return both glyph and character bounding boxes for each letter, glyph is the actual outline of the letter whereas character is the surrounding space as defined in the pdf

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

@@ -70,7 +70,7 @@
                         break;
                     }
 
-                    var myX = pageLetter.Rectangle.BottomLeft.X;
+                    var myX = pageLetter.CharacterRectangle.Left;
                     var theirX = pdfBoxData[index].X;
 
                     var myLetter = pageLetter.Value;
@@ -87,7 +87,7 @@
 
                     Assert.Equal(theirX, myX, comparer);
 
-                    Assert.Equal(pdfBoxData[index].Width, pageLetter.Rectangle.Width, comparer);
+                    Assert.Equal(pdfBoxData[index].Width, pageLetter.CharacterRectangle.Width, comparer);
 
                     index++;
                 }
@@ -97,6 +97,8 @@
         [Fact]
         public void LetterPositionsAreCorrectXfinium()
         {
+            var comparer = new DecimalComparer(1);
+
             using (var document = PdfDocument.Open(GetFilename()))
             {
                 var page = document.GetPage(1);
@@ -111,7 +113,7 @@
                         break;
                     }
 
-                    var myX = pageLetter.Rectangle.Left;
+                    var myX = pageLetter.CharacterRectangle.Left;
                     var theirX = positions[index].X;
 
                     var myLetter = pageLetter.Value;
@@ -123,9 +125,9 @@
                     }
 
                     Assert.Equal(theirLetter, myLetter);
-                    Assert.Equal(theirX, myX, 2);
+                    Assert.Equal(theirX, myX, comparer);
 
-                    Assert.Equal(positions[index].Width, pageLetter.Rectangle.Width, 1);
+                    Assert.Equal(positions[index].Width, pageLetter.CharacterRectangle.Width, 1);
 
                     index++;
                 }

src/UglyToad.PdfPig.Tests/Integration/SinglePageSimpleGoogleChromeTests.cs

@@ -134,12 +134,12 @@ namespace UglyToad.PdfPig.Tests.Integration
                     }
 
                     Assert.Equal(datum.Text, letter.Value);
-                    Assert.Equal(datum.X, letter.Rectangle.BottomLeft.X, 2);
+                    Assert.Equal(datum.X, letter.CharacterRectangle.BottomLeft.X, 2);
 
-                    var transformed = page.Height - letter.Rectangle.BottomLeft.Y;
+                    var transformed = page.Height - letter.CharacterRectangle.BottomLeft.Y;
                     Assert.Equal(datum.Y, transformed, 2);
 
-                    Assert.Equal(datum.Width, letter.Rectangle.Width, 2);
+                    Assert.Equal(datum.Width, letter.CharacterRectangle.Width, 2);
 
                     Assert.Equal(datum.FontName, letter.FontName);
 
@@ -179,13 +179,13 @@ namespace UglyToad.PdfPig.Tests.Integration
                     }
 
                     Assert.Equal(datum.Text, letter.Value);
-                    Assert.Equal(datum.X, letter.Rectangle.BottomLeft.X, 2);
+                    Assert.Equal(datum.X, letter.CharacterRectangle.BottomLeft.X, 2);
 
-                    var transformed = page.Height - letter.Rectangle.BottomLeft.Y;
+                    var transformed = page.Height - letter.CharacterRectangle.BottomLeft.Y;
                     Assert.Equal(datum.Y, transformed, 2);
 
                     // Until we get width from glyphs we're a bit out.
-                    Assert.True(Math.Abs(datum.Width - letter.Rectangle.Width) < 0.03m);
+                    Assert.True(Math.Abs(datum.Width - letter.CharacterRectangle.Width) < 0.03m);
 
                     index++;
                 }

src/UglyToad.PdfPig.Tests/Integration/SinglePageSimpleOpenOfficeTests.cs

@@ -48,27 +48,27 @@
 
                 Assert.Equal("I", page.Letters[0].Value);
 
