Fix the handling of CRCRLF - DEVSIX-1727

Author: michaeldemey

io/src/main/java/com/itextpdf/io/util/TextUtil.java

@@ -44,6 +44,7 @@ This file is part of the iText (R) project.
 package com.itextpdf.io.util;
 
 import com.itextpdf.io.font.otf.Glyph;
+import com.itextpdf.io.font.otf.GlyphLine;
 
 import java.util.ArrayList;
 import java.util.List;
@@ -215,6 +216,13 @@ public static boolean isNewLine(Glyph glyph) {
         return unicode == '\n' || unicode == '\r';
     }
 
+    public static boolean isCarriageReturnFollowedByLineFeed(GlyphLine glyphLine, int carriageReturnPosition) {
+        return glyphLine.size() > 1
+                && carriageReturnPosition <= glyphLine.size() - 2
+                && glyphLine.get(carriageReturnPosition).getUnicode() == '\r'
+                && glyphLine.get(carriageReturnPosition + 1).getUnicode() == '\n';
+    }
+
     public static boolean isSpaceOrWhitespace(Glyph glyph) {
         //\r, \n, and \t are whitespaces, but not space chars.
         //\u00a0 is SpaceChar, but not whitespace.

io/src/test/java/com/itextpdf/io/util/TextUtilTest.java

@@ -0,0 +1,66 @@
+package com.itextpdf.io.util;
+
+import com.itextpdf.io.font.otf.Glyph;
+import com.itextpdf.io.font.otf.GlyphLine;
+import com.itextpdf.test.annotations.type.UnitTest;
+
+import java.util.Arrays;
+
+import org.junit.Assert;
+import org.junit.Before;
+import org.junit.Test;
+import org.junit.experimental.categories.Category;
+
+@Category(UnitTest.class)
+public class TextUtilTest {
+
+    private Glyph carriageReturn;
+    private Glyph lineFeed;
+
+    @Before
+    public void before() {
+        this.carriageReturn = new Glyph(0, 0, '\r');
+        this.lineFeed = new Glyph(0, 0, '\n');
+    }
+
+    @Test
+    public void carriageReturnFollowedByLineFeedTest() {
+        helper(true, 0,
+                carriageReturn, lineFeed);
+    }
+
+    @Test
+    public void carriageReturnFollowedByCarriageReturnAndThenLineFeedTest() {
+        helper(false, 0,
+                carriageReturn, carriageReturn, lineFeed);
+    }
+
+    @Test
+    public void carriageReturnPrecededByCarriageReturnAndFollowedByLineFeedTest() {
+        helper(true, 1,
+                carriageReturn, carriageReturn, lineFeed);
+    }
+
+    @Test
+    public void carriageReturnFollowedByNothingTest() {
+        helper(false, 0,
+                carriageReturn);
+    }
+
+    @Test
+    public void carriageReturnPrecededByLineFeedTest() {
+        helper(false, 0,
+                lineFeed, carriageReturn);
+    }
+
+    @Test
+    public void carriageReturnPrecededByTextFollowedByLineFeedTest() {
+        helper(true, 1,
+                new Glyph(0,0, 'a'), carriageReturn, lineFeed);
+    }
+
+    private void helper(boolean expected, int currentCRPosition, Glyph...glyphs) {
+        GlyphLine glyphLine = new GlyphLine(Arrays.asList(glyphs));
+        Assert.assertTrue(expected == TextUtil.isCarriageReturnFollowedByLineFeed(glyphLine, currentCRPosition));
+    }
+}

layout/src/main/java/com/itextpdf/layout/renderer/TextRenderer.java

@@ -49,6 +49,7 @@ This file is part of the iText (R) project.
 import com.itextpdf.io.font.TrueTypeFont;
 import com.itextpdf.io.font.otf.Glyph;
 import com.itextpdf.io.font.otf.GlyphLine;
+import com.itextpdf.io.util.MessageFormatUtil;
 import com.itextpdf.io.util.TextUtil;
 import com.itextpdf.kernel.colors.Color;
 import com.itextpdf.kernel.font.PdfFont;
@@ -83,11 +84,8 @@ This file is part of the iText (R) project.
 import com.itextpdf.layout.property.Underline;
 import com.itextpdf.layout.property.UnitValue;
 import com.itextpdf.layout.splitting.ISplitCharacters;
-import org.slf4j.Logger;
-import org.slf4j.LoggerFactory;
-
-import com.itextpdf.io.util.MessageFormatUtil;
 import com.itextpdf.layout.tagging.LayoutTaggingHelper;
+
 import java.util.ArrayList;
 import java.util.Collection;
 import java.util.Collections;
@@ -96,6 +94,9 @@ This file is part of the iText (R) project.
 import java.util.List;
 import java.util.Map;
 
+import org.slf4j.Logger;
+import org.slf4j.LoggerFactory;
+
 /**
  * This class represents the {@link IRenderer renderer} object for a {@link Text}
  * object. It will draw the glyphs of the textual content on the {@link DrawContext}.
@@ -235,6 +236,8 @@ public LayoutResult layout(LayoutContext layoutContext) {
         boolean forcePartialSplitOnFirstChar = false;
         // true in situations like "Hello\nWorld"
         boolean ignoreNewLineSymbol = false;
+        // true when \r\n are found
+        boolean crlf = false;
 
