May 2011 - Pharo-dev - lists.pharo.org

[Pharo-project] OB variable references browser bug [fix]

by Igor Stasenko

There is a DNU in OB, when you do following: - go to some class - click its 'variables' , to open variable references browser - select any variable and go to one of the methods which using it - modify this method to no longer reference that variable and accept it. And at that moment you got DNU. The possible fix is attached. But note, that this method actually invoked as a result of handing another error. I don't know, maybe it would be better to prevent an original error from occur, rather than putting a workaround like this. I'm not an expert in OB, so i just fixed so it kinda works, at least its not triggers DNU anymore. To see the original error, simply put 'self halt' in the method i attached, and then look at the stack in debugger. -- Best regards, Igor Stasenko AKA sig.

May 9, 2011

Re: [Pharo-project] Problem with CompiledMethodTrailer

by Igor Stasenko

the one to the left seems like broken. it not ends with return self. On 9 May 2011 11:22, Mariano Martinez Peck <marianopeck(a)gmail.com> wrote: > > > On Mon, May 9, 2011 at 10:53 AM, Mariano Martinez Peck > <marianopeck(a)gmail.com> wrote: >> >> >> On Mon, May 9, 2011 at 10:27 AM, Nicolas Cellier >> <nicolas.cellier.aka.nice(a)gmail.com> wrote: >>> >>> Hi Mariano, >>> you should try these two things before repeating yourself: >>> >>> (CompiledMethod allInstances select: Â [:each | each trailer kind = >>> #VarLengthSourcePointer] ) first trailer kind. >>> >>> | temp | >>> temp := (CompiledMethod allInstances select: Â [:each | each trailer >>> kind = #VarLengthSourcePointer] ) first. >>> temp == (temp methodClass >> temp selector) >>> >> >> Thanks Nicolas. Yes,Â in fact all of you were right. They are not the same >> CM instance. They are different. >> It is wierd that they are only present in Pharo1.3: >> >> >> ((CompiledMethod allInstances select:Â [:each | each trailer kind = >> #VarLengthSourcePointer] ) ) size >> ->>>Â Â 11 >> >> And in Pharo 1.2, cero and and 1.1.1, 2. >> >> I am trying to inspect pointers to them to see who is pointing to >> them...but I cannot find anything :( >> >> I will continue.... >> > > More data: there are only these kind of compiled methods in the dev image. > Not in core. > > In addition, if I do: ScriptLoader new cleanUpForRelease, then inspect > ((CompiledMethod allInstances select:Â [:each | each trailer kind = > #VarLengthSourcePointer] ) ) > > and then I click on the first element in the inspector.... VM CRASH!!!Â with > both, InterpreterVM and CogVM. > > I can see the difference in the trailer of the CMs (I attach screenshot), > but still cannot find who is pointing to that weird instances :( > > grrr... thanks for the help > > > >> >> thanks >> >> >> >>> It may be possible that CompiledMethod>>printString is lying to you... >>> >>> Nicolas >>> >>> 2011/5/9 Mariano Martinez Peck <marianopeck(a)gmail.com>: >>> > >>> > >>> > On Mon, May 9, 2011 at 6:14 AM, Igor Stasenko <siguctua(a)gmail.com> >>> > wrote: >>> >> >>> >> On 9 May 2011 00:52, Mariano Martinez Peck <marianopeck(a)gmail.com> >>> >> wrote: >>> >> > Hi guys, Maybe it is too late and I am not seeing something >>> >> > obvious...so >>> >> > another look is really appreciated. If I do: >>> >> > >>> >> > (CompiledMethod allInstances select:Â [:each | each trailer kind = >>> >> > #VarLengthSourcePointer] ) first >>> >> > ---> (CodeLoader class>>#initialize "a CompiledMethod(311951360)") >>> >> > >>> >> > But then: >>> >> > >>> >> > (CodeLoader class >> #initialize) trailer kind >>> >> > ----> #SourcePointer >>> >> > >>> >> > WHYYY??Â why it is answering #SourcePointer instead of >>> >> > #VarLengthSourcePointerÂ ?? >>> >> > >>> >> > grrr how can that be possible???Â It happens the same to each of the >>> >> > CompiledMethods I select as #VarLengthSourcePointer.Â When I then >>> >> > ask >>> >> > the >>> >> > #kindÂ they always answer #SourcePointerÂ instead of >>> >> > #VarLengthSourcePointer >>> >> > >>> >> is it same compiled methods? >>> >> Because it is absolutely normal if they using different source >>> >> pointer. >>> >> >>> > >>> > I didn't understand your question. Yes, it is exactly the same >>> > CompiledMethod. >>> > Look, if I do: >>> > >>> > (CompiledMethod allInstances select:Â [:each | each trailer kind = >>> > #VarLengthSourcePointer] ) first >>> > >>> > I get >>> > ---> (CodeLoader class>>#initialize "a CompiledMethod(311951360)") >>> > >>> > So...it means that such CompiledMethod answers #VarLengthSourcePointer >>> > when >>> > send "trailer kind". Ok? >>> > >>> > But immediatly after, if I ask the trailer kind to the SAME compiled >>> > method: >>> > >>> > (CodeLoader class >> #initialize) trailer kind >>> > >>> > I get >>> > ----> #SourcePointer >>> > >>> > WHYYY??Â why it is answering #SourcePointer instead of >>> > #VarLengthSourcePointerÂ ?? >>> > >>> >> >>> >> > thanks in advance, >>> >> > >>> >> > -- >>> >> > Mariano >>> >> > http://marianopeck.wordpress.com >>> >> > >>> >> > >>> >> >>> >> >>> >> >>> >> -- >>> >> Best regards, >>> >> Igor Stasenko AKA sig. >>> >> >>> > >>> > >>> > >>> > -- >>> > Mariano >>> > http://marianopeck.wordpress.com >>> > >>> > >>> >> >> >> >> -- >> Mariano >> http://marianopeck.wordpress.com >> > > > > -- > Mariano > http://marianopeck.wordpress.com > > -- Best regards, Igor Stasenko AKA sig.

