June 2008 - Pharo-dev - lists.pharo.org

Re: [Pharo-project] [ANN] new update

by Norbert Hartl

On Wed, 2008-06-25 at 23:50 +0200, Adrian Lienhard wrote: > I think closed is better. Everyone knows what it means. Ok, I changed the issues to "closed" Norbert > Adrian > > On Jun 25, 2008, at 23:44 , Norbert Hartl wrote: > > > On Wed, 2008-06-25 at 23:34 +0200, Adrian Lienhard wrote: > >> Thanks, Stef! > >> Looks good. > >> > >> In the bug tracker, the status of 3 harvested changes is > >> "statusVerified". This does not seem to be a valid state. Shouldn't > >> it > >> be "Closed" instead?! > >> > > Yes, I created the "Closed" status exactly for this. I don't think > > there is much to do after it is harvested therefor closed. Would > > it be better to call it "harvested" ? > > > > Norbert > >> Adrian > >> > >> On Jun 25, 2008, at 22:12 , StÃ©phane Ducasse wrote: > >> > >>> here is a couple of fixed. > >>> BTW yesterday I issued another image. > >>> > >>> Stef > >>> > >>> > >>> > >>> Couple of changes to make more tests work plus fix of finalization > >>> logic: > >>> - changed WeakRegistry to use WeakIdentityKeyDictionary (http://bugs.squeak.org/view.php?id=6347 > >>> ) > >>> - FinalizerTest reverted to real test case > >>> - added missing class variables to DigitalSigntureAlgorithm > >>> - Fix issue #13: TraitsTests pollutes ProtoObject subclasses (Mantis > >>> 7090) > >>> - make BitmapStreamTest clean up the file > >>> bitmapStreamTest.ref.extSeg > >>> ------------------------------------------------------ > >>> update10043 > >>> MVC cleaning step 3 > >>> ------------------------------------------------------ > >>> update10044 > >>> cleaning unimplemented calls > >>> ------------------------------------------------------ > >>> update10045 > >>> Some more cleaning made by Norbert > >>> ------------------------------------------------------ > >>> update10046 > >>> KernelTests-Numbers -- Nan and hex tests broken > >>> > >>> Collection methods #select: and #collect: have no test > >>> > >>> MC fixes > >>> > >>> Remove testAllMethodsWithSourceString > >>> > >>> > >>> > >>> _______________________________________________ > >>> Pharo-project mailing list > >>> Pharo-project(a)lists.gforge.inria.fr > >>> http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project > >> > >> > >> _______________________________________________ > >> Pharo-project mailing list > >> Pharo-project(a)lists.gforge.inria.fr > >> http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project > > > > > > _______________________________________________ > > Pharo-project mailing list > > Pharo-project(a)lists.gforge.inria.fr > > http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project > > > _______________________________________________ > Pharo-project mailing list > Pharo-project(a)lists.gforge.inria.fr > http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project

June 25, 2008

[Pharo-project] ReleaseTest>>testUnimplementedCalls

by Norbert Hartl

I thought once more about this test. I'm reluctant to have this test removed. On the other hand is Adrien right. He states that a lot of test classes are doing calls to methods they created while the test runs. One solution would be to change the test to use perform: to invoke those methods. But that would making tests a bit tedious. So this is not my preferred solution. I tweaked SystemNavigation>>allUnimplementedNonPrimitiveCalls to allUnimplementedNonPrimitiveCallsX "Answer an Array of each message that is sent by an expression in a method but is not implemented by any object in the system." | all meth dict | dict := Dictionary new. all := self systemNavigation allImplementedMessages. self systemNavigation allBehaviorsDo: [:cl | cl selectorsDo: [:sel | meth := cl compiledMethodAt: sel. meth primitive = 0 ifTrue: [ meth messages do: [:m | (all includes: m) ifFalse: [ ((dict at: cl ifAbsentPut: [ Dictionary new ]) at: sel ifAbsentPut: [ OrderedCollection new]) add: m ] ] ] ] ]. ^ dict So the test would be like (SystemNavigation default allUnimplementedNonPrimitiveCallsX associationsSelect: [:a| (a key allSuperclasses includes: TestCase) not ]) isEmpty Would this or something similar be useful or is this just a lot of noise without much of a benefit? Norbert