-                Assert.Equal(90.1m, page.Letters[0].Rectangle.BottomLeft.X, comparer);
-                Assert.Equal(709.2m, page.Letters[0].Rectangle.BottomLeft.Y, comparer);
+                Assert.Equal(90.1m, page.Letters[0].GlyphRectangle.BottomLeft.X, comparer);
+                Assert.Equal(709.2m, page.Letters[0].GlyphRectangle.BottomLeft.Y, comparer);
 
-                Assert.Equal(94.0m, page.Letters[0].Rectangle.TopRight.X, comparer);
-                Assert.Equal(719.89m, page.Letters[0].Rectangle.TopRight.Y, comparer);
+                Assert.Equal(94.0m, page.Letters[0].GlyphRectangle.TopRight.X, comparer);
+                Assert.Equal(719.89m, page.Letters[0].GlyphRectangle.TopRight.Y, comparer);
 
                 Assert.Equal("a", page.Letters[5].Value);
 
-                Assert.Equal(114.5m, page.Letters[5].Rectangle.BottomLeft.X, comparer);
-                Assert.Equal(709.2m, page.Letters[5].Rectangle.BottomLeft.Y, comparer);
+                Assert.Equal(114.5m, page.Letters[5].GlyphRectangle.BottomLeft.X, comparer);
+                Assert.Equal(709.2m, page.Letters[5].GlyphRectangle.BottomLeft.Y, comparer);
 
-                Assert.Equal(119.82m, page.Letters[5].Rectangle.TopRight.X, comparer);
-                Assert.Equal(714.89m, page.Letters[5].Rectangle.TopRight.Y, comparer);
+                Assert.Equal(119.82m, page.Letters[5].GlyphRectangle.TopRight.X, comparer);
+                Assert.Equal(714.89m, page.Letters[5].GlyphRectangle.TopRight.Y, comparer);
 
                 Assert.Equal("f", page.Letters[16].Value);
 
-                Assert.Equal(169.9m, page.Letters[16].Rectangle.BottomLeft.X, comparer);
-                Assert.Equal(709.2m, page.Letters[16].Rectangle.BottomLeft.Y, comparer);
+                Assert.Equal(169.9m, page.Letters[16].GlyphRectangle.BottomLeft.X, comparer);
+                Assert.Equal(709.2m, page.Letters[16].GlyphRectangle.BottomLeft.Y, comparer);
 
-                Assert.Equal(176.89m, page.Letters[16].Rectangle.TopRight.X, comparer);
-                Assert.Equal(719.89m, page.Letters[16].Rectangle.TopRight.Y, comparer);
+                Assert.Equal(176.89m, page.Letters[16].GlyphRectangle.TopRight.X, comparer);
+                Assert.Equal(719.89m, page.Letters[16].GlyphRectangle.TopRight.Y, comparer);
             }
         }
 

src/UglyToad.PdfPig.Tests/Integration/Type0FontTests.cs

@@ -57,8 +57,8 @@ namespace UglyToad.PdfPig.Tests.Integration
             {
                 var page = document.GetPage(1);
 
-                Assert.True((bool) page.Letters.Any(x => x.Rectangle.Width != 0));
-                Assert.True((bool) page.Letters.Any(x => x.Rectangle.Height != 0));
+                Assert.True((bool) page.Letters.Any(x => x.GlyphRectangle.Width != 0));
+                Assert.True((bool) page.Letters.Any(x => x.GlyphRectangle.Height != 0));
             }
         }
     }

src/UglyToad.PdfPig.Tests/Integration/VisualVerification/GenerateLetterBoundingBoxImages.cs

