Gh-426 improve multi byte text rendering (#429)

Author: pcorless

core/core-awt/src/main/java/org/icepdf/core/pobjects/HexStringObject.java

@@ -234,15 +234,17 @@ public StringBuilder getLiteralStringBuffer(final int fontFormat, FontFile font)
             int charOffset = 2;
             int length = getLength();
             int charValue;
+            boolean notUCS2 = font.getToUnicode() != null
+                    && font.getToUnicode().getName() != null
+                    && !font.getToUnicode().getName().contains("UCS2");
             StringBuilder tmp = new StringBuilder(length);
             // attempt to detect mulibyte encoded strings.
             for (int i = 0; i < length; i += charOffset) {
                 String first = stringData.substring(i, i + 2);
                 if (first.charAt(0) != '0') {
                     // check range for possible 2 byte char ie mixed mode.
                     charValue = getUnsignedInt(first);
-                    if (font.getByteEncoding() == FontFile.ByteEncoding.MIXED_BYTE &&
-                            font.canDisplay((char) charValue) && font.getSource() != null) {
+                    if (notUCS2 && font.canDisplay((char) charValue) && font.getSource() != null) {
                         tmp.append((char) charValue);
                     } else {
                         charValue = getUnsignedInt(i, 4);

core/core-awt/src/main/java/org/icepdf/core/pobjects/LiteralStringObject.java

@@ -186,28 +186,16 @@ public StringBuilder getLiteralStringBuffer(final int fontFormat, FontFile font)
             int length = getLength();
             int charValue;
             StringBuilder tmp = new StringBuilder(length);
-            if (font.getByteEncoding() == FontFile.ByteEncoding.MIXED_BYTE) {
-                int charOffset = 1;
-                for (int i = 0; i < length; i += charOffset) {
-                    // check range for possible 2 byte char.
+            for (int i = 0; i < length; i += 1) {
+                // check range for possible 2 byte char.
+                charValue = getUnsignedInt(i, 2);
+                if (font.canDisplay((char) charValue)) {
+                    tmp.append((char) charValue);
+                    i += 1;
+                } else {
                     charValue = getUnsignedInt(i, 1);
                     if (font.canDisplay((char) charValue)) {
                         tmp.append((char) charValue);
-                    } else {
-                        int charValue2 = getUnsignedInt(i, 2);
-                        if (font.canDisplay((char) charValue2)) {
-                            tmp.append((char) charValue2);
-                            i += 1;
-                        }
-                    }
-                }
-            } else {
-                // we have default 2bytes.
-                int charOffset = 2;
-                for (int i = 0; i < length; i += charOffset) {
-                    int charValue2 = getUnsignedInt(i, 2);
-                    if (font.canDisplay((char) charValue2)) {
-                        tmp.append((char) charValue2);
                     }
                 }
             }

core/core-awt/src/main/java/org/icepdf/core/pobjects/fonts/CMap.java

@@ -25,64 +25,4 @@
 public interface CMap {
 
     Name TYPE = new Name("CMap");
-
-    void init();
-
-    /**
-     * Maps the character id to an underlying unicode value if available.
-     *
-     * @param ch character code to find unicode value of.
-     * @return unicode value of ch if available otherwise original ch is returned unaltered.
-     */
-    char toSelector(char ch);
-
-    char toSelector(char ch, boolean isCFF);
-
-    char fromSelector(char ch);
-
-    /**
-     * Maps the character id to an underlying to unicode table. This method should
-     * be called when looking for a unicode value for a CID.  This method differs
-     * slightly from #toSelector in that it can return at String rather than a
-     * single character code.
-     *
-     * @param ch character id to look for corresponding unicode values.
-     * @return unicode value of specified character code.
-     */
-    String toUnicode(char ch);
-
-    /**
-     * Determines if the cid should be interpreted as a one or two byte character.
-     * Some CID fonts use the one byte notation but the two byte is the most
-     * common bar far.
-     *
-     * @return true if the cid should be considered as having a one byte length.
-     */
-    boolean isOneByte();
-
-    /**
-     * Determines if the cid should be interpreted as a one or two byte character.
-     * Some CID fonts use the one byte notation but the two byte is the most
-     * common bar far.
-     *
-     * @return true if the cid should be considered as having a two byte length.
-     */
-    boolean isTwoByte();
-
-    /**
-     * Determines if the cid should be interpreted as a one or two byte character.
-     * Some CID fonts use the one byte notation but the two byte is the most
-     * common bar far.  A mixed byte string must be parsed differently as the font
-     * can be used to determine the number of bytes used for each character.
-     *
-     * @return true if the cid should be considered as having a mixed byte length.
-     */
-    boolean isMixedByte();
-
-    /**
-     * Utility method ot check if a CMap contain s any data, specifically usable toUnicode data.
-     *
-     * @return true if the mapping contains at least one entry.
-     */
-    boolean isEmptyMapping();
 }

core/core-awt/src/main/java/org/icepdf/core/pobjects/fonts/Font.java

@@ -15,7 +15,7 @@
  */
 package org.icepdf.core.pobjects.fonts;
 
