Re: [piccolo2d-dev] Re: Release Piccolo2D.Java 1.3
Chris wrote: This release is generally solid, with the exception of the pswing package. Since issue 163 has been reopened, and since I've investigated further, I think that pswing has some new problems that weren't in 1.2 (issue 163 being an example). I can't +1 the release, since my client (PhET) relies heavily on pswing. But I can't -1 the release because I think that fixing pswing and bringing it up to the same standards as the reset of Piccolo2D is going to take considerable time, and I hate to see 1.3 delayed any longer. If issue 163 were fixed, would that change your vote to +1? I wouldn't mind waiting 1.3 until that were the case. If anyone else feels strongly about pswing, then my lack of confidence in its current state may influence your vote. Otherwise, I think PhET should take the lead in fixing improving whatever version of pswing ships with 1.3. PhET uses it heavily, in some complex situations, and contributed it in (more or less) its current state. I would prefer to see PhET using a proper Piccolo2D release rather than maintaining a private fork, let us know what it will take to make that happen. michael -- Piccolo2D Developers Group: http://groups.google.com/group/piccolo2d-dev?hl=en
[piccolo2d-dev] Re: Release Piccolo2D.Java 1.3
Yes, if 163 were fixed, that would change my vote to +1. 163 is the only know problem with pswing. The other gripes I have with it are internals, and that can certainly wait. On Mar 9, 9:42 am, Michael Heuer heue...@gmail.com wrote: Chris wrote: This release is generally solid, with the exception of the pswing package. Since issue 163 has been reopened, and since I've investigated further, I think that pswing has some new problems that weren't in 1.2 (issue 163 being an example). I can't +1 the release, since my client (PhET) relies heavily on pswing. But I can't -1 the release because I think that fixing pswing and bringing it up to the same standards as the reset of Piccolo2D is going to take considerable time, and I hate to see 1.3 delayed any longer. If issue 163 were fixed, would that change your vote to +1? I wouldn't mind waiting 1.3 until that were the case. If anyone else feels strongly about pswing, then my lack of confidence in its current state may influence your vote. Otherwise, I think PhET should take the lead in fixing improving whatever version of pswing ships with 1.3. PhET uses it heavily, in some complex situations, and contributed it in (more or less) its current state. I would prefer to see PhET using a proper Piccolo2D release rather than maintaining a private fork, let us know what it will take to make that happen. michael -- Piccolo2D Developers Group: http://groups.google.com/group/piccolo2d-dev?hl=en
[piccolo2d-dev] Re: Release Piccolo2D.Java 1.3
--- [ ] +1 I support this release [ ] +0 [X] -0 [ ] -1 I oppose this release because... I'm going to abstain from this vote. This release is generally solid, with the exception of the pswing package. Since issue 163 has been reopened, and since I've investigated further, I think that pswing has some new problems that weren't in 1.2 (issue 163 being an example). I can't +1 the release, since my client (PhET) relies heavily on pswing. But I can't -1 the release because I think that fixing pswing and bringing it up to the same standards as the reset of Piccolo2D is going to take considerable time, and I hate to see 1.3 delayed any longer. If anyone else feels strongly about pswing, then my lack of confidence in its current state may influence your vote. Otherwise, I think PhET should take the lead in fixing improving whatever version of pswing ships with 1.3. PhET uses it heavily, in some complex situations, and contributed it in (more or less) its current state. On Mar 2, 8:57 pm, Michael Heuer heue...@gmail.com wrote: This is a vote for releasing Piccolo2D.Java 1.3 based on release candidate 4 (version 1.3-rc4). Since release candidate 3 new issues 163 and 165 have been fixed and verified. 1.3-rc4 is available from the downloads page: http://code.google.com/p/piccolo2d/downloads/list?can=2q=1.3-rc4 --- [ ] +1 I support this release [ ] +0 [ ] -0 [ ] -1 I oppose this release because... Votes from Piccolo2D project committers are binding, however votes from other contributors and users are welcomed. The vote must receive at least three +1 binding votes and no -1 binding votes. Vote will close at 12:00 GMT Tuesday 09 March 2010. On behalf of the Piccolo2D developers, michael -- Piccolo2D Developers Group: http://groups.google.com/group/piccolo2d-dev?hl=en
[piccolo2d-dev] Re: Release Piccolo2D.Java 1.3
My client (PhET) uses PSwing quite heavily. But I'm not going to -1 in order to get the 163 fix, for 3 reasons: (1) The consensus within PhET is that we're comfortable with patching our 1.3 copy to resolve 163. (2) Imho PSwing needs an internal overhaul, to bring the code up to the standards of Piccolo. Poor name choices is my biggest beef, luckily with private stuff. (3) There are additional PSwing issues that need to be resolved. We have not yet reported these, but are tracking them internally. The big 2 issues are focus traversal and memory leaks when PSwings are removed from the scenegraph. Chris On Feb 26, 4:16 pm, Michael Heuer heue...@gmail.com wrote: Michael Heuer wrote: Vote will close at 12:00 GMT Friday 26 February 2010. I would like to extend to vote deadline until Monday 01 March 2010 to allow for feedback on recently fixed issues 163 and 165. The current vote would pass as it stands and the fixes for those issues would not be released until a later version. michael -- Piccolo2D Developers Group: http://groups.google.com/group/piccolo2d-dev?hl=en
[piccolo2d-dev] Re: Release Piccolo2D.Java 1.3
I'm re-voting, based on our discussion in issue 161. --- [ ] +1 I support this release [ ] +0 [ ] -0 [ X] -1 I oppose this release because (as described in issue 161) full bounds behavior was changed by fixing issue 155. -- Piccolo2D Developers Group: http://groups.google.com/group/piccolo2d-dev?hl=en
