May 2012 - Pharo-dev - lists.pharo.org

Re: [Pharo-project] class message

by Marcus Denker

On May 26, 2012, at 8:59 PM, Milan Mimica wrote: > On 26 May 2012 20:53, StÃ©phane Ducasse <stephane.ducasse(a)inria.fr> wrote: > > > > What has changed in implementation "class" message in between Pharo 1.3 and 1.4? Is it compiled inline now or something like that? > > > > no, the other way around, now #class is NOT bytecoded, and therefore is managed as a normal message send. > > > > That is much better. Thanks! I like progress even if it breaks old code. > > :) > But we always try to minimize else we would go much faster and btw class should not break older code :) > > It did break Mocketry (http://www.squeaksource.com/Mocketry) > I would also suspect it breaks things like WriteBarrier and Magma and similar stuff performing magic. > normally it should do the opposite: as #class is not special, magic is much easier. Marcus -- Marcus Denker -- http://marcusdenker.de

May 26, 2012

Re: [Pharo-project] class message

by Igor Stasenko

On 26 May 2012 20:59, Milan Mimica <milan.mimica(a)gmail.com> wrote: > On 26 May 2012 20:53, StÃ©phane Ducasse <stephane.ducasse(a)inria.fr> wrote: >> >> > >> > What has changed in implementation "class" message in between Pharo 1.3 >> > and 1.4? Is it compiled inline now or something like that? >> > >> > no, the other way around, now #class is NOT bytecoded, and therefore is >> > managed as a normal message send. >> > >> > That is much better. Thanks! I like progress even if it breaks old code. >> >> :) >> But we always try to minimize else we would go much faster and btw class >> should not break older code :) > > > It did breakÂ Mocketry (http://www.squeaksource.com/Mocketry) > I would also suspect it breaks things like WriteBarrier and Magma and > similar stuff performing magic. but how you can make an omelet without breaking eggs? :) > -- > Milan Mimica > http://sparklet.sf.net -- Best regards, Igor Stasenko.

May 26, 2012

Re: [Pharo-project] class message

by Milan Mimica

On 26 May 2012 20:53, StÃ©phane Ducasse <stephane.ducasse(a)inria.fr> wrote: > > > > What has changed in implementation "class" message in between Pharo 1.3 > and 1.4? Is it compiled inline now or something like that? > > > > no, the other way around, now #class is NOT bytecoded, and therefore is > managed as a normal message send. > > > > That is much better. Thanks! I like progress even if it breaks old code. > > :) > But we always try to minimize else we would go much faster and btw class > should not break older code :) > It did break Mocketry (http://www.squeaksource.com/Mocketry) I would also suspect it breaks things like WriteBarrier and Magma and similar stuff performing magic. -- Milan Mimica http://sparklet.sf.net

May 26, 2012

Re: [Pharo-project] RPackage problem with obsolete classes

by StÃ©phane Ducasse

Hi mariano This afternoon we worked on improving the tests + the integration but this is more complex than it looks. Now I will continue to push RPackage. Stef On May 26, 2012, at 8:01 PM, Mariano Martinez Peck wrote: > > > On Sat, May 26, 2012 at 3:02 PM, Guillermo Polito <guillermopolito(a)gmail.com> wrote: > This test illustrate the problem: > > testMethodPackageFromObsoleteClass > | pack method | > pack := self createNewPackageNamed: 'P1'. > foo := self createNewClassNamed: #FooForTest inPackage: pack. > foo compileSilently: 'bar ^42'. > method := foo>>#bar. > foo removeFromSystem. > self assert: (method packageFromOrganizer: pack organizer)equals: foo package. > > > This seems to happen in non silent tests that > > - create a class/trait for test > - sends an announcement with a compiled method instance (like, method recategorized) > - remove the class/trait because they are for test only :) > - then, since the announcements are handled asychronously, a method from an obsolete category reaches somewhere, and RPackage explodes. > > > This is *exactlly* the problem I faced when running Fuel tests in Pharo 2.0. > I could have time to fix it, nor anyone else. Therefore, what I did is to realy fix Fuel to have ALL tests completly silent. And of course that fixed the problem. > Nice that at least you could write a test :) > > -- > Mariano > http://marianopeck.wordpress.com >

