Rework font selection strategies and add a new one (char by char) and add logic for diacritics

DEVSIX-2052

Author: introfog

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

@@ -38,6 +38,17 @@ public final class TextUtil {
     private TextUtil() {
     }
 
+    /**
+     * Checks if the passed code point corresponds to diacritic.
+     *
+     * @param codePoint the code point to check
+     *
+     * @return {@code true} if passed code point is diacritic, {@code false} otherwise
+     */
+    public static boolean isDiacritic(int codePoint) {
+        return codePoint > 0x0300 && codePoint <= 0x036F;
+    }
+
     /**
      * Check if the value of a character belongs to a certain interval
      * that indicates it's the higher part of a surrogate pair.
@@ -71,7 +82,7 @@ public static char lowSurrogate(int codePoint) {
 
     /**
      * Checks if two subsequent characters in a String are
-     * are the higher and the lower character in a surrogate
+     * the higher and the lower character in a surrogate
      * pair (and therefore eligible for conversion to a UTF 32 character).
      *
      * @param text the String with the high and low surrogate characters
@@ -86,7 +97,7 @@ && isSurrogateHigh(text.charAt(idx))
 
     /**
      * Checks if two subsequent characters in a character array are
-     * are the higher and the lower character in a surrogate
+     * the higher and the lower character in a surrogate
      * pair (and therefore eligible for conversion to a UTF 32 character).
      *
      * @param text the character array with the high and low surrogate characters

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

@@ -112,6 +112,12 @@ public void isMarkNegativeTest() {
         Assert.assertFalse(TextUtil.isMark(glyph));
     }
 
+    @Test
+    public void isDiacriticTest() {
+        Assert.assertTrue(TextUtil.isDiacritic("\u0303".charAt(0)));
+        Assert.assertFalse(TextUtil.isDiacritic("\u006b".charAt(0)));
+    }
+
     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/font/ComplexFontSelectorStrategy.java

@@ -25,6 +25,7 @@ This file is part of the iText (R) project.
 import com.itextpdf.io.font.otf.Glyph;
 import com.itextpdf.io.util.TextUtil;
 import com.itextpdf.kernel.font.PdfFont;
+import com.itextpdf.layout.font.selectorstrategy.FirstMatchFontSelectorStrategy;
 
 import java.util.ArrayList;
 import java.util.List;
@@ -33,7 +34,9 @@ This file is part of the iText (R) project.
  * Complex FontSelectorStrategy split text based on {@link java.lang.Character.UnicodeScript}.
  * If unicode script changes, a new font will be found.
  * If there is no suitable font, only one notdef glyph from {@link FontSelector#bestMatch()} will be added.
+ * @deprecated was replaced by {@link FirstMatchFontSelectorStrategy}.
  */
+@Deprecated
 public class ComplexFontSelectorStrategy extends FontSelectorStrategy {
 
     private PdfFont font;

layout/src/main/java/com/itextpdf/layout/font/FontProvider.java

@@ -22,19 +22,22 @@ This file is part of the iText (R) project.
  */
 package com.itextpdf.layout.font;
 
+import com.itextpdf.commons.utils.FileUtil;
 import com.itextpdf.io.font.FontCache;
 import com.itextpdf.io.font.FontProgram;
 import com.itextpdf.io.font.FontProgramFactory;
 import com.itextpdf.io.font.PdfEncodings;
 import com.itextpdf.io.font.Type1Font;
 import com.itextpdf.io.font.constants.StandardFonts;
-import com.itextpdf.commons.utils.FileUtil;
 import com.itextpdf.kernel.exceptions.PdfException;
 import com.itextpdf.kernel.font.PdfFont;
 import com.itextpdf.kernel.font.PdfFontFactory;
 import com.itextpdf.kernel.font.PdfFontFactory.EmbeddingStrategy;
 import com.itextpdf.kernel.pdf.PdfDocument;
 import com.itextpdf.layout.exceptions.LayoutExceptionMessageConstant;
+import com.itextpdf.layout.font.selectorstrategy.FirstMatchFontSelectorStrategy.FirstMathFontSelectorStrategyFactory;
+import com.itextpdf.layout.font.selectorstrategy.IFontSelectorStrategy;
+import com.itextpdf.layout.font.selectorstrategy.IFontSelectorStrategyFactory;
 
 import java.io.IOException;
 import java.util.ArrayList;
@@ -53,7 +56,7 @@ This file is part of the iText (R) project.
  * <p>
  * It is allowed to use only one {@link FontProvider} per document. If additional fonts per element needed,
  * another instance of  {@link FontSet} can be used. For more details see {@link com.itextpdf.layout.properties.Property#FONT_SET},
- * {@link #getPdfFont(FontInfo, FontSet)}, {@link #getStrategy(String, List, FontCharacteristics, FontSet)}.
+ * {@link #getPdfFont(FontInfo, FontSet)}, {@link #createFontSelectorStrategy(List, FontCharacteristics, FontSet)}.
  * <p>
  * Note, FontProvider does not close created {@link FontProgram}s, because of possible conflicts with {@link FontCache}.
  */
@@ -69,6 +72,8 @@ public class FontProvider {
     protected final String defaultFontFamily;
     protected final Map<FontInfo, PdfFont> pdfFonts;
 
