For #25 - Add method to builder to specify a custom line breaker. [ci…

… skip]

See also implementation of line breaker using ICU4J in rtl-support
module.

README.md

@@ -50,6 +50,7 @@ CHANGELOG
 
 head - 0.0.1-RC5-SNAPSHOT
 ========
++ [Add method to builder to specify a custom line breaker](https://github.com/danfickle/openhtmltopdf/issues/25) Thanks @Magotchi
 
 0.0.1-RC4
 ========

openhtmltopdf-core/src/main/java/com/openhtmltopdf/extend/FSTextBreaker.java

@@ -0,0 +1,11 @@
+package com.openhtmltopdf.extend;
+
+/**
+ * Represents a text breaker, such as those on line break opportunities.
+ * Implementations usually wrap a BreakIterator of some kind.
+ * Will be reused many times during a run.
+ */
+public interface FSTextBreaker {
+	public int next();
+	public void setText(String newText);
+}

openhtmltopdf-core/src/main/java/com/openhtmltopdf/layout/Breaker.java

@@ -20,10 +20,9 @@
  */
 package com.openhtmltopdf.layout;
 
-import java.text.BreakIterator;
-
 import com.openhtmltopdf.css.constants.IdentValue;
 import com.openhtmltopdf.css.style.CalculatedStyle;
+import com.openhtmltopdf.extend.FSTextBreaker;
 import com.openhtmltopdf.render.FSFont;
 
 /**
@@ -109,7 +108,7 @@ private static void doBreakText(LayoutContext c,
             boolean tryToBreakAnywhere) {
         FSFont font = style.getFSFont(c);
         String currentString = context.getStartSubstring();
-        BreakIterator iterator = getWordStream(currentString);
+        FSTextBreaker iterator = getLineBreakStream(currentString, c.getSharedContext());
         int left = 0;
         int right = tryToBreakAnywhere ? 1 : iterator.next();
         int lastWrap = 0;
@@ -174,11 +173,9 @@ private static void doBreakText(LayoutContext c,
         return;
     }
 
-	public static BreakIterator getWordStream(String s) {
-		BreakIterator i = new UrlAwareLineBreakIterator();
+	public static FSTextBreaker getLineBreakStream(String s, SharedContext shared) {
+		FSTextBreaker i = shared.getLineBreaker();
 		i.setText(s);
 		return i;
 	}
-
 }
-

openhtmltopdf-core/src/main/java/com/openhtmltopdf/layout/SharedContext.java

@@ -23,7 +23,9 @@
 import java.awt.HeadlessException;
 import java.awt.Rectangle;
 import java.awt.Toolkit;
+import java.text.BreakIterator;
 import java.util.HashMap;
+import java.util.Locale;
 import java.util.Map;
 
 import org.w3c.dom.Document;
@@ -36,6 +38,7 @@
 import com.openhtmltopdf.css.style.EmptyStyle;
 import com.openhtmltopdf.css.value.FontSpecification;
 import com.openhtmltopdf.extend.FSCanvas;
+import com.openhtmltopdf.extend.FSTextBreaker;
 import com.openhtmltopdf.extend.FontContext;
 import com.openhtmltopdf.extend.FontResolver;
 import com.openhtmltopdf.extend.NamespaceHandler;
@@ -126,6 +129,7 @@ public class SharedContext {
 	private boolean defaultPageSizeIsInches;
 
 	private String replacementText = "#";
+	private FSTextBreaker lineBreaker = new UrlAwareLineBreakIterator(BreakIterator.getLineInstance(Locale.US));
 
     public SharedContext() {
     }
@@ -634,6 +638,14 @@ public void setDefaultPageSize(Float pageWidth, Float pageHeight, boolean isInch
 		this.defaultPageSizeIsInches = isInches;
 	}
 
+	public FSTextBreaker getLineBreaker() {
+		return lineBreaker;
+	}
+
+	public void setLineBreaker(FSTextBreaker breaker) {
+		this.lineBreaker = breaker;
+	}
+
 	/**
 	 * This registers the shared context with a thread local so it
 	 * can be used anywhere. It should be matched with a call to 

openhtmltopdf-core/src/main/java/com/openhtmltopdf/layout/UrlAwareLineBreakIterator.java

@@ -1,37 +1,26 @@
 package com.openhtmltopdf.layout;
 
 import java.text.BreakIterator;
-import java.text.CharacterIterator;
+import com.openhtmltopdf.extend.FSTextBreaker;
 
 
 /**
  * BreakIterator implementation that improves line breaking for URLs. Break points are supported
  * before path fragments.
  */
-public class UrlAwareLineBreakIterator extends BreakIterator {
+public class UrlAwareLineBreakIterator implements FSTextBreaker {
 
     private static final String BREAKING_CHARS = ".,:;!?- \n\r\t/";
 
