December 2013 - Pharo-dev - lists.pharo.org

Re: [Pharo-dev] Responsible development

by Benjamin

I did not say I did it :) Sorry for the misunderstanding. I was saying that theoretically it should not freeze the UI thread if you fork. But then it also mean you should freeze the UI thread from time to time while doing critical operations So itâs actually a bit tricky to implement Ben On 03 Dec 2013, at 11:40, kilon alios <kilon.alios(a)gmail.com> wrote: > looks like your post to the mailing list was not as pointless as you thought ;) > > Great work Benjamin > > > On Tue, Dec 3, 2013 at 12:38 PM, Yuriy Tymchuk <yuriy.tymchuk(a)me.com> wrote: > > On 03 Dec 2013, at 11:08, Benjamin <Benjamin.VanRyseghem.Pharo(a)gmail.com> wrote: > >> On 03 Dec 2013, at 10:01, StÃ©phane Ducasse <stephane.ducasse(a)inria.fr> wrote: >> >>> >>> On Dec 2, 2013, at 8:37 PM, Benjamin <Benjamin.VanRyseghem.Pharo(a)gmail.com> wrote: >>> >>>> On 02 Dec 2013, at 20:27, StÃ©phane Ducasse <stephane.ducasse(a)inria.fr> wrote: >>>> >>>>>>>> We try now to have responsive UIs in the sense the tools like Nautilus try to >>>>>>>> run things in a separate thread. >>>>>>>> >>>>>>>> I will do an experiment and fork each Nautilus opening to see if it can save my ass :P >>>>>>> :) >>>>>>> >>>>>>> personnally I would be really against because just forking is just a way to have a lot more mess in the future. >>>>>> >>>>>> Why ? >>>>> >>>>> Because you do not know when you invariants should hold. Normally you expect them to hold once the system is loaded. >>>>> Because loading for example act as an atomic action when you modify the system. Now if your thread can see and modify >>>>> different versions of the state be prepared to have really strange and difficult bugs to find. >>>>> >>>>> I prefer to have cache than to have forked processes around. >>>> >>>> Cache will not help you killing Nautilus when it freezes your image >>> >>> fork neither. >> >> It should not freeze your image anymore, only its own thread > > Wow, that was fast > >> >> Ben >> >>> >>>> (why cache by the way ?) >>> >>> I thought the discussion was about speeding up nautilus when performing start up actions. >>>> >>>> Ben >>>> >>>>> >>>>> Stef > >

