8361381: GlyphLayout behavior differs on JDK 11+ compared to JDK 8

Reviewed-by: prr, serb
This commit is contained in:
Volker Simonis 2025-09-08 13:30:45 +00:00
parent 03c54d4288
commit bcff857ba0
3 changed files with 136 additions and 8 deletions

View File

@ -577,6 +577,7 @@ class ExtendedTextSourceLabel extends ExtendedTextLabel implements Decoration.La
* not all do, and it cannot be relied upon.
* - each glyph maps to a single character, when multiple glyphs exist for a character they all map to it, but
* no two characters map to the same glyph
* This was only true for the old, ICU layout engine which inserted 0xffff glyphs for ligaturized characters!
* - multiple glyphs mapping to the same character need not be in sequence (thai, tamil have split characters)
* - glyphs may be arbitrarily reordered (Indic reorders glyphs)
* - all glyphs share the same bidi level
@ -712,8 +713,6 @@ class ExtendedTextSourceLabel extends ExtendedTextLabel implements Decoration.La
while (gx != gxlimit) {
// start of new cluster
int clusterExtraGlyphs = 0;
minIndex = indices[gx];
maxIndex = minIndex;
@ -730,14 +729,11 @@ class ExtendedTextSourceLabel extends ExtendedTextLabel implements Decoration.La
while (gx != gxlimit &&
((glyphinfo[gp + advx] == 0) ||
(indices[gx] <= maxIndex) ||
(maxIndex - minIndex > clusterExtraGlyphs))) {
(indices[gx] <= maxIndex))) {
++clusterExtraGlyphs; // have an extra glyph in this cluster
if (DEBUG) {
System.err.println("gp=" +gp +" adv=" + glyphinfo[gp + advx] +
" gx="+ gx+ " i[gx]="+indices[gx] +
" clusterExtraGlyphs="+clusterExtraGlyphs);
" gx="+ gx+ " i[gx]="+indices[gx]);
}
// adjust advance only if new glyph has non-zero advance

View File

@ -23,7 +23,7 @@
/* @test
* @summary Test getGlyphCharIndex() results from layout
* @bug 8152680
* @bug 8152680 8361381
*/
import java.awt.Font;
@ -40,5 +40,22 @@ public class GetGlyphCharIndexTest {
if (idx0 != 0) {
throw new RuntimeException("Expected 0, got " + idx0);
}
// This is the encoding-independent Khmer string "បានស្នើសុំនៅតែត្រូវបានបដិសេធ"
// We can't check for more details like e.g. correct line breaking because it is font and platform dependent,
// but we can at least chack that the created GlyphVector has monotonically increasing character indices.
// This is guaranteed by HarfBuzz's HB_BUFFER_CLUSTER_LEVEL_MONOTONE_CHARACTERS cluster level which is used
// in the OpenJDK layout implementation.
String khmer = "\u1794\u17b6\u1793\u179f\u17d2\u1793\u17be\u179f\u17bb\u17c6\u1793\u17c5" +
"\u178f\u17c2\u178f\u17d2\u179a\u17bc\u179c\u1794\u17b6\u1793\u1794\u178a\u17b7\u179f\u17c1\u1792";
font = new Font(Font.DIALOG, Font.PLAIN, 12);
gv = font.layoutGlyphVector(frc, khmer.toCharArray(), 0, khmer.length(), 0);
int[] indices = gv.getGlyphCharIndices(0, gv.getNumGlyphs(), null);
for (int i = 0; i < (indices.length - 1); i++) {
if (indices[i] > indices[i + 1]) {
throw new RuntimeException("Glyph character indices are supposed to be monotonically growing, but character index at position " +
i + " is bigger then the one at position " + (i + 1) + ", i.e. " + indices[i] + " > " + indices[i + 1] + ".");
}
}
}
}