May 26, 2012

Re: [Pharo-project] class message

by Mariano Martinez Peck

On Sat, May 26, 2012 at 8:36 PM, Milan Mimica <milan.mimica(a)gmail.com>wrote: > On 26 May 2012 20:32, Mariano Martinez Peck <marianopeck(a)gmail.com> wrote: > >> >> >> On Sat, May 26, 2012 at 8:31 PM, Milan Mimica <milan.mimica(a)gmail.com>wrote: >> >>> Hello! >>> >>> What has changed in implementation "class" message in between Pharo 1.3 >>> and 1.4? Is it compiled inline now or something like that? >>> >> >> no, the other way around, now #class is NOT bytecoded, and therefore is >> managed as a normal message send. >> > > That is much better. Thanks! I like progress even if it breaks old code. > breaks? can you give us an example? beause it has to be something quite hacky or low level, has't it? > > > -- > Milan Mimica > http://sparklet.sf.net > -- Mariano http://marianopeck.wordpress.com

May 26, 2012

Re: [Pharo-project] class message

by StÃ©phane Ducasse

> > What has changed in implementation "class" message in between Pharo 1.3 and 1.4? Is it compiled inline now or something like that? > > no, the other way around, now #class is NOT bytecoded, and therefore is managed as a normal message send. > > That is much better. Thanks! I like progress even if it breaks old code. :) But we always try to minimize else we would go much faster and btw class should not break older code :) > > > -- > Milan Mimica > http://sparklet.sf.net

May 26, 2012

Re: [Pharo-project] class message

by Milan Mimica

On 26 May 2012 20:32, Mariano Martinez Peck <marianopeck(a)gmail.com> wrote: > > > On Sat, May 26, 2012 at 8:31 PM, Milan Mimica <milan.mimica(a)gmail.com>wrote: > >> Hello! >> >> What has changed in implementation "class" message in between Pharo 1.3 >> and 1.4? Is it compiled inline now or something like that? >> > > no, the other way around, now #class is NOT bytecoded, and therefore is > managed as a normal message send. > That is much better. Thanks! I like progress even if it breaks old code. -- Milan Mimica http://sparklet.sf.net

May 26, 2012

Re: [Pharo-project] class message

by Mariano Martinez Peck

On Sat, May 26, 2012 at 8:31 PM, Milan Mimica <milan.mimica(a)gmail.com>wrote: > Hello! > > What has changed in implementation "class" message in between Pharo 1.3 > and 1.4? Is it compiled inline now or something like that? > no, the other way around, now #class is NOT bytecoded, and therefore is managed as a normal message send. > > > -- > Milan Mimica > http://sparklet.sf.net > -- Mariano http://marianopeck.wordpress.com

May 26, 2012

[Pharo-project] class message

by Milan Mimica

Hello! What has changed in implementation "class" message in between Pharo 1.3 and 1.4? Is it compiled inline now or something like that? -- Milan Mimica http://sparklet.sf.net

May 26, 2012

Re: [Pharo-project] RPackage problem with obsolete classes

by Mariano Martinez Peck

On Sat, May 26, 2012 at 3:02 PM, Guillermo Polito <guillermopolito(a)gmail.com > wrote: > This test illustrate the problem: > > testMethodPackageFromObsoleteClass > | pack method | > pack := self createNewPackageNamed: 'P1'. > foo := self createNewClassNamed: #FooForTest inPackage: pack. > foo compileSilently: 'bar ^42'. > method := foo>>#bar. > foo removeFromSystem. > self assert: (method packageFromOrganizer: pack organizer)equals: foo > package. > > > This seems to happen in non silent tests that > > - create a class/trait for test > - sends an announcement with a compiled method instance (like, method > recategorized) > - remove the class/trait because they are for test only :) > - then, since the announcements are handled asychronously, a method from > an obsolete category reaches somewhere, and RPackage explodes. > > This is *exactlly* the problem I faced when running Fuel tests in Pharo 2.0. I could have time to fix it, nor anyone else. Therefore, what I did is to realy fix Fuel to have ALL tests completly silent. And of course that fixed the problem. Nice that at least you could write a test :) -- Mariano http://marianopeck.wordpress.com

May 26, 2012