Dec. 3, 2013

Re: [Pharo-dev] transcript is plain bad

by phil＠highoctane.be

What's wrong with Strings anyway? But yeah, I understand your point. What I have done for my own stuff is to have a TLog aspect with info, debug, error, warn and, well, Strings. But double dispatch could take care of that, like printOn:, storeOn: etc. Do you have some code that I can look at? TLog>>info: anObject anObject logOn: Logger default info. Phil On Tue, Dec 3, 2013 at 11:27 AM, StÃ©phane Ducasse <stephane.ducasse(a)inria.fr > wrote: > > On Dec 3, 2013, at 10:34 AM, phil(a)highoctane.be wrote: > > I am working on this one. > > Don't kill the transcript, it is useful when working without a console. > > > I will not kill it but it will just be dull outputer for my logger. In > fact I'm missing a user interface (the idea was to build it with fernando) > and right now > the transcript just displays the printstring of my log objects. So I can > even control what I want to output. > more important I can filter the log ojbjects. For example I need to know > when my framework finish to deal with a class and now this is mix with the > compiler > logherea > > Now, adding the features you mention is not hard. And the top right > triangle menu of Transcript could get a clue from the Workspace one with a > Save As entry. > > > No i just DO NOT WANT STRINGS!!!! > > I once had an issue with things like (out of the top of my head) > FileStream stdout nextPutAll: 'hello'; cr; flush. when I went headless. The > image worked at one point and the froze/locked. Until this is sorted out in > full, keep a transcript handy. > > > I gave the image to Guille who sent me back a version "unlocked" with Oz, > but it still dies on Windows for example. Weird. > > Phil > > > > --- > Philippe Back > Dramatic Performance Improvements > Mob: +32(0) 478 650 140 | Fax: +32 (0) 70 408 027 > Mail:phil@highoctane.be | Web: http://philippeback.eu > Blog: http://philippeback.be | Twitter: @philippeback > Youtube: http://www.youtube.com/user/philippeback/videos > > High Octane SPRL > rue cour Boisacq 101 | 1301 Bierges | Belgium > > Pharo Consortium Member - http://consortium.pharo.org/ > Featured on the Software Process and Measurement Cast - > http://spamcast.libsyn.com > Sparx Systems Enterprise Architect and Ability Engineering EADocX Value > Added Reseller > > > > > On Tue, Dec 3, 2013 at 10:10 AM, StÃ©phane Ducasse < > stephane.ducasse(a)inria.fr> wrote: > >> >> On Dec 2, 2013, at 8:46 PM, phil(a)highoctane.be wrote: >> >> And missing a ton of methods. >> >> Teh StackInterpreterSimulator is agreeing with this statement. >> >> >> send us the ones that are missing. >> >> Please note that we should kill the transcript in the future because this >> is a ^%$&*((^%$* outdated way to manage >> feedback: plain string, close the window data is lost, limited in length, >> no filteringâ¦.. >> bad bad bad does not scale horrible and so frustrating. I will not send >> you the current transcript I have because I deal with private code >> but this does not work. >> >> Stef >> >> >> >> >> >> >> On Mon, Dec 2, 2013 at 8:05 PM, Esteban Lorenzano <estebanlm(a)gmail.com>wrote: >> >>> yes, the transcript is ThreadSafeTranscript >>> >>> >>> On Mon, Dec 2, 2013 at 8:00 PM, Yuriy Tymchuk <yuriy.tymchuk(a)me.com>wrote: >>> >>>> Buy the way, is absence of Transcript class - ok? >>>> >>>> I know how it all works now, but is it ok? :) >>>> >>>> uko >>>> >>>> >>>> On 02 Dec 2013, at 19:58, Esteban Lorenzano <estebanlm(a)gmail.com> >>>> wrote: >>>> >>>> yes, please :) >>>> >>>> >>>> On Mon, Dec 2, 2013 at 7:56 PM, StÃ©phane Ducasse < >>>> stephane.ducasse(a)inria.fr> wrote: >>>> >>>>> >>>>> > The return of the toothpick is near. o/ >>>>> >>>>> No I will finish my logger :) >>>>> >>>>> Stef >>>>> >>>>> >>>> >>>> >>> >> >> > >

Dec. 3, 2013

Re: [Pharo-dev] Opal hook?

by StÃ©phane Ducasse

On Dec 3, 2013, at 11:20 AM, ClÃ©ment Bera <bera.clement(a)gmail.com> wrote: > Then for your project replace ProtoObject by StubRootClass in UndefinedObject>>subclass:instanceVariableNames:classVariableNames:poolDictionaries:category: it will work. yes but this is ugly because I do not want override. So I will introduce an exception so that we can do that in a clean way. > An alternative is to catch OCSemanticWarning or its subclass OCUndeclaredVariableWarning while loading the new classes and execute another code instead of the default action in the exception handling block do you know if the OCUndeclaredVariable is raised for unknown superclass? How to undefined and OCUndeclaredVariableWarning relate? Stef > > > 2013/12/3 StÃ©phane Ducasse <stephane.ducasse(a)inria.fr> > > Well if the class does not exists it calls the method on UndefinedObject (UndeclaredBinding). And you have: > > I will check and see if I can set up a hook because having classes subclasses of protoobject is not a good idea > when loading broken code. > > > UndefinedObject>>subclass: nameOfClass > > instanceVariableNames: instVarNames > > classVariableNames: classVarNames > > poolDictionaries: poolDictnames > > category: category > > "Calling this method is now considered an accident. If you really want to create a class with a nil superclass, then create the class and then set the superclass using #superclass:" > > self traceCr: ('Attempt to create ', nameOfClass, ' as a subclass of nil. Possibly a class is being loaded before its superclass.'). > > ^ProtoObject > > subclass: nameOfClass > > instanceVariableNames: instVarNames > > classVariableNames: classVarNames > > poolDictionaries: poolDictnames > > category: category > > > > The best is to set manually the superclass after with #superclass: > > No because I have no idea what are the classes whose superclasses are not in the image. > With a proper hook I could just say > if the superclass is unknow please use the "StubRootClass" of my framework. > else let the system handles it. > > Stef >