-
+import org.apache.fontbox.cmap.CMap;
 import org.icepdf.core.pobjects.Dictionary;
 import org.icepdf.core.pobjects.DictionaryEntries;
 import org.icepdf.core.pobjects.Name;
@@ -394,6 +394,15 @@ public int getCharacterCount() {
         return lastchar - firstchar + 1;
     }
 
+    /**
+     * <p>Returns the CMap used to map characters to glyphs.</p>
+     *
+     * @return CMap used to map characters to glyphs.
+     */
+    public boolean hasUnicodeCMap() {
+        return toUnicodeCMap != null;
+    }
+
     /**
      * <p>Returns true if the writing mode is vertical; false, otherwise</p>
      *

core/core-awt/src/main/java/org/icepdf/core/pobjects/fonts/FontFactory.java

@@ -103,10 +103,6 @@ else if (FONT_SUBTYPE_CID_FONT_TYPE_0.equals(subtype)) {
         } else if (FONT_SUBTYPE_CID_FONT_TYPE_2.equals(subtype)) {
             font = new TypeCidType2Font(library, entries);
         }
-        if (font == null) {
-            // create OFont implementation.
-            font = new org.icepdf.core.pobjects.fonts.ofont.Font(library, entries);
-        }
         return font;
     }
 

core/core-awt/src/main/java/org/icepdf/core/pobjects/fonts/FontFile.java

@@ -15,6 +15,8 @@
  */
 package org.icepdf.core.pobjects.fonts;
 
+import org.apache.fontbox.cmap.CMap;
+
 import java.awt.*;
 import java.awt.geom.AffineTransform;
 import java.awt.geom.Point2D;
@@ -57,8 +59,6 @@ FontFile deriveFont(Map<Integer, Float> widths, int firstCh, float missingWidth,
 
     boolean canDisplay(char ech);
 
-    void setIsCid();
-
     CMap getToUnicode();
 
     String toUnicode(String displayText);

core/core-awt/src/main/java/org/icepdf/core/pobjects/fonts/FontManager.java

@@ -23,8 +23,8 @@
 import java.io.File;
 import java.io.IOException;
 import java.net.URL;
-import java.util.List;
 import java.util.*;
+import java.util.List;
 import java.util.logging.Level;
 import java.util.logging.Logger;
 import java.util.prefs.Preferences;
@@ -107,27 +107,37 @@ public class FontManager {
             };
 
     private static final String[] JAPANESE_FONT_NAMES = {
+            // windows
             "Arial Unicode MS", "PMingLiU", "MingLiU",
             "MS PMincho", "MS Mincho", "Kochi Mincho", "Hiragino Mincho Pro",
-            "KozMinPro Regular Acro", "HeiseiMin W3 Acro", "Adobe Ming Std Acro"
+            "KozMinPro Regular Acro", "HeiseiMin W3 Acro", "Adobe Ming Std Acro",
+            // linux
+            "ipaexmincho", "Kochi Gothic", "Hiragino Kaku Gothic Pro",
+
     };
 
     private static final String[] CHINESE_SIMPLIFIED_FONT_NAMES = {
             "Arial Unicode MS", "PMingLiU", "MingLiU",
             "SimSun", "NSimSun", "Kochi Mincho", "STFangsong", "STSong Light Acro",
-            "Adobe Song Std Acro", "stsong"
+            "Adobe Song Std Acro", "stsong",
+            // linux
+            "ipaexmincho", "Kochi Gothic", "Hiragino Kaku Gothic Pro",
     };
 
     private static final String[] CHINESE_TRADITIONAL_FONT_NAMES = {
             "Arial Unicode MS", "PMingLiU", "MingLiU",
             "SimSun", "NSimSun", "Kochi Mincho", "BiauKai", "MSungStd Light Acro",
-            "Adobe Song Std Acro"
+            "Adobe Song Std Acro",
+            // linux
+            "umingcn", "ipaexmincho", "Kochi Gothic", "Hiragino Kaku Gothic Pro",
     };
 
     private static final String[] KOREAN_FONT_NAMES = {
             "Arial Unicode MS", "Dotum", "Gulim", "New Gulim", "GulimChe", "Batang",
             "BatangChe", "HYSMyeongJoStd Medium Acro", "Adobe Myungjo Std Acro",
-            "AppleGothic", "Malgun Gothic", "UnDotum", "UnShinmun", "Baekmuk Gulim"
+            "AppleGothic", "Malgun Gothic", "UnDotum", "UnShinmun", "Baekmuk Gulim",
+            // linux
+            "ipaexmincho", "Kochi Gothic", "Hiragino Kaku Gothic Pro",
     };
 
     /**

core/core-awt/src/main/java/org/icepdf/core/pobjects/fonts/ofont/Encoding.java

@@ -18,9 +18,8 @@
 import java.util.HashMap;
 
 /**
- * @deprecated
+ *
  */
-@Deprecated
 public class Encoding implements org.icepdf.core.pobjects.fonts.Encoding {
     private final char[] enc;