-    private BreakIterator delegate = BreakIterator.getLineInstance();
+    private final BreakIterator delegate;
     private String text;
     private Range currentRange;
 
-
-    public int preceding(int offset) {
-        throw new UnsupportedOperationException("Not yet implemented");
-    }
-
-
-    public int last() {
-        throw new UnsupportedOperationException("Not yet implemented");
+    public UrlAwareLineBreakIterator(BreakIterator breaker) {
+    	delegate = breaker;
     }
-
-
-    public int previous() {
-        throw new UnsupportedOperationException("Not yet implemented");
-    }
-
-
+
+    @Override
     public int next() {
         checkNotAheadOfDelegate();
 
@@ -103,42 +92,7 @@ private boolean advanceDelegate() {
         return next == BreakIterator.DONE;
     }
 
-
-    public int next(int n) {
-        throw new UnsupportedOperationException("Not yet implemented");
-    }
-
-
-    public boolean isBoundary(int offset) {
-        throw new UnsupportedOperationException("Not yet implemented");
-    }
-
-
-    public int following(int offset) {
-        throw new UnsupportedOperationException("Not yet implemented");
-    }
-
-
-    public int first() {
-        throw new UnsupportedOperationException("Not yet implemented");
-    }
-
-
-    public void setText(CharacterIterator newText) {
-        throw new UnsupportedOperationException("Not yet implemented");
-    }
-
-
-    public int current() {
-        throw new UnsupportedOperationException("Not yet implemented");
-    }
-
-
-    public CharacterIterator getText() {
-        return delegate.getText();
-    }
-
-
+    @Override
     public void setText(String newText) {
         delegate.setText(newText);
         text = newText;
@@ -148,56 +102,46 @@ public void setText(String newText) {
 
     private static class Range {
 
-        int start;
-        int stop;
-
+        private final int start;
+        private final int stop;
 
         public Range(int start, int stop) {
             this.start = start;
             this.stop = Math.max(start, stop);
         }
 
-
         public Range(int referencePoint, int startOffset, int stopOffset) {
             this(referencePoint + startOffset, referencePoint + stopOffset);
         }
 
-
         public Range withStart(int start) {
             return new Range(start, stop);
         }
 
-
         public Range withStop(int stop) {
             return new Range(start, stop);
         }
 
-
         public Range incrementStart() {
             int newStart = start + 1;
             return new Range(newStart, Math.max(newStart, stop));
         }
 
-
         public Range decrementStop() {
             int newStop = stop + -1;
             return new Range(Math.min(start, newStop), newStop);
         }
 
-
         public int getStart() {
             return start;
         }
 
-
         public int getStop() {
             return stop;
         }
 
-
         public String toString() {
             return "[" + start + ", " + stop + ")";
         }
     }
-
 }

openhtmltopdf-core/src/main/java/com/openhtmltopdf/render/InlineBox.java

@@ -28,6 +28,7 @@
 import com.openhtmltopdf.css.extend.ContentFunction;
 import com.openhtmltopdf.css.parser.FSFunction;
 import com.openhtmltopdf.css.style.CalculatedStyle;
+import com.openhtmltopdf.extend.FSTextBreaker;
 import com.openhtmltopdf.layout.Breaker;
 import com.openhtmltopdf.layout.LayoutContext;
 import com.openhtmltopdf.layout.Styleable;
@@ -247,7 +248,7 @@ private int calcMinWidthFromWordLength(
         int lastWord = 0;
 
         String text = getText(trimLeadingSpace);
-        BreakIterator breakIterator = Breaker.getWordStream(text);
+        FSTextBreaker breakIterator = Breaker.getLineBreakStream(text, c.getSharedContext());
 
         // Breaker should be used
         while ( (current = breakIterator.next()) != BreakIterator.DONE) {

openhtmltopdf-core/src/test/java/com/openhtmltopdf/layout/UrlAwareLineBreakIteratorTest.java

@@ -1,7 +1,9 @@
 package com.openhtmltopdf.layout;
 
 import java.text.BreakIterator;
+import java.util.Locale;
 
+import com.openhtmltopdf.extend.FSTextBreaker;
 import com.openhtmltopdf.layout.UrlAwareLineBreakIterator;
 
 import junit.framework.TestCase;
@@ -107,7 +109,7 @@ public void testNext_IncompleteUrl() throws Exception {
 
 
     private void assertBreaksCorrectly(String input, String[] segments) {
-        BreakIterator iterator = new UrlAwareLineBreakIterator();
+        FSTextBreaker iterator = new UrlAwareLineBreakIterator(BreakIterator.getLineInstance(Locale.US));
         iterator.setText(input);
 
         int segmentIndex = 0;

openhtmltopdf-pdfbox/src/main/java/com/openhtmltopdf/pdfboxout/PdfBoxRenderer.java