Dec. 3, 2013

Re: [Pharo-dev] Grease conflicts #packages with RPackage

by StÃ©phane Ducasse

The real problem is that a CLASS now responds to the packages message. Stef On Dec 3, 2013, at 11:10 AM, Stephan Eggermont <stephan(a)stack.nl> wrote: > Mariano wrote: >> You will only have problems with class side #packages. >> Are all those 35 class side? > > 5 out of 35. 4 of which are in Pharo 3 > > Stephan >

Dec. 3, 2013

Re: [Pharo-dev] Responsible development

by kilon alios

looks like your post to the mailing list was not as pointless as you thought ;) Great work Benjamin On Tue, Dec 3, 2013 at 12:38 PM, Yuriy Tymchuk <yuriy.tymchuk(a)me.com> wrote: > > On 03 Dec 2013, at 11:08, Benjamin <Benjamin.VanRyseghem.Pharo(a)gmail.com> > wrote: > > On 03 Dec 2013, at 10:01, StÃ©phane Ducasse <stephane.ducasse(a)inria.fr> > wrote: > > > On Dec 2, 2013, at 8:37 PM, Benjamin <Benjamin.VanRyseghem.Pharo(a)gmail.com> > wrote: > > On 02 Dec 2013, at 20:27, StÃ©phane Ducasse <stephane.ducasse(a)inria.fr> > wrote: > > We try now to have responsive UIs in the sense the tools like Nautilus try > to > run things in a separate thread. > > I will do an experiment and fork each Nautilus opening to see if it can > save my ass :P > > :) > > personnally I would be really against because just forking is just a way > to have a lot more mess in the future. > > > Why ? > > > Because you do not know when you invariants should hold. Normally you > expect them to hold once the system is loaded. > Because loading for example act as an atomic action when you modify the > system. Now if your thread can see and modify > different versions of the state be prepared to have really strange and > difficult bugs to find. > > I prefer to have cache than to have forked processes around. > > > Cache will not help you killing Nautilus when it freezes your image > > > fork neither. > > > It should not freeze your image anymore, only its own thread > > > Wow, that was fast > > > Ben > > > (why cache by the way ?) > > > I thought the discussion was about speeding up nautilus when performing > start up actions. > > > Ben > > > Stef > > >

