bug-classpath
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[Bug awt/56229] New: Area.add sometimes incorrectly creates non-singular


From: scottg at mantatest dot com
Subject: [Bug awt/56229] New: Area.add sometimes incorrectly creates non-singular shape
Date: Wed, 06 Feb 2013 16:44:12 +0000


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=56229



             Bug #: 56229

           Summary: Area.add sometimes incorrectly creates non-singular

                    shape

    Classification: Unclassified

           Product: classpath

           Version: 0.98

            Status: UNCONFIRMED

          Severity: normal

          Priority: P3

         Component: awt

        AssignedTo: address@hidden

        ReportedBy: address@hidden





Created attachment 29372

  --> http://gcc.gnu.org/bugzilla/attachment.cgi?id=29372

Command-line test program: OK on OpenJDK/Sun JRE, etc., but spurious false at

0.50 on Classpath



I ran the attached command-line (no graphics shown) program using:

 - Fedora 15

 - JamVM 1.5.4

 - Classpath 0.98



What it does:

 - Create a GeneralPath containing a 1x1 square, with the lower-left corner at

the origin

 - Create an Ellipse2D.Float as a radius-0.5 circle, with its center at Y=1

 - Modify the center X of the circle, and for each position:

  - Create an Area from the circle

  - Area.add the square

  - Use Area.isSingular to see whether the shape is properly closed



On OpenJDK, the result is true from X = -1.45 to +1.45, and otherwise false.



With Classpath, there's a spurious false when X = 0.50.



My original observation was in a different environment: old (2005) GCJ, with

just the java.awt.geom stuff copied over from Classpath 0.98.  The observed

symptom was that certain shapes wouldn't add properly, resulting in various

internal gaps in the resulting shape.



reply via email to

[Prev in Thread] Current Thread [Next in Thread]