[piccolo2d-dev] Re: Release Piccolo2D.Java 1.3 failed
All 1.3-rc1 issues that I reported have been resolved. Many thanks! What is the timeline for 1.3-rc2, and are there any other issues still pending that must be resolved? Chris On Feb 1, 4:09 pm, cmal...@pixelzoom.com cmal...@pixelzoom.com wrote: See issue 160 for the endless series of events issue. All problems we've identified are now in the issues database. 158 and 159 are resolved. 160 is open. Chris On Feb 1, 11:03 am, cmal...@pixelzoom.com cmal...@pixelzoom.com wrote: An update on where we're at with the PSwing issues... Issue 158 was resolved by Allain last week, and fixed a couple of our problems. Issue 159 was opened a few minutes ago, and is related to PSwing transform/picking problems. I am still investigating one additional issue, where a PSwing Component is receiving an endless series of events that toggle its visibility on and off. As soon as I isolate, I'll create an issue. That summarizes all of the problems we've encountered with 1.3-rc1. Chris On Jan 28, 12:29 pm, Michael Heuer heue...@gmail.com wrote: The following people voted on release 1.3rc1: Michael Heuer +1 Allain Lalonde +1 Chris Malley -1 The vote failed since it did not receive at least three +1 binding votes and no -1 binding votes. New issues related to PSwing were discovered and are being addressed. Another release candidate (1.3rc2) will be created when these issues are closed and validated. On behalf of the Piccolo2D developers, michael -- Piccolo2D Developers Group: http://groups.google.com/group/piccolo2d-dev?hl=en
Re: [piccolo2d-dev] Re: Release Piccolo2D.Java 1.3 failed
Chris wrote: All 1.3-rc1 issues that I reported have been resolved. Many thanks! What is the timeline for 1.3-rc2, and are there any other issues still pending that must be resolved? If no new issues are reported this week, I should be able to have 1.3-rc2 ready by Friday. Looking ahead to deploying the 1.3 release to the maven central repository, it would be useful if we as developers could create a KEYS file and discuss how we might be able to sign each others' keys. http://www.apache.org/dev/openpgp.html http://www.apache.org/dev/release-signing.html At a minimum, the release artifacts uploaded to maven central will need to be signed by my key, acting as release manager. michael -- Piccolo2D Developers Group: http://groups.google.com/group/piccolo2d-dev?hl=en
[piccolo2d-dev] Re: Release Piccolo2D.Java 1.3 failed
An update on where we're at with the PSwing issues... Issue 158 was resolved by Allain last week, and fixed a couple of our problems. Issue 159 was opened a few minutes ago, and is related to PSwing transform/picking problems. I am still investigating one additional issue, where a PSwing Component is receiving an endless series of events that toggle its visibility on and off. As soon as I isolate, I'll create an issue. That summarizes all of the problems we've encountered with 1.3-rc1. Chris On Jan 28, 12:29 pm, Michael Heuer heue...@gmail.com wrote: The following people voted on release 1.3rc1: Michael Heuer +1 Allain Lalonde +1 Chris Malley -1 The vote failed since it did not receive at least three +1 binding votes and no -1 binding votes. New issues related to PSwing were discovered and are being addressed. Another release candidate (1.3rc2) will be created when these issues are closed and validated. On behalf of the Piccolo2D developers, michael -- Piccolo2D Developers Group: http://groups.google.com/group/piccolo2d-dev?hl=en
[piccolo2d-dev] Re: Release Piccolo2D.Java 1.3
--- [ ] +1 I support this release [ ] +0 [ ] -0 [X] -1 I oppose this release because... PSwing appears to have some new problems, and PhET (my client) relies heavily on PSwing. Specifically: (1) Visibility issues; there are nodes that should be visible and are not, and vice-versa. (2) Bounds issues; computing PSwing offsets for the purposes of layout is resulting in additional whitespace. I've observed problem (1) in 3 applications, problem (2) in 1 application, on both Windows and Mac platforms. I've been trying to isolate the problems in small test applications, but have been unsuccessful so far. I'm going to continue with that effort a little longer, then have a look at PSwing changes. -- Piccolo2D Developers Group: http://groups.google.com/group/piccolo2d-dev?hl=en
Re: [piccolo2d-dev] Re: Release Piccolo2D.Java 1.3
So far, I have tested 3 of my Piccolo applications, and found only 1 problem in one of the 3 applications. My problem is exhibited with the release candidate 1.3 rc1, but not with our previously used snapshot r390 (circa 9/9/2008), and it also appears to be related to PSwing; a panel with 2 radio buttons is shown in the scene graph, but is noninteractive. This problem by itself wouldn't suggest to me just cause for opposing rc1, unless there is no application-side fix (rather than piccolo or pswing fix) that can resolve or workaround the problem (I don't yet know whether there is an application-side fix). However, based on cmalley's remarks below, I don't think this version should be promoted to a release, and perhaps we'll be able to isolate and resolve my issue as we are working on the related pswing issues. Sam Reid On 1/26/2010 2:35 PM, cmal...@pixelzoom.com wrote: --- [ ] +1 I support this release [ ] +0 [ ] -0 [X] -1 I oppose this release because... PSwing appears to have some new problems, and PhET (my client) relies heavily on PSwing. Specifically: (1) Visibility issues; there are nodes that should be visible and are not, and vice-versa. (2) Bounds issues; computing PSwing offsets for the purposes of layout is resulting in additional whitespace. I've observed problem (1) in 3 applications, problem (2) in 1 application, on both Windows and Mac platforms. I've been trying to isolate the problems in small test applications, but have been unsuccessful so far. I'm going to continue with that effort a little longer, then have a look at PSwing changes. -- Piccolo2D Developers Group: http://groups.google.com/group/piccolo2d-dev?hl=en