@@ -60,6 +60,7 @@
 import com.openhtmltopdf.context.StyleReference;
 import com.openhtmltopdf.css.style.CalculatedStyle;
 import com.openhtmltopdf.extend.FSCache;
+import com.openhtmltopdf.extend.FSTextBreaker;
 import com.openhtmltopdf.extend.FSUriResolver;
 import com.openhtmltopdf.extend.HttpStreamFactory;
 import com.openhtmltopdf.extend.NamespaceHandler;
@@ -76,7 +77,6 @@
 import com.openhtmltopdf.resource.XMLResource;
 import com.openhtmltopdf.simple.extend.XhtmlNamespaceHandler;
 import com.openhtmltopdf.util.Configuration;
-import com.openhtmltopdf.util.ThreadCtx;
 import com.openhtmltopdf.util.XRLog;
 
 public class PdfBoxRenderer {
@@ -178,13 +178,14 @@ public PdfBoxRenderer(float dotsPerPoint, int dotsPerPixel, boolean useSubsets,
 
     /**
      * Do not use this method. It is constantly changing as options are added to the builder.
+     * @param lineBreaker 
      */
     public PdfBoxRenderer(boolean textDirection, boolean testMode,
             boolean useSubsets, HttpStreamFactory httpStreamFactory,
             BidiSplitterFactory splitterFactory, BidiReorderer reorderer, String html,
             Document document, String baseUri, String uri, File file,
             OutputStream os, FSUriResolver _resolver, FSCache _cache, SVGDrawer svgImpl,
-            Float pageWidth, Float pageHeight, boolean isPageSizeInches, float pdfVersion, String replacementText) {
+            Float pageWidth, Float pageHeight, boolean isPageSizeInches, float pdfVersion, String replacementText, FSTextBreaker lineBreaker) {
 
         _pdfDoc = new PDDocument();
         _pdfDoc.setVersion(pdfVersion);
@@ -244,6 +245,10 @@ public PdfBoxRenderer(boolean textDirection, boolean testMode,
             this._outputDevice.setBidiReorderer(_reorderer);
         }
 
+        if (lineBreaker != null) {
+            _sharedContext.setLineBreaker(lineBreaker);
+        }
+
         this._defaultTextDirection = textDirection ? BidiSplitter.RTL : BidiSplitter.LTR;
 
         if (html != null) {

openhtmltopdf-pdfbox/src/main/java/com/openhtmltopdf/pdfboxout/PdfRendererBuilder.java

@@ -8,10 +8,10 @@
 import com.openhtmltopdf.bidi.BidiReorderer;
 import com.openhtmltopdf.bidi.BidiSplitterFactory;
 import com.openhtmltopdf.extend.FSCache;
+import com.openhtmltopdf.extend.FSTextBreaker;
 import com.openhtmltopdf.extend.FSUriResolver;
 import com.openhtmltopdf.extend.HttpStreamFactory;
 import com.openhtmltopdf.extend.SVGDrawer;
-import com.openhtmltopdf.swing.NaiveUserAgent;
 
 public class PdfRendererBuilder
 {
@@ -42,6 +42,7 @@ public static enum PageSizeUnits { MM, INCHES };
     private boolean _isPageSizeInches;
     private float _pdfVersion = 1.7f;
     private String _replacementText;
+    private FSTextBreaker _lineBreaker;
 
     /**
      * Run the XHTML/XML to PDF conversion and output to an output stream set by toStream.
@@ -67,7 +68,8 @@ public void run() throws Exception {
     public PdfBoxRenderer buildPdfRenderer() {
         return new PdfBoxRenderer(_textDirection, _testMode, _useSubsets, _httpStreamFactory, _splitter, _reorderer,
                 _html, _document, _baseUri, _uri, _file, _os, _resolver, _cache, _svgImpl,
-                _pageWidth, _pageHeight, _isPageSizeInches, _pdfVersion, _replacementText);
+                _pageWidth, _pageHeight, _isPageSizeInches, _pdfVersion, _replacementText,
+                _lineBreaker);
     }
 
     /**
@@ -256,4 +258,19 @@ public PdfRendererBuilder useReplacementText(String replacement) {
         this._replacementText = replacement;
         return this;
     }
+
+    /**
+     * Specify the line breaker. By default a Java default BreakIterator line instance is used
+     * with US locale. Additionally, this is wrapped with UrlAwareLineBreakIterator to also
+     * break before the forward slash (/) character so that long URIs can be broken on to multiple lines.
+     * 
+     * You may want to use a BreakIterator with a different locale (wrapped by UrlAwareLineBreakIterator or not)
+     * or a more advanced BreakIterator from icu4j (see the rtl-support module for an example).
+     * @param breaker
+     * @return
+     */
+    public PdfRendererBuilder useLineBreaker(FSTextBreaker breaker) {
+        this._lineBreaker = breaker;
+        return this;
+    }
 }