         // For example, if a first character is a RTL mark (U+200F), and the second is a newline, we need to break anyway
         int firstPrintPos = currentTextPos;
@@ -271,9 +274,17 @@ public LayoutResult layout(LayoutContext layoutContext) {
                         // Notice that in that case we do not need to ignore the new line symbol ('\n')
                         forcePartialSplitOnFirstChar = true;
                     }
+
                     if (line.start == -1) {
                         line.start = currentTextPos;
                     }
+
+                    crlf = TextUtil.isCarriageReturnFollowedByLineFeed(text, currentTextPos);
+
+                    if ( crlf) {
+                        currentTextPos++;
+                    }
+
                     line.end = Math.max(line.end, firstCharacterWhichExceedsAllowedWidth - 1);
                     break;
                 }
@@ -417,7 +428,9 @@ public LayoutResult layout(LayoutContext layoutContext) {
                         if (line.start == -1) {
                             line.start = currentTextPos;
                         }
-                        currentTextPos = (forcePartialSplitOnFirstChar || null == overflowX || OverflowPropertyValue.FIT.equals(overflowX)) ? firstCharacterWhichExceedsAllowedWidth : nonBreakablePartEnd+1;
+                        if ( !crlf ) {
+                            currentTextPos = (forcePartialSplitOnFirstChar || null == overflowX || OverflowPropertyValue.FIT.equals(overflowX)) ? firstCharacterWhichExceedsAllowedWidth : nonBreakablePartEnd+1;
+                        }
                         line.end = Math.max(line.end, currentTextPos);
                         wordSplit = !forcePartialSplitOnFirstChar && (text.end != currentTextPos);
                         if (wordSplit || !(forcePartialSplitOnFirstChar || null == overflowX || OverflowPropertyValue.FIT.equals(overflowX))) {
@@ -475,7 +488,7 @@ public LayoutResult layout(LayoutContext layoutContext) {
             result = new TextLayoutResult(LayoutResult.FULL, occupiedArea, null, null, isPlacingForcedWhileNothing ? this : null);
         } else {
             TextRenderer[] split;
-            if (ignoreNewLineSymbol) {
+            if (ignoreNewLineSymbol || crlf ) {
                 // ignore '\n'
                 split = splitIgnoreFirstNewLine(currentTextPos);
             } else {
@@ -984,13 +997,8 @@ static float[] calculateAscenderDescender(PdfFont font) {
     }
 
     private TextRenderer[] splitIgnoreFirstNewLine(int currentTextPos) {
-        if (text.get(currentTextPos).getUnicode() == '\r') {
-            int next = currentTextPos + 1 < text.end ? text.get(currentTextPos + 1).getUnicode() : -1;
-            if (next == '\n') {
-                return split(currentTextPos + 2);
-            } else {
-                return split(currentTextPos + 1);
-            }
+        if ( TextUtil.isCarriageReturnFollowedByLineFeed(text, currentTextPos)) {
+            return split(currentTextPos + 2);
         } else {
             return split(currentTextPos + 1);
         }

layout/src/test/java/com/itextpdf/layout/NewLineTest.java

@@ -0,0 +1,81 @@
+package com.itextpdf.layout;
+
+import com.itextpdf.kernel.pdf.PdfDocument;
+import com.itextpdf.kernel.pdf.PdfWriter;
+import com.itextpdf.kernel.pdf.WriterProperties;
+import com.itextpdf.kernel.utils.CompareTool;
+import com.itextpdf.layout.element.Paragraph;
+import com.itextpdf.test.ExtendedITextTest;
+import com.itextpdf.test.annotations.type.IntegrationTest;
+
+import java.io.FileOutputStream;
+import java.io.IOException;
+
+import org.junit.Assert;
+import org.junit.BeforeClass;
+import org.junit.Test;
+import org.junit.experimental.categories.Category;
+
+@Category(IntegrationTest.class)
+public class NewLineTest extends ExtendedITextTest {
+
+    public static final String sourceFolder = "./src/test/resources/com/itextpdf/layout/NewLineTest/";
+    public static final String destinationFolder = "./target/test/com/itextpdf/layout/NewLineTest/";
+
+    @BeforeClass
+    public static void beforeClass() {
+        createOrClearDestinationFolder(destinationFolder);
+    }
+
+    @Test
+    public void r() throws IOException, InterruptedException {
+        test("\r", "r.pdf");
+    }
+
+    @Test
+    public void n() throws IOException, InterruptedException {
+        test("\n", "n.pdf");
+    }
+
+    @Test
+    public void rn() throws IOException, InterruptedException {
+        test("\r\n", "rn.pdf");
+    }
+
+    @Test
+    public void rrn() throws IOException, InterruptedException {
+        test("\r\r\n", "rrn.pdf");
+    }
+
+    @Test
+    public void nn() throws IOException, InterruptedException {
+        test("\n\n", "nn.pdf");
+    }
+
+    @Test
+    public void rnn() throws IOException, InterruptedException {
+        test("\r\n\n", "rnn.pdf");
+    }
+
+    @Test
+    public void rnrn() throws IOException, InterruptedException {
+        test("\r\n\r\n", "rnrn.pdf");
+    }
+
+    private void test(String newlineCharacters, String fileName) throws IOException, InterruptedException {
+        String outFileName = destinationFolder + fileName;
+        String cmpFileName = sourceFolder + "cmp_" + fileName;
+        String diffPrefix = "diff_" + fileName + "_";
+
+        PdfDocument pdf = new PdfDocument(new PdfWriter(new FileOutputStream(outFileName), new WriterProperties().setCompressionLevel(0)));
+        Document document = new Document(pdf);
+
+        Paragraph paragraph = new Paragraph().add(
+                "This line is before." + newlineCharacters + "This line is after.");
+                
+        document.add(paragraph);
+        document.close();
+
+        Assert.assertNull(new CompareTool().compareVisually(outFileName, cmpFileName, destinationFolder, diffPrefix));
+    }
+}