@@ -57,8 +57,8 @@
                 {
                     foreach (var word in page.Letters)
                     {
-                        graphics.DrawRectangle(redPen, new Rectangle((int)word.Rectangle.Left,
-                            792 - (int)word.Rectangle.Top, (int)Math.Max(1, word.Rectangle.Width), (int)word.Rectangle.Height));
+                        graphics.DrawRectangle(redPen, new Rectangle((int)word.GlyphRectangle.Left,
+                            792 - (int)word.GlyphRectangle.Top, (int)Math.Max(1, word.GlyphRectangle.Width), (int)word.GlyphRectangle.Height));
                     }
 
                     var imageName = $"{file}.jpg";

src/UglyToad.PdfPig/Content/Letter.cs

@@ -13,9 +13,14 @@
         public string Value { get; }
 
         /// <summary>
-        /// Position of the bounding box.
+        /// Position of the bounding box for the glyph.
         /// </summary>
-        public PdfRectangle Rectangle { get; }
+        public PdfRectangle GlyphRectangle { get; }
+
+        /// <summary>
+        /// The bounding box for the entire character.
+        /// </summary>
+        public PdfRectangle CharacterRectangle { get; }
 
         /// <summary>
         /// Size as defined in the PDF file. This is not equivalent to font size in points but is relative to other font sizes on the page.
@@ -35,13 +40,14 @@
         /// <summary>
         /// Create a new letter to represent some text drawn by the Tj operator.
         /// </summary>
-        internal Letter(string value, PdfRectangle rectangle, decimal fontSize, string fontName, decimal pointSize)
+        internal Letter(string value, PdfRectangle glyphRectangle, PdfRectangle characterRectangle, decimal fontSize, string fontName, decimal pointSize)
         {
             Value = value;
-            Rectangle = rectangle;
+            GlyphRectangle = glyphRectangle;
             FontSize = fontSize;
             FontName = fontName;
             PointSize = pointSize;
+            CharacterRectangle = characterRectangle;
         }
 
         /// <summary>
@@ -49,7 +55,7 @@
         /// </summary>
         public override string ToString()
         {
-            return $"{Value} {Rectangle} {FontName} {PointSize}";
+            return $"{Value} {GlyphRectangle} {FontName} {PointSize}";
         }
     }
 }

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

@@ -42,6 +42,8 @@
 
         decimal GetWidthFromDictionary(int cid);
 
+        decimal GetWidthFromFont(int characterIdentifier);
+
         PdfRectangle GetBoundingBox(int characterIdentifier);
     }
 }

src/UglyToad.PdfPig/Fonts/CidFonts/ICidFontProgram.cs

@@ -12,7 +12,9 @@
 
         bool TryGetBoundingBox(int characterIdentifier, Func<int, int> characterIdentifierToGlyphIndex, out PdfRectangle boundingBox);
 
-        bool TryGetBoundingAdvancedWidth(int characterCode, out decimal width);
+        bool TryGetBoundingAdvancedWidth(int characterIdentifier, Func<int, int> characterIdentifierToGlyphIndex, out decimal width);
+
+        bool TryGetBoundingAdvancedWidth(int characterIdentifier, out decimal width);
 
         int GetFontMatrixMultiplier();
     }

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

@@ -46,10 +46,14 @@
             FontMatrix = TransformationMatrix.FromValues(scale, 0, 0, scale, 0, 0);
         }
 
-        public decimal GetWidthFromFont(int characterCode)
+        public decimal GetWidthFromFont(int characterIdentifier)
         {
+            if (fontProgram.TryGetBoundingAdvancedWidth(characterIdentifier, cidToGid.GetGlyphIndex, out var width))
+            {
+                return width;
+            }
             // TODO: Read the font width from the font program.
-            throw new System.NotImplementedException();
+            return GetWidthFromDictionary(characterIdentifier);
         }
 
         public decimal GetWidthFromDictionary(int characterIdentifier)

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

@@ -69,22 +69,22 @@
             return ToUnicode.TryGet(characterCode, out value);
         }
 
-        public PdfRectangle GetBoundingBox(int characterCode)
+        public CharacterBoundingBox GetBoundingBox(int characterCode)
         {
             var matrix = GetFontMatrix();
 
             var boundingBox = GetBoundingBoxInGlyphSpace(characterCode);
 
-            return matrix.Transform(boundingBox);
-        }
+            boundingBox = matrix.Transform(boundingBox);
 