June 25, 2008

Re: [Pharo-project] milestone 1

by Adrian Lienhard

I agree. For Milestone 1 I suggest: - carefully check list of 3.10 (and maybe Croquet) fixes and integrate the ones that make sense (notice, on our wiki there exist pages listing 3.10.* and Croquet fixes). - Check for important fixes posted on Mantis that were not integrated in 3.10 or were posted more recently. - Make all remaining tests pass or mark them as expected failures (only 2 left!) Anything else that would be valuable and does not take too much effort? This version would then be a good candidate for a first release. It would be good if we have a first release as this allows people to start moving to Pharo. And its also good that we do not have fundamental changes that would require code migration. Adrian PS: Stef, I do not understand what you mean by "putting Pharo in the flap" On Jun 25, 2008, at 09:11 , StÃ©phane Ducasse wrote: > for milestone 1 > - it would be good to have a Pharo in the flap (if any flap) and any > other better looking lesssqueakish look :L) > if one of you has time :) > > > - when do we want to finish. It would be good to have milestone 1 > out for esug or before. > - I would really like to see the items of 3.10 examined. > > for milestone 2 > > - better building process > > - something that I would like us to do is to have a task force > writing some good class comments. > If you can think about that it would be good. > > Purpose of the class > Main methods > Main collaborators > Example > Internal representation > > Stef > > _______________________________________________ > Pharo-project mailing list > Pharo-project(a)lists.gforge.inria.fr > http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project

June 25, 2008

Re: [Pharo-project] [ANN] new update

by Adrian Lienhard

I think closed is better. Everyone knows what it means. Adrian On Jun 25, 2008, at 23:44 , Norbert Hartl wrote: > On Wed, 2008-06-25 at 23:34 +0200, Adrian Lienhard wrote: >> Thanks, Stef! >> Looks good. >> >> In the bug tracker, the status of 3 harvested changes is >> "statusVerified". This does not seem to be a valid state. Shouldn't >> it >> be "Closed" instead?! >> > Yes, I created the "Closed" status exactly for this. I don't think > there is much to do after it is harvested therefor closed. Would > it be better to call it "harvested" ? > > Norbert >> Adrian >> >> On Jun 25, 2008, at 22:12 , StÃ©phane Ducasse wrote: >> >>> here is a couple of fixed. >>> BTW yesterday I issued another image. >>> >>> Stef >>> >>> >>> >>> Couple of changes to make more tests work plus fix of finalization >>> logic: >>> - changed WeakRegistry to use WeakIdentityKeyDictionary (http://bugs.squeak.org/view.php?id=6347 >>> ) >>> - FinalizerTest reverted to real test case >>> - added missing class variables to DigitalSigntureAlgorithm >>> - Fix issue #13: TraitsTests pollutes ProtoObject subclasses (Mantis >>> 7090) >>> - make BitmapStreamTest clean up the file >>> bitmapStreamTest.ref.extSeg >>> ------------------------------------------------------ >>> update10043 >>> MVC cleaning step 3 >>> ------------------------------------------------------ >>> update10044 >>> cleaning unimplemented calls >>> ------------------------------------------------------ >>> update10045 >>> Some more cleaning made by Norbert >>> ------------------------------------------------------ >>> update10046 >>> KernelTests-Numbers -- Nan and hex tests broken >>> >>> Collection methods #select: and #collect: have no test >>> >>> MC fixes >>> >>> Remove testAllMethodsWithSourceString >>> >>> >>> >>> _______________________________________________ >>> Pharo-project mailing list >>> Pharo-project(a)lists.gforge.inria.fr >>> http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project >> >> >> _______________________________________________ >> Pharo-project mailing list >> Pharo-project(a)lists.gforge.inria.fr >> http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project > > > _______________________________________________ > Pharo-project mailing list > Pharo-project(a)lists.gforge.inria.fr > http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project

June 25, 2008

Re: [Pharo-project] [ANN] new update

by Norbert Hartl

On Wed, 2008-06-25 at 23:34 +0200, Adrian Lienhard wrote: > Thanks, Stef! > Looks good. > > In the bug tracker, the status of 3 harvested changes is > "statusVerified". This does not seem to be a valid state. Shouldn't it > be "Closed" instead?! > Yes, I created the "Closed" status exactly for this. I don't think there is much to do after it is harvested therefor closed. Would it be better to call it "harvested" ? Norbert > Adrian > > On Jun 25, 2008, at 22:12 , StÃ©phane Ducasse wrote: > > > here is a couple of fixed. > > BTW yesterday I issued another image. > > > > Stef > > > > > > > > Couple of changes to make more tests work plus fix of finalization > > logic: > > - changed WeakRegistry to use WeakIdentityKeyDictionary (http://bugs.squeak.org/view.php?id=6347 > > ) > > - FinalizerTest reverted to real test case > > - added missing class variables to DigitalSigntureAlgorithm > > - Fix issue #13: TraitsTests pollutes ProtoObject subclasses (Mantis > > 7090) > > - make BitmapStreamTest clean up the file bitmapStreamTest.ref.extSeg > > ------------------------------------------------------ > > update10043 > > MVC cleaning step 3 > > ------------------------------------------------------ > > update10044 > > cleaning unimplemented calls > > ------------------------------------------------------ > > update10045 > > Some more cleaning made by Norbert > > ------------------------------------------------------ > > update10046 > > KernelTests-Numbers -- Nan and hex tests broken > > > > Collection methods #select: and #collect: have no test > > > > MC fixes > > > > Remove testAllMethodsWithSourceString > > > > > > > > _______________________________________________ > > Pharo-project mailing list > > Pharo-project(a)lists.gforge.inria.fr > > http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project > > > _______________________________________________ > Pharo-project mailing list > Pharo-project(a)lists.gforge.inria.fr > http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project

