[Pharo-project] Metacello Configuration Pharo1.2 solo sprint Portland, Oregon ... Saturday and Sunday
miguel.coba at gmail.com
Sat Feb 26 05:13:52 CET 2011
El vie, 25-02-2011 a las 17:27 -0800, Dale Henrichs escribió:
> Okay, okay.... I've been trying to keep a half dozen balls in the air
> for the last few weeks and what that means is that I don't give enough
> attention to any of them ...
> Pharo1.2 is close and Metacello is one of the bad boys with the failing
> tests sooooo, I am going to do a solo sprint this weekend:
> 1. a quick release of Metacello with expectedFailures for the tests
> that are failing in Pharo1.2. I have already fixed the tests
> (only test changes are needed), but I now need to port the Gofer
> changes to Squeak and GemStone which is a bigger job. A quick
> surgical set of expectedFailures is the best thing to do here.
> 2. in build #57 of Pharo1.2 I noticed some odd things that I want to
> track down...
> 3. I expect that some of the #57 issues can be revealed with the
> validator, so I am thinking that I will run the validator over
> all of the configurations that are used in Pharo1.2 and go about
> identify and/or fixing the meaningful issues that I find. My main
> goal will be to repair things that are broken, so I will be
> reluctant to make semantic changes without checking with the
> maintainers, but if the problem is really gruesome I will fix it
> and apologize later:)
> 4. I'm not going to spend a lot of time reading/writing email ...
> this will be a sprint not a coffee klatch. With that said, if you
> have a Metacello config issue that is related to Pharo1.2 drop me
> an email with your question/thought but don't be offended if
> I don't respond at all or just GRUNT in reply ... I may even take
> action without letting you know...
> 5. While working on the above task, I will work on the
> MetacelloBrowser adding menu items for the operations that I
> performso it will then be easier for other folks to perform the
> same types of jobs in the future. I may do something funky like
> hide the menu items behind a boolean so as not to pollute the
> menu items with a whole bunch of advanced type operations (that
> will be named poorly:)...
> I realize that I may be closing the barn door after the horse is gone -
> build 57 is yellow afterall - but cleaning up validation issues is long
> I don't plan on fixing Metacello bugs unless I find something that just
> begs to be fixed ... the sprint is mainly about running the validator
> and cleaning up the real nasty bits and track down the Metacello
> anomalies in #57...
More information about the Pharo-project