Dec. 3, 2013

Re: [Pharo-dev] Responsible development

by Yuriy Tymchuk

On 03 Dec 2013, at 11:08, Benjamin <Benjamin.VanRyseghem.Pharo(a)gmail.com> wrote: > On 03 Dec 2013, at 10:01, StÃ©phane Ducasse <stephane.ducasse(a)inria.fr> wrote: > >> >> On Dec 2, 2013, at 8:37 PM, Benjamin <Benjamin.VanRyseghem.Pharo(a)gmail.com> wrote: >> >>> On 02 Dec 2013, at 20:27, StÃ©phane Ducasse <stephane.ducasse(a)inria.fr> wrote: >>> >>>>>>> We try now to have responsive UIs in the sense the tools like Nautilus try to >>>>>>> run things in a separate thread. >>>>>>> >>>>>>> I will do an experiment and fork each Nautilus opening to see if it can save my ass :P >>>>>> :) >>>>>> >>>>>> personnally I would be really against because just forking is just a way to have a lot more mess in the future. >>>>> >>>>> Why ? >>>> >>>> Because you do not know when you invariants should hold. Normally you expect them to hold once the system is loaded. >>>> Because loading for example act as an atomic action when you modify the system. Now if your thread can see and modify >>>> different versions of the state be prepared to have really strange and difficult bugs to find. >>>> >>>> I prefer to have cache than to have forked processes around. >>> >>> Cache will not help you killing Nautilus when it freezes your image >> >> fork neither. > > It should not freeze your image anymore, only its own thread Wow, that was fast > > Ben > >> >>> (why cache by the way ?) >> >> I thought the discussion was about speeding up nautilus when performing start up actions. >>> >>> Ben >>> >>>> >>>> Stef

Dec. 3, 2013

Re: [Pharo-dev] transcript is plain bad

by StÃ©phane Ducasse

On Dec 3, 2013, at 10:34 AM, phil(a)highoctane.be wrote: > I am working on this one. > > Don't kill the transcript, it is useful when working without a console. I will not kill it but it will just be dull outputer for my logger. In fact I'm missing a user interface (the idea was to build it with fernando) and right now the transcript just displays the printstring of my log objects. So I can even control what I want to output. more important I can filter the log ojbjects. For example I need to know when my framework finish to deal with a class and now this is mix with the compiler logherea > Now, adding the features you mention is not hard. And the top right triangle menu of Transcript could get a clue from the Workspace one with a Save As entry. No i just DO NOT WANT STRINGS!!!! > I once had an issue with things like (out of the top of my head) FileStream stdout nextPutAll: 'hello'; cr; flush. when I went headless. The image worked at one point and the froze/locked. Until this is sorted out in full, keep a transcript handy. > > I gave the image to Guille who sent me back a version "unlocked" with Oz, but it still dies on Windows for example. Weird. > > Phil > > > > --- > Philippe Back > Dramatic Performance Improvements > Mob: +32(0) 478 650 140 | Fax: +32 (0) 70 408 027 > Mail:phil@highoctane.be | Web: http://philippeback.eu > Blog: http://philippeback.be | Twitter: @philippeback > Youtube: http://www.youtube.com/user/philippeback/videos > > High Octane SPRL > rue cour Boisacq 101 | 1301 Bierges | Belgium > > Pharo Consortium Member - http://consortium.pharo.org/ > Featured on the Software Process and Measurement Cast - http://spamcast.libsyn.com > Sparx Systems Enterprise Architect and Ability Engineering EADocX Value Added Reseller > > > > > On Tue, Dec 3, 2013 at 10:10 AM, StÃ©phane Ducasse <stephane.ducasse(a)inria.fr> wrote: > > On Dec 2, 2013, at 8:46 PM, phil(a)highoctane.be wrote: > >> And missing a ton of methods. >> >> Teh StackInterpreterSimulator is agreeing with this statement. > > send us the ones that are missing. > > Please note that we should kill the transcript in the future because this is a ^%$&*((^%$* outdated way to manage > feedback: plain string, close the window data is lost, limited in length, no filteringâ¦.. > bad bad bad does not scale horrible and so frustrating. I will not send you the current transcript I have because I deal with private code > but this does not work. > > Stef >> >> >> >> >> >> On Mon, Dec 2, 2013 at 8:05 PM, Esteban Lorenzano <estebanlm(a)gmail.com> wrote: >> yes, the transcript is ThreadSafeTranscript >> >> >> On Mon, Dec 2, 2013 at 8:00 PM, Yuriy Tymchuk <yuriy.tymchuk(a)me.com> wrote: >> Buy the way, is absence of Transcript class - ok? >> >> I know how it all works now, but is it ok? :) >> >> uko >> >> >> On 02 Dec 2013, at 19:58, Esteban Lorenzano <estebanlm(a)gmail.com> wrote: >> >>> yes, please :) >>> >>> >>> On Mon, Dec 2, 2013 at 7:56 PM, StÃ©phane Ducasse <stephane.ducasse(a)inria.fr> wrote: >>> >>> > The return of the toothpick is near. o/ >>> >>> No I will finish my logger :) >>> >>> Stef >>> >>> >> >> >> > >

