This is the mail archive of the
mailing list for the Mauve project.
Re: Mauve regressions for J2SDK 1.4.2
- From: Sascha Brawer <brawer at dandelis dot ch>
- To: <mauve-discuss at sources dot redhat dot com>, Stephen Crawley <crawley at dstc dot edu dot au>
- Date: Mon, 1 Mar 2004 09:36:40 +0100
- Subject: Re: Mauve regressions for J2SDK 1.4.2
- References: <200402290732.i1T7WQOR011001@piglet.dstc.edu.au>
Stephen Crawley <email@example.com> wrote on Sun, 29 Feb 2004 17:32:28
>As a sanity check of the mauve testcases, I ran them against the Java
>compiler and JVM in Sun's JDK 1.4.2. The results are attached.
>Without looking at the 106 failures in detail, I suspect that most are
>due to bugs in the testcases themselves.
I can only comment on the FAILs for java.awt.geom: In the Sun JDK
1.4.1_01, the methods CubicCurve2D.solveCubic and
QuadCurve2D.solveQuadratic are indeed not able to find many existing
solutions. In other words, those bugs are real, not in the test case.
Maybe we need an XFAIL mechanism? This would allow us to specify which
testlets are known to fail on what VMs.
Sascha Brawer, firstname.lastname@example.org, http://www.dandelis.ch/people/brawer/