-        public decimal GetWidth(int characterCode)
-        {
-            var cid = CMap.ConvertToCid(characterCode);
+            var characterIdentifier = CMap.ConvertToCid(characterCode);
 
-            var fromFont = CidFont.GetWidthFromDictionary(cid);
+            var width = CidFont.GetWidthFromFont(characterIdentifier);
 
-            return fromFont;
+            var advanceWidth = new PdfRectangle(0, 0, width, 0);
+            advanceWidth = matrix.Transform(advanceWidth);
+
+            return new CharacterBoundingBox(boundingBox, advanceWidth);
         }
 
         public PdfRectangle GetBoundingBoxInGlyphSpace(int characterCode)

src/UglyToad.PdfPig/Fonts/IFont.cs

@@ -15,8 +15,21 @@
 
         bool TryGetUnicode(int characterCode, out string value);
 
-        PdfRectangle GetBoundingBox(int characterCode);
+        CharacterBoundingBox GetBoundingBox(int characterCode);
 
         TransformationMatrix GetFontMatrix();
     }
+
+    internal class CharacterBoundingBox
+    {
+        public PdfRectangle GlyphBounds { get; }
+
+        public PdfRectangle CharacterBounds { get; }
+
+        public CharacterBoundingBox(PdfRectangle glyphBounds, PdfRectangle characterBounds)
+        {
+            GlyphBounds = glyphBounds;
+            CharacterBounds = characterBounds;
+        }
+    }
 }

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

@@ -1,6 +1,4 @@
-using System;
-
-namespace UglyToad.PdfPig.Fonts.Simple
+namespace UglyToad.PdfPig.Fonts.Simple
 {
     using Cmap;
     using Composite;
@@ -88,15 +86,32 @@ namespace UglyToad.PdfPig.Fonts.Simple
             return true;
         }
 
-        public PdfRectangle GetBoundingBox(int characterCode)
+        public CharacterBoundingBox GetBoundingBox(int characterCode)
         {
             var fontMatrix = GetFontMatrix();
 
             var boundingBox = GetBoundingBoxInGlyphSpace(characterCode);
 
-            var result = fontMatrix.Transform(boundingBox);
+            boundingBox = fontMatrix.Transform(boundingBox);
 
-            return result;
+            decimal width;
+            if (font == null)
+            {
+                 width = widths[characterCode];
+            }
+            else
+            {
+                if (!font.TryGetBoundingAdvancedWidth(characterCode, out width))
+                {
+                    width = boundingBox.Width;
+                }
+            }
+            
+            var advancedRectangle = new PdfRectangle(0, 0, width, 0);
+
+            advancedRectangle = fontMatrix.Transform(advancedRectangle);
+
+            return new CharacterBoundingBox(boundingBox, advancedRectangle);
         }
 
         private PdfRectangle GetBoundingBoxInGlyphSpace(int characterCode)

src/UglyToad.PdfPig/Fonts/Simple/Type1FontSimple.cs

@@ -86,9 +86,13 @@
             return true;
         }
 
-        public PdfRectangle GetBoundingBox(int characterCode)
+        public CharacterBoundingBox GetBoundingBox(int characterCode)
         {
-            return fontMatrix.Transform(GetBoundingBoxInGlyphSpace(characterCode));
+            var boundingBox = GetBoundingBoxInGlyphSpace(characterCode);
+
+            boundingBox = fontMatrix.Transform(boundingBox);
+
+            return new CharacterBoundingBox(boundingBox, boundingBox);
         }
 
         private PdfRectangle GetBoundingBoxInGlyphSpace(int characterCode)

src/UglyToad.PdfPig/Fonts/Simple/Type1Standard14Font.cs

@@ -44,9 +44,13 @@
             return true;
         }
 
-        public PdfRectangle GetBoundingBox(int characterCode)
+        public CharacterBoundingBox GetBoundingBox(int characterCode)
         {
-            return fontMatrix.Transform(GetBoundingBoxInGlyphSpace(characterCode));
+            var boundingBox = GetBoundingBoxInGlyphSpace(characterCode);
+
+            boundingBox = fontMatrix.Transform(boundingBox);
+
+            return new CharacterBoundingBox(boundingBox, boundingBox);
         }
 
