Re: [OpenJDK 2D-Dev] CubicCurve2D.solveCubic and containment/intersection bugs.

2011-01-26 Thread Jim Graham
Oh, and I just looked at the tests now and they look good too... ...jim On 1/26/2011 12:29 PM, Denis Lila wrote: The URL is the same: http://icedtea.classpath.org/~dlila/webrevs/cc2d/webrev/ Looks good, other than the minor issues above (most of which are requests for

Re: [OpenJDK 2D-Dev] CubicCurve2D.solveCubic and containment/intersection bugs.

2011-01-26 Thread Jim Graham
Whoops, one last optimization: Line 1161 - "num > 1"? One last suggestion: Line 1271 - if the vals are opposite sign, is it worth bisectRoot'ing? A couple final comments: Line 1173 - add "num > 1"? Line 1251 - a referring comment should be added to line 1154 and/or 1157 otherwise someone mi

[OpenJDK 2D-Dev] hg: jdk7/2d/jdk: 6940890: Java doesn't pick up the correct fontconfig files in latest Solaris Next builds

2011-01-26 Thread philip . race
Changeset: d0e158473b6f Author:prr Date: 2011-01-26 13:26 -0800 URL: http://hg.openjdk.java.net/jdk7/2d/jdk/rev/d0e158473b6f 6940890: Java doesn't pick up the correct fontconfig files in latest Solaris Next builds Reviewed-by: bae, igor ! src/share/classes/sun/java2d/SunGraphicsEn

Re: [OpenJDK 2D-Dev] CubicCurve2D.solveCubic and containment/intersection bugs.

2011-01-26 Thread Denis Lila
Hi Jim. > It doesn't really affect anything. I can get used to it, but I'm > worried that it might raise a "Why did the previous engineer make this > final? Is it OK if I have to make it non-final?" maintenance question > that might make a future bug fixer stumble a little. It might also > convey

[OpenJDK 2D-Dev] hg: jdk7/2d/jdk: 7014738: Update jdk repo application manifests with Windows 7 compatibility section.

2011-01-26 Thread philip . race
Changeset: bad0ddc6f573 Author:prr Date: 2011-01-26 11:46 -0800 URL: http://hg.openjdk.java.net/jdk7/2d/jdk/rev/bad0ddc6f573 7014738: Update jdk repo application manifests with Windows 7 compatibility section. Reviewed-by: bae, igor ! src/windows/resource/java.manifest

Re: [OpenJDK 2D-Dev] CubicCurve2D.solveCubic and containment/intersection bugs.

2011-01-26 Thread Jim Graham
On 1/26/2011 7:31 AM, Denis Lila wrote: You make a lot of local variables final - is that for performance? Does it actually have any impact (I've only ever used final on a local variable when it was required for an anonymous inner class to work). Not really - I just like using final for variabl

Re: [OpenJDK 2D-Dev] CubicCurve2D.solveCubic and containment/intersection bugs.

2011-01-26 Thread Denis Lila
Hi Jim. > You make a lot of local variables final - is that for performance? > Does > it actually have any impact (I've only ever used final on a local > variable when it was required for an anonymous inner class to work). Not really - I just like using final for variables that aren't supposed to