May 9, 2011

Re: [Pharo-project] How to load Nautilus?

by Benjamin

On May 9, 2011, at 11:51 AM, Mariano Martinez Peck wrote: > > > On Mon, May 9, 2011 at 11:46 AM, Benjamin <benjamin.vanryseghem.pharo(a)gmail.com> wrote: > > On May 7, 2011, at 10:46 PM, Mariano Martinez Peck wrote: > >> I took the latest PharoCore 1.3 and evaluated: >> >> Gofer new >> squeaksource: 'Nautilus'; >> package: 'ConfigurationOfNautilus'; >> load. >> >> >> ConfigurationOfNautilus project lastVersion load >> >> >> But it says >> >> This package depends on the following classes: >> SystemAnnouncer >> You must resolve these dependencies before you will be able to load these definitions: >> GroupAnnouncer >> >> >> Select Proceed to continue, or close this window to cancel the operation. >> >> >> >> Then after in the load, I have a DNU in MCWorkingCopy >> modified: aBooleanm, because MonticelloAnnouncer is nil (not defined) >> >> So....how I am supposed to load Nautilus? > > Nautilus is based on RPackage so: > > - dl PharoCore 1.3 from Hudson > - load RPackage 3.0 > - load Nautilus (stable or default, should be the same) > > Here it works > > > But the idea is that it works outside your pc ;) > > So...if Nautilus is based on RPackage, then you must add it to ConfigurationOfNautilus as a depdeny. Otherwise, I cannot guess that you need RPackage. If fact, I didn'tknow until now. > I don't know where RPackage is, nor which group should I load. > > :) Yep, I will do that :) Ben > > > >> >> And if I want the integration with Refactoring Browser, OCompletion, Shout, etc, what should I do? load those packages and magically Nautilus will work ? > > OCompletion and Shout normally impact only the textMorph. > I haven't test OCompletion (I do not use it, but I will do it right now), but for Shout, load the Shout package, and no, it's not magic, but Nautilus works. > > ok cool > > > For refactoring browser, since Nautilus is based on BrowserEnvironment, it should work too. But I have to check and steal some menu's entries from OB. > > > Cool. I think this is really important :) > >> What happens if I load Nautilus in a dev image? > > No idea, let's try :) > > > Ben > > > > -- > Mariano > http://marianopeck.wordpress.com >

