Add support for UTF-8/UTF-16 strings through DOCS

src/net/sf/jcgm/core/CGM.java

@@ -80,6 +80,8 @@ public class CGM implements Cloneable {
 	private boolean realPrecisionProcessed = false;
 	private RestrictedTextType.Type restrictedTextType = RestrictedTextType.Type.BASIC;
 	private DeviceViewportSpecificationMode.Mode deviceViewportSpecificationMode = DeviceViewportSpecificationMode.Mode.FractionOfDrawingSurface;
+	/** ISO/IEC 8632-1 §6.3.4.5: the default encoding is ISO 8859-1. */
+	private String currentCharSet = "ISO8859-1";
 
 	private final List<ICommandListener> commandListeners = new ArrayList<ICommandListener>();
 
@@ -731,6 +733,14 @@ public void setEndMetafileReached() {
 	public boolean shouldFlattenCurve() {
 		return this.shouldFlattenCurve;
 	}
+
+	public void setCurrentCharSet(String charSet) {
+		this.currentCharSet = charSet;
+	}
+
+	public String getCurrentCharSet() {
+		return this.currentCharSet;
+	}
 }
 
 /*

src/net/sf/jcgm/core/Command.java

@@ -166,11 +166,39 @@ final protected String makeString() {
 			c[i] = makeByte();
 		}
 
+		return makeString(c);
+	}
+
+	final protected String makeString(byte[] characters) {
+		// attempt to detect an encoding switch using DOCS (DECIDE OTHER CODING SYSTEM). See ISO/IEC 2022 or ECMA-35.
+		// this specifically tests for the most common ones, UTF-8 and UTF-16 as described by
+		// "ISO-IR: Internation Register of Coded Character Sets to be used with Escape Sequences"
+		// in chapter 2.8.2 "Coding Systems without Standard return".
+		// they take the form of ESC 2/5 2/15 F ("<ESC> % / F") where F specifies the coding system.
+		if (characters.length >= 4 && characters[0] == 0x1B && characters[1] == 0x25 && characters[2] == 0x2F)
+		{
+			if (characters[3] >= 0x47 && characters[3] <= 0x49)
+			{
+				// ESC 2/5 2/15 4/7: UTF-8 Level 1
+				// ESC 2/5 2/15 4/8: UTF-8 Level 2
+				// ESC 2/5 2/15 4/9: UTF-8 Level 3
+				this.cgm.setCurrentCharSet("UTF-8");
+			}
+			else
+			{
+				// ESC 2/5 2/15 4/10: UTF-16 Level 1
+				// ESC 2/5 2/15 4/11: UTF-16 Level 2
+				// ESC 2/5 2/15 4/12: UTF-16 Level 3
+				this.cgm.setCurrentCharSet("UTF-16BE");
+			}
+			characters = Arrays.copyOfRange(characters, 4, characters.length);
+		}
+
 		try {
-			return new String(c, "ISO8859-1");
+			return new String(characters, this.cgm.getCurrentCharSet());
 		}
 		catch (UnsupportedEncodingException e) {
-			return new String(c);
+			return new String(characters);
 		}
 	}
 

src/net/sf/jcgm/core/FontList.java

@@ -81,10 +81,12 @@ public FontList(int ec, int eid, int l, DataInput in, CGM cgm)
 		count = 0;
 		i = 0;
 		while (i < this.args.length) {
-			char a[] = new char[this.args[i]];
+			byte a[] = new byte[this.args[i]];
 			for (int j = 0; j < this.args[i]; j++)
-				a[j] = (char) this.args[i + j + 1];
-			this.fontNames[count] = new String(a);
+				a[j] = (byte) this.args[i + j + 1];
+			// NOTE: this should be a fixed string, but some metafile generators use a regular string (based on the current encoding).
+			//       with how makeString produces its result, this should be ok to use and be more forgiving.
+			this.fontNames[count] = makeString(a);
 			count++;
 			i += this.args[i] + 1;
 		}

src/net/sf/jcgm/examples/Analyzer.java

@@ -31,6 +31,7 @@
 import java.io.IOException;
 import java.io.InputStream;
 import java.io.PrintWriter;
+import java.nio.charset.Charset;
 import java.util.Map;
 import java.util.Set;
 import java.util.TreeMap;
@@ -94,13 +95,15 @@ public class Analyzer implements ICommandListener {
 				this.messagesFile = new PrintWriter(new BufferedWriter(
 						new FileWriter(new File(this.outputPath
 								.getAbsolutePath()
-								+ File.separator + "messages.txt"))));
+								+ File.separator + "messages.txt"),
+								Charset.forName("UTF-8"))));
 
 				if (verbose) {
 					this.commandFile = new PrintWriter(new BufferedWriter(
 							new FileWriter(new File(this.outputPath
 									.getAbsolutePath()
-									+ File.separator + "commands.txt"))));
+									+ File.separator + "commands.txt"),
+									Charset.forName("UTF-8"))));
 				}
 			}