June 25, 2008

Re: [Pharo-project] [ANN] new update

by Adrian Lienhard

Thanks, Stef! Looks good. In the bug tracker, the status of 3 harvested changes is "statusVerified". This does not seem to be a valid state. Shouldn't it be "Closed" instead?! Adrian On Jun 25, 2008, at 22:12 , StÃ©phane Ducasse wrote: > here is a couple of fixed. > BTW yesterday I issued another image. > > Stef > > > > Couple of changes to make more tests work plus fix of finalization > logic: > - changed WeakRegistry to use WeakIdentityKeyDictionary (http://bugs.squeak.org/view.php?id=6347 > ) > - FinalizerTest reverted to real test case > - added missing class variables to DigitalSigntureAlgorithm > - Fix issue #13: TraitsTests pollutes ProtoObject subclasses (Mantis > 7090) > - make BitmapStreamTest clean up the file bitmapStreamTest.ref.extSeg > ------------------------------------------------------ > update10043 > MVC cleaning step 3 > ------------------------------------------------------ > update10044 > cleaning unimplemented calls > ------------------------------------------------------ > update10045 > Some more cleaning made by Norbert > ------------------------------------------------------ > update10046 > KernelTests-Numbers -- Nan and hex tests broken > > Collection methods #select: and #collect: have no test > > MC fixes > > Remove testAllMethodsWithSourceString > > > > _______________________________________________ > Pharo-project mailing list > Pharo-project(a)lists.gforge.inria.fr > http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project

June 25, 2008

[Pharo-project] [ANN] new update

by StÃ©phane Ducasse

here is a couple of fixed. BTW yesterday I issued another image. Stef Couple of changes to make more tests work plus fix of finalization logic: - changed WeakRegistry to use WeakIdentityKeyDictionary (http://bugs.squeak.org/view.php?id=6347 ) - FinalizerTest reverted to real test case - added missing class variables to DigitalSigntureAlgorithm - Fix issue #13: TraitsTests pollutes ProtoObject subclasses (Mantis 7090) - make BitmapStreamTest clean up the file bitmapStreamTest.ref.extSeg ------------------------------------------------------ update10043 MVC cleaning step 3 ------------------------------------------------------ update10044 cleaning unimplemented calls ------------------------------------------------------ update10045 Some more cleaning made by Norbert ------------------------------------------------------ update10046 KernelTests-Numbers -- Nan and hex tests broken Collection methods #select: and #collect: have no test MC fixes Remove testAllMethodsWithSourceString

June 25, 2008

[Pharo-project] Standing against tempest by example :)

by StÃ©phane Ducasse

Begin forwarded message: > From: "Samuel MORELLO" <samuel.morello(a)ouvanous.com> > Date: June 25, 2008 2:09:06 PM CEDT > To: "StÃ©phane Ducasse" <stephane.ducasse(a)inria.fr> > > http://www.metacafe.com/watch/yt-oNpH5tx4Pu8/lighthouses_and_waves_during_v… > > -- > Samuel > +33 (0)6 20 65 22 99 >

June 25, 2008

Re: [Pharo-project] can you check the last update

by Damien Cassou

On Wed, Jun 25, 2008 at 1:00 PM, Norbert Hartl <norbert(a)hartl.name> wrote: > On Wed, 2008-06-25 at 12:52 +0200, Damien Cassou wrote: >> On Wed, Jun 25, 2008 at 11:05 AM, Norbert Hartl <norbert(a)hartl.name> wrote: >> > On Wed, 2008-06-25 at 10:55 +0200, Damien Cassou wrote: >> >> > On Wed, Jun 25, 2008 at 10:50 AM, Norbert Hartl <norbert(a)hartl.name> wrote: >> >> >> - removed unnecessary columns Difficulty and Priority from Issue >> >> >> >> Should I remove them completely from the issue tracking system? >> >> >> >> Fixed = Developer made requested changes, QA should verify >> >> Verified = QA has verified that the fix worked >> >> Invalid = This was not a valid issue report >> >> Duplicate = This report duplicates an existing issue >> >> WontFix = We decided to not take action on this issue >> >> Closed = The fix has been treated and issue is closed >> >> >> > >> > I don't understan. I removed the columns in the view as >> > they are only disturbing. What do you mean by showing the >> > list of statusses? >> >> I meant: >> >> Type-Defect = Report of a software defect >> Type-Enhancement = Request for enhancement >> Type-Task = Work item that doesn't change the code or docs >> Type-Patch = Source code patch for review >> Type-Other = Some other kind of issue >> Priority-High = Strongly want to resolve in the specified milestone >> Priority-Medium = Normal priority >> Priority-Low = Might slip to later milestone >> Difficulty-VeryEasy = Anybody can fix that in a few seconds >> Difficulty-Easy = Only little knowledge and time required >> Difficulty-Medium = Knowledge and time required >> Difficulty-Hard = Requires more than one expert >> > Why you want to remove them? Now they are only visible in the admin area > where to don't disturb anyone. But do as you think is best. They can still be set at the bottom of bug raeporats. Just click in the text-fields -- Damien Cassou Peter von der AhÃ©: Â«I'm beginning to see why Gilad wished us good luckÂ». (http://blogs.sun.com/ahe/entry/override_snafu)

