- Pharo-dev - lists.pharo.org

[pharo-project/pharo-core] ccaa84: 20625

by GitHub

Branch: refs/heads/2.0 Home: https://github.com/pharo-project/pharo-core Commit: ccaa8405f452676d149e42759a742dc3db2c9ea5 https://github.com/pharo-project/pharo-core/commit/ccaa8405f452676d149e4275… Author: Jenkins Build Server <board(a)pharo-project.org> Date: 2013-10-25 (Fri, 25 Oct 2013) Changed paths: R NautilusCommon.package/RBBrowserEnvironment.extension/instance/packageOrganizer.st R NautilusCommon.package/RBBrowserEnvironment.extension/instance/packages.st M NautilusCommon.package/RBBrowserEnvironment.extension/methodProperties.json R NautilusCommon.package/RBBrowserEnvironmentWrapper.extension/instance/packages.st R NautilusCommon.package/RBBrowserEnvironmentWrapper.extension/methodProperties.json R NautilusCommon.package/RBBrowserEnvironmentWrapper.extension/properties.json A Refactoring-Environment.package/RBBrowserEnvironment.class/instance/packageOrganizer.st A Refactoring-Environment.package/RBBrowserEnvironment.class/instance/packages.st M Refactoring-Environment.package/RBBrowserEnvironment.class/methodProperties.json A Refactoring-Environment.package/RBBrowserEnvironmentWrapper.class/instance/packages.st M Refactoring-Environment.package/RBBrowserEnvironmentWrapper.class/methodProperties.json A Refactoring-Environment.package/RBSelectorEnvironment.class/instance/packages.st M Refactoring-Environment.package/RBSelectorEnvironment.class/methodProperties.json M ScriptLoader20.package/ScriptLoader.class/instance/commentForCurrentUpdate.st A ScriptLoader20.package/ScriptLoader.class/instance/script598.st A ScriptLoader20.package/ScriptLoader.class/instance/update20625.st M ScriptLoader20.package/ScriptLoader.class/methodProperties.json Log Message: ----------- 20625 11535 RBBrowserEnvironment packages method is too broad backport https://pharo.fogbugz.com/f/cases/11535

Oct. 25, 2013

Re: [Pharo-dev] ensureCreateDirectory?