Dec. 3, 2013

Re: [Pharo-dev] Opal hook?

by ClÃ©ment Bera

Then for your project replace ProtoObject by StubRootClass in UndefinedObject>>subclass:instanceVariableNames:classVariableNames: poolDictionaries:category: it will work. An alternative is to catch OCSemanticWarning or its subclass OCUndeclaredVariableWarning while loading the new classes and execute another code instead of the default action in the exception handling block 2013/12/3 StÃ©phane Ducasse <stephane.ducasse(a)inria.fr> > > Well if the class does not exists it calls the method on UndefinedObject > (UndeclaredBinding). And you have: > > I will check and see if I can set up a hook because having classes > subclasses of protoobject is not a good idea > when loading broken code. > > > UndefinedObject>>subclass: nameOfClass > > instanceVariableNames: instVarNames > > classVariableNames: classVarNames > > poolDictionaries: poolDictnames > > category: category > > "Calling this method is now considered an accident. If you really > want to create a class with a nil superclass, then create the class and > then set the superclass using #superclass:" > > self traceCr: ('Attempt to create ', nameOfClass, ' as a subclass > of nil. Possibly a class is being loaded before its superclass.'). > > ^ProtoObject > > subclass: nameOfClass > > instanceVariableNames: instVarNames > > classVariableNames: classVarNames > > poolDictionaries: poolDictnames > > category: category > > > > The best is to set manually the superclass after with #superclass: > > No because I have no idea what are the classes whose superclasses are not > in the image. > With a proper hook I could just say > if the superclass is unknow please use the "StubRootClass" of my > framework. > else let the system handles it. > > Stef >

Dec. 3, 2013

Re: [Pharo-dev] Grease conflicts #packages with RPackage

by Stephan Eggermont

Mariano wrote: > You will only have problems with class side #packages. > Are all those 35 class side? 5 out of 35. 4 of which are in Pharo 3 Stephan

Dec. 3, 2013

Re: [Pharo-dev] Responsible development

by Benjamin

On 03 Dec 2013, at 10:01, StÃ©phane Ducasse <stephane.ducasse(a)inria.fr> wrote: > > On Dec 2, 2013, at 8:37 PM, Benjamin <Benjamin.VanRyseghem.Pharo(a)gmail.com> wrote: > >> On 02 Dec 2013, at 20:27, StÃ©phane Ducasse <stephane.ducasse(a)inria.fr> wrote: >> >>>>>> We try now to have responsive UIs in the sense the tools like Nautilus try to >>>>>> run things in a separate thread. >>>>>> >>>>>> I will do an experiment and fork each Nautilus opening to see if it can save my ass :P >>>>> :) >>>>> >>>>> personnally I would be really against because just forking is just a way to have a lot more mess in the future. >>>> >>>> Why ? >>> >>> Because you do not know when you invariants should hold. Normally you expect them to hold once the system is loaded. >>> Because loading for example act as an atomic action when you modify the system. Now if your thread can see and modify >>> different versions of the state be prepared to have really strange and difficult bugs to find. >>> >>> I prefer to have cache than to have forked processes around. >> >> Cache will not help you killing Nautilus when it freezes your image > > fork neither. It should not freeze your image anymore, only its own thread Ben > >> (why cache by the way ?) > > I thought the discussion was about speeding up nautilus when performing start up actions. >> >> Ben >> >>> >>> Stef

Dec. 3, 2013