JDK-4858912 : JCK1.4a-runtime api/java_awt/interactive/GraphicsTests.html#GraphicsTests fails
  • Type: Bug
  • Component: client-libs
  • Sub-Component: 2d
  • Affected Version: 1.4.1,1.4.2,5.0
  • Priority: P4
  • Status: Closed
  • Resolution: Duplicate
  • OS: linux,windows_xp
  • CPU: x86
  • Submitted: 2003-05-06
  • Updated: 2005-10-05
  • Resolved: 2005-10-05
Related Reports
Duplicate :  
Relates :  
Relates :  
Description
Name: iaR10016			Date: 05/06/2003


Filed By       : J2SE-SQA [###@###.###
JDK            : JDK1.4.2-b22, JDK1.5.0-b05, JDK1.4.1
JCK            : JCK1.4a-runtime
Platform[s]    : all
switch/Mode    : default
JCK test owner : http://javaweb.eng/jck/usr/owners.jto
Falling test[s]: api/java_awt/interactive/GraphicsTests.html#GraphicsTests [GraphicsTest0001]

It seems like nethods Graphics.fillPolygon(int[], int[], int),
Graphics.fillRoundRect(int, int, int, int, int, int), Graphics.drawRoundRect(int, int, int, int, int, int)
works incorrectly with big numbers.

Please, run the script below and do the following steps to reprodyce the failure:

1. set tested Width and Height for Clip area to "2147483647" (max integer number) and click "Apply" button

2. Select "RoundRect" in "Area" choice list, set (for example) Height to 2000000 and click "Apply".
    RoundRect does not look as expected.

3. Click "Fill" and then click "Apply".
    Filled RoundRect does not look as expected.

4. Select "Poligon" in "Area" choice list, set (for example) X3 to "2000000" and click "Apply".
    Poligon looks as expected.
    Click "Fill" and then click "Apply".
    Filled poligon does not look as expected.

Test source location:
=====================
/java/re/jck/1.4a/archive/fcs/binaries/JCK-runtime-14a/tests/api/java_awt/interactive/GraphicsTests.java

jtr file location:
==================
/net/jtgb4u4c.sfbay/export/sail15/results/mantis/b22/jck14a/linux/redhat8.0_gnomens4.7_linux-13/workDir/api-interactive/java_awt/interactive/GraphicsTests_GraphicsTests.jtr

How to reproduce:
=================
Run the following script (you may need to change it's variables)
--- script start ---
#!/bin/bash
JCK="/java/re/jck/1.4a/archive/fcs/binaries/JCK-runtime-14a"
JDK="/java/re/jdk/1.5.0/promoted/latest/binaries/linux-i586"
export CLASSPATH="$JCK/classes:$JCK/lib/javatest.jar"
$JDK/bin/java javasoft.sqe.tests.api.java.awt.interactive.GraphicsTests       -TestCaseID GraphicsTest0001       -TestDirURL file:$JCK/tests/api/java_awt/interactive/GraphicsTests.html
--- script end ---

Specific machine info:
======================
Hostname: linux-13
OS: RedHat Linux 8.0

======================================================================

Comments
EVALUATION This is a duplicate of the well known problem 4265778.
05-10-2005

EVALUATION These tests belong to 2D. ###@###.### 2003-05-06 Name: spR10137 Date: 08/27/2003 This effect is reproducable with all values with set bit in 16th position. The reason is XDrawLine call. Xlib declares this call with int coordinates, but X protocol truncate them to short. So any value with 16th bit set becomes the great negative value while this truncation. I see two ways for bug: a) we can try to compensate this effect for X11 renderer; b) we can close this bug because it's not 2d problem but X protocol. ====================================================================== Name: spR10137 Date: 10/24/2003 I propose to dplit this bug into at least three to localize the problems. 1) Problem with RoundRect (clamping problem) - moved to bugId 4937390; 2) Problem with Poligons (X11 case) - moved to bugId 4943522; 3) Same problems in windows case; this case does not refer to clamping to short, we have wrong results with very strange values; need to investigate this case more carefully. ======================================================================
21-08-2004