by Norbert Hartl

Am 25.10.2013 um 15:52 schrieb Sven Van Caekenberghe <sven(a)stfx.eu>: > The problem with #ensure style selectors is that they are confusing because they both test something and maybe do something to create a certain condition. The only way to take away the confusion is to make a very long selector. That is why comments are also important. > Maybe. I didnât participate in this discussion. And I donât know about this obscure consistency rules that this particular selector follows. Languages and consistency rules are probably troublesome most of the time. In this particular case I would it have called just like others,too. That would be ensureDirectoryExists. You can indeed âmake sureâ that a âparticular objectâ has a âcertain stateâ after the method is done. But in a single case it is always easy to argue. I would need to see the full list to get in trouble . > Like Stef said, this was already debated for quite some time. The current choice is maybe not perfect, there is nothing stopping anyone from suggesting a better alternative. Good luck with getting a consensus. > Oh, hell, no. Iâm stupid but not completely insane :) Norbert > On 25 Oct 2013, at 15:28, Norbert Hartl <norbert(a)hartl.name> wrote: > >> Stef, >> >> I really donât want to stretch this into an argue â¦. but I have problems understanding that. There are refactorings of method selectors because it is said it is important that includesSubString: is written as includesSubstring: and there are other examples. Now we have a selector that people find confusing and that is not important? I donât understand it and the only difference I can see by now is the way you or your team is involved in the action. So the only thing I would understand is sensitivities meaning you are sick of complaints regarding your work. Otherwise I donât understand that. >> >> Norbert >> >> Am 25.10.2013 um 15:04 schrieb StÃ©phane Ducasse <stephane.ducasse(a)inria.fr>: >> >>> can we focus on real problems? >>> The method has a comment. >>> >>> Stef >>> >>> On Oct 25, 2013, at 10:39 AM, Norbert Hartl <norbert(a)hartl.name> wrote: >>> >>>> >>>> Am 25.10.2013 um 10:11 schrieb Nicolas Cellier <nicolas.cellier.aka.nice(a)gmail.com>: >>>> >>>>> So action=create object=directory >>>>> you want to be sure that you create a directory? >>>>> What if it already exists? >>>>> Since you want to be sure to create, will you erase and re-create? >>>>> You see the kind of misunderstanding this could create... >>>>> IOW, self ensureCreateMisunderstanding >>>>> >>>> +1 I didnât like it in the first place. It is just confusing. And probably an example how hard it can be if two or more non-native speaker try to find something that makes sense. >>>> >>>> Norbert >>>>> >>>>> 2013/10/25 StÃ©phane Ducasse <stephane.ducasse(a)inria.fr> >>>>> Consistency >>>>> ensureActionObject >>>>> >>>>> ensureCreateDirectory >>>>> ensureCreateFile >>>>> >>>>> We discuss a lot with camillo when we did it and I think that this is good right now. >>>>> >>>>>> Out of curiosity, what was the motivation behind Issue 10924: Simplifying >>>>>> ensure* and create protocol? >>>>>> >>>>>> For me #ensureCreateDirectory is less intention revealing than >>>>>> #ensureDirectory because the "ensure" means "create only if necessary", not >>>>>> "definitely create" a directory. What am I missing? Thanks. >>>>>> >>>>>> >>>>>> >>>>>> ----- >>>>>> Cheers, >>>>>> Sean >>>>>> -- >>>>>> View this message in context: http://forum.world.st/ensureCreateDirectory-tp4716925.html >>>>>> Sent from the Pharo Smalltalk Developers mailing list archive at Nabble.com. >>>>>> >>>>> >>>>> >>>>> >>>> >>> >> > >

Oct. 25, 2013

[update 2.0] #20625

by Marcus Denker

20625 ----- 11535 RBBrowserEnvironment packages method is too broad backport https://pharo.fogbugz.com/f/cases/11535 Diff information: http://ss3.gemstone.com/ss/Pharo20/Refactoring-Environment-MarcusDenker.14.… http://ss3.gemstone.com/ss/Pharo20/NautilusCommon-MarcusDenker.122.diff

Oct. 25, 2013

Re: [Pharo-dev] DeepCopier>>isItTimeToCheckVariables

by Mariano Martinez Peck

Martin watching the DeepCopier??? mmmm I am afraid to ask why ;) On Fri, Oct 25, 2013 at 6:10 AM, Martin Dias <tinchodias(a)gmail.com> wrote: > Hi, > > This method in DeepCopier has no senders (in Pharo 3) and it's the only > who references the class variable NextVariableCheckTime. > > The time stamp says 1999... Unused code to remove? > > > isItTimeToCheckVariables > > | now isIt | > NextVariableCheckTime ifNil: [ > NextVariableCheckTime := Time totalSeconds. > ^ true]. > now := Time totalSeconds. > isIt := NextVariableCheckTime < now. > isIt ifTrue: ["update time for next check" > NextVariableCheckTime := now + self intervalForChecks]. > ^isIt > > > Cheers, > MartÃn > -- Mariano http://marianopeck.wordpress.com

Oct. 25, 2013

Re: [Pharo-dev] ensureCreateDirectory?