+    private IFontSelectorStrategyFactory fontSelectorStrategyFactory;
+
     /**
      * Creates a new instance of FontProvider.
      *
@@ -105,6 +110,7 @@ public FontProvider(FontSet fontSet, String defaultFontFamily) {
         pdfFonts = new HashMap<>();
         fontSelectorCache = new FontSelectorCache(this.fontSet);
         this.defaultFontFamily = defaultFontFamily;
+        this.fontSelectorStrategyFactory = new FirstMathFontSelectorStrategyFactory();
     }
 
     /**
@@ -364,7 +370,9 @@ public boolean getDefaultEmbeddingFlag() {
      *                        font selector strategy instance and will not be otherwise preserved in font provider.
      *
      * @return {@link FontSelectorStrategy} instance.
+     * @deprecated use {@link #createFontSelectorStrategy(List, FontCharacteristics, FontSet)}
      */
+    @Deprecated
     public FontSelectorStrategy getStrategy(String text, List<String> fontFamilies, FontCharacteristics fc, FontSet additionalFonts) {
         return new ComplexFontSelectorStrategy(text, getFontSelector(fontFamilies, fc, additionalFonts), this, additionalFonts);
     }
@@ -379,7 +387,9 @@ public FontSelectorStrategy getStrategy(String text, List<String> fontFamilies,
      * @param fc instance of {@link FontCharacteristics} to create {@link FontSelector} for sequences of glyphs.
      *
      * @return {@link FontSelectorStrategy} instance.
+     * @deprecated use {@link #createFontSelectorStrategy(List, FontCharacteristics, FontSet)}
      */
+    @Deprecated
     public FontSelectorStrategy getStrategy(String text, List<String> fontFamilies, FontCharacteristics fc) {
         return getStrategy(text, fontFamilies, fc, null);
     }
@@ -393,11 +403,44 @@ public FontSelectorStrategy getStrategy(String text, List<String> fontFamilies,
      * @param fontFamilies target font families to create {@link FontSelector} for sequences of glyphs.
      *
      * @return {@link FontSelectorStrategy} instance.
+     * @deprecated use {@link #createFontSelectorStrategy(List, FontCharacteristics, FontSet)}
      */
+    @Deprecated
     public FontSelectorStrategy getStrategy(String text, List<String> fontFamilies) {
         return getStrategy(text, fontFamilies, null);
     }
 
+    /**
+     * Sets factory which will be used in {@link #createFontSelectorStrategy(List, FontCharacteristics, FontSet)}
+     * method.
+     *
+     * @param factory the factory which will be used to create font selector strategies
+     */
+    public void setFontSelectorStrategyFactory(IFontSelectorStrategyFactory factory) {
+        this.fontSelectorStrategyFactory = factory;
+    }
+
+    /**
+     * Creates the {@link IFontSelectorStrategy} to split text into sequences of glyphs, already tied
+     * to the fonts which contain them. The fonts can be taken from the added fonts to the font provider and
+     * are chosen based on font-families list and desired font characteristics.
+     *
+     * @param fontFamilies target font families to create {@link FontSelector} for sequences of glyphs.
+     * @param fc instance of {@link FontCharacteristics} to create {@link FontSelector} for sequences of glyphs.
+     * @param additionalFonts set which provides fonts additionally to the fonts added to font provider.
+     *                        Combined set of font provider fonts and additional fonts is used when choosing
+     *                        a single font for a sequence of glyphs. Additional fonts will only be used for the given
+     *                        font selector strategy instance and will not be otherwise preserved in font provider.
+     *
+     * @return {@link IFontSelectorStrategy} instance
+     */
+    public IFontSelectorStrategy createFontSelectorStrategy(List<String> fontFamilies,
+            FontCharacteristics fc, FontSet additionalFonts) {
+
+        final FontSelector fontSelector = getFontSelector(fontFamilies, fc, additionalFonts);
+        return fontSelectorStrategyFactory.createFontSelectorStrategy(this, fontSelector, additionalFonts);
+    }
+
     /**
      * Create {@link FontSelector} or get from cache.
      *

layout/src/main/java/com/itextpdf/layout/font/FontSelectorStrategy.java

@@ -24,13 +24,16 @@ This file is part of the iText (R) project.
 
 import com.itextpdf.io.font.otf.Glyph;
 import com.itextpdf.kernel.font.PdfFont;
+import com.itextpdf.layout.font.selectorstrategy.IFontSelectorStrategy;
 
 import java.util.List;
 
 /**
  * {@link FontSelectorStrategy} is responsible for splitting text into sub texts with one particular font.
  * {@link #nextGlyphs()} will create next sub text and set current font.
+ * @deprecated replaced by {@link IFontSelectorStrategy}.
  */
+@Deprecated
 public abstract class FontSelectorStrategy {
 
     protected String text;