May 9, 2011

[Pharo-project] There are 443 open Bug Reports

by Marcus Denker

There are 443 open Bug Reports: http://code.google.com/p/pharo/issues/list It would *really* be nice if more people would look at them. Especially, if *you* reported a bug in the past, why don't *you* check the status of your own report? - Maybe it has already been fixed? - Maybe it's blocked on some triviality? - Maybe it has been fixed and the fix is ready but just drowned in 443 open reports? - Maybe the report has a strange tilte? - Or is formulated strangely? - Or maybe after some discussion, it's not clear just what the next action is (even though, yes, after reading the discussion and thinking for 11 minutes it's obvious?) Marcus -- Marcus Denker -- http://www.marcusdenker.de INRIA Lille -- Nord Europe. Team RMoD.

May 9, 2011

Re: [Pharo-project] How to load Nautilus?

by Marcus Denker

On May 9, 2011, at 11:54 AM, Tudor Girba wrote: >> >> >> - dl PharoCore 1.3 from Hudson >> - load RPackage 3.0 >> - load Nautilus (stable or default, should be the same) > > You should add ConfigurationOfRPackage to ConfigurationOfNautilus so that all these load seamlessly. > And most important: there should be Hudson project. How can people live without that? Marcus -- Marcus Denker -- http://www.marcusdenker.de INRIA Lille -- Nord Europe. Team RMoD.

May 9, 2011

Re: [Pharo-project] How to load Nautilus?

by Tudor Girba

Hi Ben, On 9 May 2011, at 11:46, Benjamin wrote: > > On May 7, 2011, at 10:46 PM, Mariano Martinez Peck wrote: > >> I took the latest PharoCore 1.3 and evaluated: >> >> Gofer new >> squeaksource: 'Nautilus'; >> package: 'ConfigurationOfNautilus'; >> load. >> >> >> ConfigurationOfNautilus project lastVersion load >> >> >> But it says >> >> This package depends on the following classes: >> SystemAnnouncer >> You must resolve these dependencies before you will be able to load these definitions: >> GroupAnnouncer >> >> >> Select Proceed to continue, or close this window to cancel the operation. >> >> >> >> Then after in the load, I have a DNU in MCWorkingCopy >> modified: aBooleanm, because MonticelloAnnouncer is nil (not defined) >> >> So....how I am supposed to load Nautilus? > > Nautilus is based on RPackage so: > > - dl PharoCore 1.3 from Hudson > - load RPackage 3.0 > - load Nautilus (stable or default, should be the same) You should add ConfigurationOfRPackage to ConfigurationOfNautilus so that all these load seamlessly. Cheers, Doru > Here it works > > >> >> And if I want the integration with Refactoring Browser, OCompletion, Shout, etc, what should I do? load those packages and magically Nautilus will work ? > > OCompletion and Shout normally impact only the textMorph. > I haven't test OCompletion (I do not use it, but I will do it right now), but for Shout, load the Shout package, and no, it's not magic, but Nautilus works. > > For refactoring browser, since Nautilus is based on BrowserEnvironment, it should work too. But I have to check and steal some menu's entries from OB. > >> What happens if I load Nautilus in a dev image? > > No idea, let's try :) > > > Ben -- www.tudorgirba.com "What we can governs what we wish."

May 9, 2011

Re: [Pharo-project] How to load Nautilus?

