUnit);

src/UglyToad.Pdf/Core/TransformationMatrix.cs

@@ -134,6 +134,35 @@
             return new TransformationMatrix(result);
         }
 
+        public decimal GetScalingFactorX()
+        {
+            var xScale = A;
+
+            /**
+             * BM: if the trm is rotated, the calculation is a little more complicated
+             *
+             * The rotation matrix multiplied with the scaling matrix is:
+             * (   x   0   0)    ( cos  sin  0)    ( x*cos x*sin   0)
+             * (   0   y   0) *  (-sin  cos  0)  = (-y*sin y*cos   0)
+             * (   0   0   1)    (   0    0  1)    (     0     0   1)
+             *
+             * So, if you want to deduce x from the matrix you take
+             * M(0,0) = x*cos and M(0,1) = x*sin and use the theorem of Pythagoras
+             *
+             * sqrt(M(0,0)^2+M(0,1)^2) =
+             * sqrt(x2*cos2+x2*sin2) =
+             * sqrt(x2*(cos2+sin2)) = <- here is the trick cos2+sin2 is one
+             * sqrt(x2) =
+             * abs(x)
+             */
+            if (!(B == 0m && C == 0m))
+            {
+                xScale = (decimal)Math.Sqrt((double)(A*A + B*B));
+            }
+
+            return xScale;
+        }
+
         public override bool Equals(object obj)
         {
             if (!(obj is TransformationMatrix m))

src/UglyToad.Pdf/Fonts/CidFonts/ICidFont.cs

@@ -1,5 +1,6 @@
 namespace UglyToad.Pdf.Fonts.CidFonts
 {
+    using Core;
     using Cos;
 
     /// <summary>
@@ -32,6 +33,8 @@
         /// </summary>
         CharacterIdentifierSystemInfo SystemInfo { get; }
 
+        TransformationMatrix FontMatrix { get; }
+
         CidFontType CidFontType { get; }
 
         FontDescriptor Descriptor { get; }

src/UglyToad.Pdf/Fonts/CidFonts/Type0CidFont.cs

@@ -1,5 +1,6 @@
 namespace UglyToad.Pdf.Fonts.CidFonts
 {
+    using Core;
     using Cos;
 
     /// <inheritdoc/>
@@ -13,9 +14,15 @@
         public CosName SubType { get; }
         public CosName BaseFont { get; }
         public CharacterIdentifierSystemInfo SystemInfo { get; }
+        public TransformationMatrix FontMatrix { get; }
         public CidFontType CidFontType => CidFontType.Type0;
         public FontDescriptor Descriptor { get; }
 
+        public Type0CidFont()
+        {
+            throw new System.NotImplementedException();
+        }
+
         public decimal GetWidthFromFont(int characterCode)
         {
             throw new System.NotImplementedException();

src/UglyToad.Pdf/Fonts/CidFonts/Type2CidFont.cs

@@ -1,6 +1,7 @@
 namespace UglyToad.Pdf.Fonts.CidFonts
 {
     using System.Collections.Generic;
+    using Core;
     using Cos;
 
     /// <inheritdoc />
@@ -18,6 +19,7 @@
         public CosName SubType { get; }
         public CosName BaseFont { get; }
         public CharacterIdentifierSystemInfo SystemInfo { get; }
+        public TransformationMatrix FontMatrix { get; }
         public CidFontType CidFontType => CidFontType.Type2;
         public FontDescriptor Descriptor { get; }
 
@@ -34,10 +36,15 @@
             this.fontProgram = fontProgram;
             this.verticalWritingMetrics = verticalWritingMetrics;
             this.widths = widths;
+
+            // TODO: This should maybe take units per em into account?
+            var scale = 1 / 1000m;
+            FontMatrix = TransformationMatrix.FromValues(scale, 0, 0, scale, 0, 0);
         }
 
         public decimal GetWidthFromFont(int characterCode)
         {
+            // TODO: Read the font width from the font program.
             throw new System.NotImplementedException();
         }
 

src/UglyToad.Pdf/Fonts/Composite/Type0Font.cs

@@ -3,6 +3,7 @@
     using System;
     using CidFonts;
     using Cmap;
+    using Core;
     using Cos;
     using Geometry;
     using IO;
@@ -84,5 +85,10 @@
             
             return fromFont;
         }
+
+        public TransformationMatrix GetFontMatrix()
+        {
+            return CidFont.FontMatrix;
+        }
     }
 }

src/UglyToad.Pdf/Fonts/IFont.cs

@@ -1,5 +1,6 @@
 namespace UglyToad.Pdf.Fonts
 {
+    using Core;
     using Cos;
     using Geometry;
     using IO;
@@ -17,5 +18,7 @@
         PdfVector GetDisplacement(int characterCode);
 
         decimal GetWidth(int characterCode);
+
+        TransformationMatrix GetFontMatrix();
     }
 }

src/UglyToad.Pdf/Fonts/Simple/TrueTypeSimpleFont.cs

@@ -3,6 +3,7 @@
     using System;
     using Cmap;
     using Composite;
+    using Core;
     using Cos;
     using Encodings;
     using Geometry;
@@ -11,6 +12,8 @@
 
     internal class TrueTypeSimpleFont : IFont
     {
+        private static readonly TransformationMatrix FontMatrix =
+            TransformationMatrix.FromValues(1/1000m, 0, 0, 1/1000m, 0, 0);
         private readonly int firstCharacterCode;
         private readonly int lastCharacterCode;
         private readonly decimal[] widths;
@@ -108,5 +111,11 @@
 
             return widths[index];
         }
+
+        public TransformationMatrix GetFontMatrix()
+        {
+            // TODO: should this also use units per em?
+            return FontMatrix;
+        }
     }
 }

src/UglyToad.Pdf/Graphics/ContentStreamProcessor.cs

@@ -89,11 +89,13 @@
             }
             
             var fontSize = currentState.FontState.FontSize;
-            var horizontalScaling = currentState.FontState.HorizontalScaling;
+            var horizontalScaling = currentState.FontState.HorizontalScaling / 100m;
             var characterSpacing = currentState.FontState.CharacterSpacing;
 
             var transformationMatrix = currentState.CurrentTransformationMatrix;
 
+            var fontMatrix = font.GetFontMatrix();
+
             // TODO: this does not seem correct, produces the correct result for now but we need to revisit.
             // see: https://stackoverflow.com/questions/48010235/pdf-specification-get-font-size-in-points
             var pointSize = decimal.Round(fontSize * transformationMatrix.A, 2);
@@ -114,12 +116,12 @@
 
                 if (font.IsVertical)
                 {
-                    throw new NotImplementedException("Vertical fonts are currently unsupported, please submit a pull request or issue with an example file.");
+                    throw new NotImplementedException("Vertical fonts are# currently unsupported, please submit a pull request or issue with an example file.");
                 }
 
                 var displacement = font.GetDisplacement(code);
-
+                
-                var width = (displacement.X * fontSize) * transformationMatrix.A;
+                var width = displacement.X * fontSize * TextMatrices.TextMatrix.GetScalingFactorX() * transformationMatrix.A;
 
                 ShowGlyph(renderingMatrix, font, code, unicode, width, fontSize, pointSize);
 
@@ -148,7 +150,7 @@
             var textState = currentState.FontState;
 
             var fontSize = textState.FontSize;
-            var horizontalScaling = textState.HorizontalScaling;
+            var horizontalScaling = textState.HorizontalScaling/100m;
             var font = resourceStore.GetFont(textState.FontName);
 
             var isVertical = font.IsVertical;