Merge pull request #61 from vadik299/master

Adding TextSequence number to each letter to determine if letters belong to the same Tj operation

src/UglyToad.PdfPig/Content/Letter.cs

@@ -65,6 +65,11 @@
         /// </summary>
         internal decimal PointSize { get; }
 
+        /// <summary>
+        /// Sequence number of the ShowText operation that printed this letter.
+        /// </summary>
+        public int TextSequence { get; }
+
         /// <summary>
         /// Create a new letter to represent some text drawn by the Tj operator.
         /// </summary>
@@ -75,7 +80,8 @@
             decimal fontSize,
             string fontName, 
             IColor color,
-            decimal pointSize)
+            decimal pointSize,
+            int textSequence)
         {
             Value = value;
             GlyphRectangle = glyphRectangle;
@@ -86,6 +92,7 @@
             FontName = fontName;
             Color = color ?? GrayColor.Black;
             PointSize = pointSize;
+            TextSequence = textSequence;
             TextDirection = GetTextDirection();
         }
 

src/UglyToad.PdfPig/Graphics/ContentStreamProcessor.cs

@@ -31,6 +31,9 @@
         private Stack<CurrentGraphicsState> graphicsStack = new Stack<CurrentGraphicsState>();
         private IFont activeExtendedGraphicsStateFont = null;
 
+        //a sequence number of ShowText operation to determine whether letters belong to same operation or not (letters that belong to different operations have less changes to belong to same word)
+        private int textSequence = 0;
+
         public TextMatrices TextMatrices { get; } = new TextMatrices();
 
         public TransformationMatrix CurrentTransformationMatrix
@@ -187,7 +190,8 @@
                     unicode, 
                     fontSize,
                     color,
-                    pointSize);
+                    pointSize,
+                    textSequence);
 
                 decimal tx, ty;
                 if (font.IsVertical)
@@ -209,6 +213,8 @@
 
         public void ShowPositionedText(IReadOnlyList<IToken> tokens)
         {
+            textSequence++;
+
             var currentState = GetCurrentState();
 
             var textState = currentState.FontState;
@@ -361,7 +367,8 @@
             string unicode,
             decimal fontSize,
             IColor color,
-            decimal pointSize)
+            decimal pointSize,
+            int textSequence)
         {
             var letter = new Letter(unicode, glyphRectangle, 
                 startBaseLine, 
@@ -370,7 +377,8 @@
                 fontSize, 
                 font.Name.Data,
                 color,
-                pointSize);
+                pointSize,
+                textSequence);
 
             Letters.Add(letter);
         }

src/UglyToad.PdfPig/Writer/PdfPageBuilder.cs

@@ -22,6 +22,10 @@
     {
         private readonly PdfDocumentBuilder documentBuilder;
         private readonly List<IGraphicsStateOperation> operations = new List<IGraphicsStateOperation>();
+
+        //a sequence number of ShowText operation to determine whether letters belong to same operation or not (letters that belong to different operations have less changes to belong to same word)
+        private static int textSequence = 0;
+
         internal IReadOnlyList<IGraphicsStateOperation> Operations => operations;
 
         /// <summary>
@@ -240,6 +244,8 @@
 
             var width = 0m;
 
+            textSequence++;
+
             for (var i = 0; i < text.Length; i++)
             {
                 var c = text[i];
@@ -261,7 +267,9 @@
 
                 var letter = new Letter(c.ToString(), documentSpace, advanceRect.BottomLeft, advanceRect.BottomRight, width, fontSize, font.Name,
                     GrayColor.Black, 
-                    fontSize);
+                    fontSize,
+                    textSequence);
+
                 letters.Add(letter);
 
                 var tx = advanceRect.Width * horizontalScaling;