by Mariano Martinez Peck

On Mon, May 9, 2011 at 11:46 AM, Benjamin < benjamin.vanryseghem.pharo(a)gmail.com> wrote: > > On May 7, 2011, at 10:46 PM, Mariano Martinez Peck wrote: > > I took the latest PharoCore 1.3 and evaluated: > > Gofer new > squeaksource: 'Nautilus'; > package: 'ConfigurationOfNautilus'; > load. > > > ConfigurationOfNautilus project lastVersion load > > > But it says > > This package depends on the following classes: > SystemAnnouncer > You must resolve these dependencies before you will be able to load these > definitions: > GroupAnnouncer > > > Select Proceed to continue, or close this window to cancel the operation. > > > > Then after in the load, I have a DNU in MCWorkingCopy >> modified: > aBooleanm, because MonticelloAnnouncer is nil (not defined) > > So....how I am supposed to load Nautilus? > > > Nautilus is based on RPackage so: > > - dl PharoCore 1.3 from Hudson > - load RPackage 3.0 > - load Nautilus (stable or default, should be the same) > > Here it works > > But the idea is that it works outside your pc ;) So...if Nautilus is based on RPackage, then you must add it to ConfigurationOfNautilus as a depdeny. Otherwise, I cannot guess that you need RPackage. If fact, I didn'tknow until now. I don't know where RPackage is, nor which group should I load. :) > > > And if I want the integration with Refactoring Browser, OCompletion, > Shout, etc, what should I do? load those packages and magically Nautilus > will work ? > > > OCompletion and Shout normally impact only the textMorph. > I haven't test OCompletion (I do not use it, but I will do it right now), > but for Shout, load the Shout package, and no, it's not magic, but Nautilus > works. > ok cool > > For refactoring browser, since Nautilus is based on BrowserEnvironment, it > should work too. But I have to check and steal some menu's entries from OB. > > Cool. I think this is really important :) > What happens if I load Nautilus in a dev image? > > > No idea, let's try :) > > > Ben > -- Mariano http://marianopeck.wordpress.com

May 9, 2011

Re: [Pharo-project] How to load Nautilus?

by Benjamin

On May 7, 2011, at 10:46 PM, Mariano Martinez Peck wrote: > I took the latest PharoCore 1.3 and evaluated: > > Gofer new > squeaksource: 'Nautilus'; > package: 'ConfigurationOfNautilus'; > load. > > > ConfigurationOfNautilus project lastVersion load > > > But it says > > This package depends on the following classes: > SystemAnnouncer > You must resolve these dependencies before you will be able to load these definitions: > GroupAnnouncer > > > Select Proceed to continue, or close this window to cancel the operation. > > > > Then after in the load, I have a DNU in MCWorkingCopy >> modified: aBooleanm, because MonticelloAnnouncer is nil (not defined) > > So....how I am supposed to load Nautilus? Nautilus is based on RPackage so: - dl PharoCore 1.3 from Hudson - load RPackage 3.0 - load Nautilus (stable or default, should be the same) Here it works > > And if I want the integration with Refactoring Browser, OCompletion, Shout, etc, what should I do? load those packages and magically Nautilus will work ? OCompletion and Shout normally impact only the textMorph. I haven't test OCompletion (I do not use it, but I will do it right now), but for Shout, load the Shout package, and no, it's not magic, but Nautilus works. For refactoring browser, since Nautilus is based on BrowserEnvironment, it should work too. But I have to check and steal some menu's entries from OB. > What happens if I load Nautilus in a dev image? No idea, let's try :) Ben

May 9, 2011

[Pharo-project] [Hudson] Updates

by Marcus Denker

Hudson Updates ============= -> I fixed the build of 1.2 Full. For the Future, I would like to establish a new rule that a failing project is removed (with all its dependencies) from Hudson after 2 weeks of faling builds. A Project with a failing build over a long has proven to the world that nobody is interested in it. In addition, automatic builds don't make sense if they are broken all the time. -> Pharo Clients now are build for 1.3 in addition to 1.2 https://pharo-ic.lille.inria.fr/hudson/view/Pharo-Clients/ NOTE: Seaside is broken. This is due to MetaCello config that needs to be done for 1.3. It would be nice if someone would take the lead in fixing this. Marcus -- Marcus Denker -- http://www.marcusdenker.de INRIA Lille -- Nord Europe. Team RMoD.

May 9, 2011

Re: [Pharo-project] [update 1.3] #13185

by Benjamin

On May 7, 2011, at 6:35 PM, StÃ©phane Ducasse wrote: > 13185 > ----- > > - Fix FlatBrowser not to use PluggableTexMorphWithLimits (since this is in nautilus for now).\ PluggableTexMorphWithLimits is in a different package than Nautilus especially because FlatMessageListBrowser used to use it :) But it's not a big deal :) Ben > - Issue 4142: Never use a leadingChar for byte char. Thanks levente and Nicolas Cellier. > - Issue 4090: Better default template for absent class comment. > > Stef

May 9, 2011