June 25, 2008

Re: [Pharo-project] can you check the last update

by StÃ©phane Ducasse

Excellent!!! I like one click todo :) >>> >>> We could do a SLICE for every fix so you could see what needs >>> to be updated. But then it is hard to be sure someone reviewed >>> it. >>> >>> Maybe the best solution is to use the issue tracker on google. >>> That's the purpose it serves. If someone has an update he can >>> open a ticket with type "patch" and status "Fixed". If another >>> one has reviewed it he can change the status to "verified". To >>> see what is ready for update you just filter the issues by >>> status "Verified" >> >> Yes we should do that >> > Ok, I changed a few things: > > - removed unnecessary columns Difficulty and Priority from Issue > list view > - added status closed so it either goes > -> defect -> fix -> verified -> close > or > -> patch -> fix -> verified -> close > - treated a few issues to have correct status > > So Stef, all you have to do is to save this link > > http://code.google.com/p/pharo/issues/list?q=status%3AVerified&can=1 > > and click it if you want to know what to do ;) > > I hope you like it. > > Norbert >>> >>> >>> Norbert >>>> Stef >>>> >>>> On Jun 25, 2008, at 9:45 AM, Norbert Hartl wrote: >>>> >>>>> Are you going to harvest the rest of the packages, too? >>>>> >>>>> KernelTests-al.67 >>>>> Monticello-al.319 >>>>> >>>>> are fixing all but 2 remaining tests. >>>>> Norbert >>>>> >>>>> On Wed, 2008-06-25 at 09:00 +0200, StÃ©phane Ducasse wrote: >>>>>> this is strange when I do a merge with SLICE-randomRemoval I >>>>>> get an >>>>>> emtpy list. >>>>>> Were SLICE-UnimplementedCallsFix1-NorbertHartl.1 and >>>>>> randomRemoval >>>>>> the >>>>>> same? >>>>>> Should I close SLICE-randomRemoval? >>>>> >>>>>> Stef >>>>>> >>>>>> On Jun 25, 2008, at 8:52 AM, StÃ©phane Ducasse wrote: >>>>>> >>>>>>> Hi all >>>>>>> >>>>>>> I got some funny unpublished packages and I'm a bit confused. So >>>>>>> could you check what I did. >>>>>>> I was harvesting MVC removal + norbert cleaning SLICE- >>>>>>> UnimplementedCallsFix1-NorbertHartl.1 >>>>>>> >>>>>>> I have to find some time (argh) to improve the process and the >>>>>>> tools. because this is far too much manual. >>>>>>> >>>>>>> Stef >>>>>>> >>>>>>> _______________________________________________ >>>>>>> Pharo-project mailing list >>>>>>> Pharo-project(a)lists.gforge.inria.fr >>>>>>> http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo- >>>>>>> project >>>>>>> >>>>>> >>>>>> >>>>>> _______________________________________________ >>>>>> Pharo-project mailing list >>>>>> Pharo-project(a)lists.gforge.inria.fr >>>>>> http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project >>>>> >>>>> >>>>> _______________________________________________ >>>>> Pharo-project mailing list >>>>> Pharo-project(a)lists.gforge.inria.fr >>>>> http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo- >>>>> project >>>>> >>>> >>>> >>>> _______________________________________________ >>>> Pharo-project mailing list >>>> Pharo-project(a)lists.gforge.inria.fr >>>> http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project >>> >>> >>> _______________________________________________ >>> Pharo-project mailing list >>> Pharo-project(a)lists.gforge.inria.fr >>> http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project >>> >> >> >> _______________________________________________ >> Pharo-project mailing list >> Pharo-project(a)lists.gforge.inria.fr >> http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project > > > _______________________________________________ > Pharo-project mailing list > Pharo-project(a)lists.gforge.inria.fr > http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project >

June 25, 2008