View File

@ -0,0 +1,115 @@
/*
* Copyright Amazon.com Inc. or its affiliates. All Rights Reserved.
* DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS FILE HEADER.
*
* This code is free software; you can redistribute it and/or modify it
* under the terms of the GNU General Public License version 2 only, as
* published by the Free Software Foundation.
*
* This code is distributed in the hope that it will be useful, but WITHOUT
* ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
* FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License
* version 2 for more details (a copy is included in the LICENSE file that
* accompanied this code).
*
* You should have received a copy of the GNU General Public License version
* 2 along with this work; if not, write to the Free Software Foundation,
* Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA.
*
* Please contact Oracle, 500 Oracle Parkway, Redwood Shores, CA 94065 USA
* or visit www.oracle.com if you need additional information or have any
* questions.
*/
/**
* @test
* @bug 8361381
* @summary GlyphLayout behavior differs on JDK 11+ compared to JDK 8
*/
import java.awt.*;
import java.awt.font.FontRenderContext;
import java.awt.font.LineBreakMeasurer;
import java.awt.font.TextAttribute;
import java.text.AttributedCharacterIterator;
import java.text.AttributedString;
import java.text.BreakIterator;
import java.util.Locale;
public class KhmerLineBreakTest {
static String khmer = "បានស្នើសុំនៅតែត្រូវបានបដិសេធ";
/*
This is part of the output we get from `ExtendedTextSourceLabel::createCharinfo()`
when running with `-Dsun.java2d.debugfonts=true`. It's a listing of the 28 code points
of the `khmer` string defined above and displays their x-position during rendering as
well as their advance. Code points with zero advance belong to the glyph cluster which
is started by the first preceding code point with a non-zero advance. There should be no
breaks at characters with zero advance, because this would break a glyph cluster.
0 ch: 1794 x: 0.0 xa: 68.115234
1 ch: 17b6 x: 68.115234 xa: 0.0
2 ch: 1793 x: 68.115234 xa: 45.410156
3 ch: 179f x: 113.52539 xa: 90.82031
4 ch: 17d2 x: 204.3457 xa: 0.0
5 ch: 1793 x: 204.3457 xa: 0.0
6 ch: 17be x: 204.3457 xa: 0.0
7 ch: 179f x: 204.3457 xa: 68.115234
8 ch: 17bb x: 272.46094 xa: 0.0
9 ch: 17c6 x: 272.46094 xa: 0.0
10 ch: 1793 x: 272.46094 xa: 90.82031
11 ch: 17c5 x: 363.28125 xa: 0.0
12 ch: 178f x: 363.28125 xa: 68.115234
13 ch: 17c2 x: 431.39648 xa: 0.0
14 ch: 178f x: 431.39648 xa: 68.115234
15 ch: 17d2 x: 499.51172 xa: 0.0
16 ch: 179a x: 499.51172 xa: 0.0
17 ch: 17bc x: 499.51172 xa: 0.0
18 ch: 179c x: 499.51172 xa: 22.705078
19 ch: 1794 x: 522.2168 xa: 68.115234
20 ch: 17b6 x: 590.33203 xa: 0.0
21 ch: 1793 x: 590.33203 xa: 45.410156
22 ch: 1794 x: 635.7422 xa: 45.410156
23 ch: 178a x: 681.15234 xa: 45.410156
24 ch: 17b7 x: 726.5625 xa: 0.0
25 ch: 179f x: 726.5625 xa: 90.82031
26 ch: 17c1 x: 817.3828 xa: 0.0
27 ch: 1792 x: 817.3828 xa: 45.410156
*/
static boolean[] possibleBreak = new boolean[]
{ true, false, true, true, false, false, false, true, false, false,
true, false, true, false, true, false, false, false, true, true,
false, true, true, true, false, true, false, true, true /* */ };
static Locale locale = new Locale.Builder().setLanguage("km").setRegion("KH").build();
static BreakIterator breakIterator = BreakIterator.getLineInstance(locale);
static FontRenderContext frc = new FontRenderContext(null, true, true);
public static void main(String[] args) {
Font[] allFonts = GraphicsEnvironment.getLocalGraphicsEnvironment().getAllFonts();
for (int i=0; i < allFonts.length; i++) {
if (allFonts[i].canDisplayUpTo(khmer) == -1) {
Font font = allFonts[i].deriveFont(Font.PLAIN, 60f);
System.out.println("Trying font: " + font.getFontName());
AttributedString attrStr = new AttributedString(khmer);
attrStr.addAttribute(TextAttribute.FONT, font);
AttributedCharacterIterator it = attrStr.getIterator();
for (int width = 200; width < 400; width += 10) {
LineBreakMeasurer measurer = new LineBreakMeasurer(it, breakIterator, frc);
System.out.print(width + " : ");
while (measurer.getPosition() < it.getEndIndex()) {
int nextOffset = measurer.nextOffset(width);
System.out.print(nextOffset + " ");
if (!possibleBreak[nextOffset]) {
System.out.println();
throw new RuntimeException("Invalid break at offset " + nextOffset + " (width = " + width + " font = " + font.getFontName() + ")");
}
measurer.setPosition(nextOffset);
}
System.out.println();
}
System.out.println("OK");
}
}
}
}