         private PdfRectangle GetBoundingBoxInGlyphSpace(int characterCode)

src/UglyToad.PdfPig/Fonts/Simple/Type3Font.cs

@@ -63,9 +63,13 @@
             return true;
         }
 
-        public PdfRectangle GetBoundingBox(int characterCode)
+        public CharacterBoundingBox GetBoundingBox(int characterCode)
         {
-            return fontMatrix.Transform(GetBoundingBoxInGlyphSpace(characterCode));
+            var boundingBox = GetBoundingBoxInGlyphSpace(characterCode);
+
+            boundingBox = fontMatrix.Transform(boundingBox);
+
+            return new CharacterBoundingBox(boundingBox, boundingBox);
         }
 
         private PdfRectangle GetBoundingBoxInGlyphSpace(int characterCode)

src/UglyToad.PdfPig/Fonts/TrueType/TrueTypeFont.cs

@@ -57,11 +57,12 @@
             return true;
         }
 
-        public bool TryGetBoundingAdvancedWidth(int characterCode, out decimal width)
+        public bool TryGetBoundingAdvancedWidth(int characterIdentifier, out decimal width) => TryGetBoundingAdvancedWidth(characterIdentifier, null, out width);
+        public bool TryGetBoundingAdvancedWidth(int characterIdentifier, Func<int, int> characterIdentifierToGlyphIndex, out decimal width)
         {
             width = 0m;
 
-            if (!TryGetGlyphIndex(characterCode, null, out var index))
+            if (!TryGetGlyphIndex(characterIdentifier, characterIdentifierToGlyphIndex, out var index))
             {
                 return false;
             }

src/UglyToad.PdfPig/Graphics/ContentStreamProcessor.cs

@@ -128,22 +128,25 @@
 
                 var boundingBox = font.GetBoundingBox(code);
 
-                var transformedDisplacement = transformationMatrix
+                var transformedGlyphBounds = transformationMatrix
                     .Transform(TextMatrices.TextMatrix
                     .Transform(renderingMatrix
-                    .Transform(boundingBox)));
+                    .Transform(boundingBox.GlyphBounds)));
+                var transformedGlyphOrigin = transformationMatrix
+                    .Transform(TextMatrices.TextMatrix
+                        .Transform(renderingMatrix.Transform(boundingBox.CharacterBounds)));
 
-                ShowGlyph(font, transformedDisplacement, unicode, fontSize, pointSize);
+                ShowGlyph(font, transformedGlyphBounds, transformedGlyphOrigin, unicode, fontSize, pointSize);
 
                 decimal tx, ty;
                 if (font.IsVertical)
                 {
                     tx = 0;
-                    ty = boundingBox.Height * fontSize + characterSpacing + wordSpacing;
+                    ty = boundingBox.CharacterBounds.Height * fontSize + characterSpacing + wordSpacing;
                 }
                 else
                 {
-                    tx = (boundingBox.Width * fontSize + characterSpacing + wordSpacing) * horizontalScaling;
+                    tx = (boundingBox.CharacterBounds.Width * fontSize + characterSpacing + wordSpacing) * horizontalScaling;
                     ty = 0;
                 }
 
@@ -211,9 +214,9 @@
             TextMatrices.TextMatrix = newMatrix;
         }
 
-        private void ShowGlyph(IFont font, PdfRectangle rectangle, string unicode, decimal fontSize, decimal pointSize)
+        private void ShowGlyph(IFont font, PdfRectangle glyphRectangle, PdfRectangle characterRectangle,  string unicode, decimal fontSize, decimal pointSize)
         {
-            var letter = new Letter(unicode, rectangle, fontSize, font.Name.Data, pointSize);
+            var letter = new Letter(unicode, glyphRectangle, characterRectangle, fontSize, font.Name.Data, pointSize);
 
             Letters.Add(letter);
         }