by Sven Van Caekenberghe

The problem with #ensure style selectors is that they are confusing because they both test something and maybe do something to create a certain condition. The only way to take away the confusion is to make a very long selector. That is why comments are also important. Like Stef said, this was already debated for quite some time. The current choice is maybe not perfect, there is nothing stopping anyone from suggesting a better alternative. Good luck with getting a consensus. On 25 Oct 2013, at 15:28, Norbert Hartl <norbert(a)hartl.name> wrote: > Stef, > > I really donât want to stretch this into an argue â¦. but I have problems understanding that. There are refactorings of method selectors because it is said it is important that includesSubString: is written as includesSubstring: and there are other examples. Now we have a selector that people find confusing and that is not important? I donât understand it and the only difference I can see by now is the way you or your team is involved in the action. So the only thing I would understand is sensitivities meaning you are sick of complaints regarding your work. Otherwise I donât understand that. > > Norbert > > Am 25.10.2013 um 15:04 schrieb StÃ©phane Ducasse <stephane.ducasse(a)inria.fr>: > >> can we focus on real problems? >> The method has a comment. >> >> Stef >> >> On Oct 25, 2013, at 10:39 AM, Norbert Hartl <norbert(a)hartl.name> wrote: >> >>> >>> Am 25.10.2013 um 10:11 schrieb Nicolas Cellier <nicolas.cellier.aka.nice(a)gmail.com>: >>> >>>> So action=create object=directory >>>> you want to be sure that you create a directory? >>>> What if it already exists? >>>> Since you want to be sure to create, will you erase and re-create? >>>> You see the kind of misunderstanding this could create... >>>> IOW, self ensureCreateMisunderstanding >>>> >>> +1 I didnât like it in the first place. It is just confusing. And probably an example how hard it can be if two or more non-native speaker try to find something that makes sense. >>> >>> Norbert >>>> >>>> 2013/10/25 StÃ©phane Ducasse <stephane.ducasse(a)inria.fr> >>>> Consistency >>>> ensureActionObject >>>> >>>> ensureCreateDirectory >>>> ensureCreateFile >>>> >>>> We discuss a lot with camillo when we did it and I think that this is good right now. >>>> >>>> > Out of curiosity, what was the motivation behind Issue 10924: Simplifying >>>> > ensure* and create protocol? >>>> > >>>> > For me #ensureCreateDirectory is less intention revealing than >>>> > #ensureDirectory because the "ensure" means "create only if necessary", not >>>> > "definitely create" a directory. What am I missing? Thanks. >>>> > >>>> > >>>> > >>>> > ----- >>>> > Cheers, >>>> > Sean >>>> > -- >>>> > View this message in context: http://forum.world.st/ensureCreateDirectory-tp4716925.html >>>> > Sent from the Pharo Smalltalk Developers mailing list archive at Nabble.com. >>>> > >>>> >>>> >>>> >>> >> >

Oct. 25, 2013

Re: [Pharo-dev] ensureCreateDirectory?

by StÃ©phane Ducasse

On Oct 25, 2013, at 3:24 PM, Nicolas Cellier <nicolas.cellier.aka.nice(a)gmail.com> wrote: > What are real problems exactly? Here is a list - better API for Compiler - better functionality for OPAL - Cleaning Morph (architecture, package structure) - using Slot for real - Removing CodeHolder hierarchy - Improving the tools (usability) - Making sure that in the future we can remove streams and use another Stream library - Removing leadingCharacter and other mess - Running SmallLint on all the packages - Building Pharo from a set of project using metacello - Removing pseudoclass and other shit - Rewriting morphTreeMorph - Enhancing the widgtes set: table for example is missing - Building a UI builder - Cleaning Text - Cleaning morph event model - Supporting GIT for real - improving Log - automatic recategorization of all the methods. - Automatic validation of packages - NativeBoost for ARM - Smoothly supporting remote browsing if you want more I'm quite confident that I can find more problems that I would love to see fixed in Pharo. Stef > If you are going to open each and every method to read the comment or the code instead of trusting the name, IMHO that is going to be a real problem. > If the focus shall be on real problems, then why change APIs which are not a real problem? > > > 2013/10/25 StÃ©phane Ducasse <stephane.ducasse(a)inria.fr> > can we focus on real problems? > The method has a comment. > > Stef > > On Oct 25, 2013, at 10:39 AM, Norbert Hartl <norbert(a)hartl.name> wrote: > >> >> Am 25.10.2013 um 10:11 schrieb Nicolas Cellier <nicolas.cellier.aka.nice(a)gmail.com>: >> >>> So action=create object=directory >>> you want to be sure that you create a directory? >>> What if it already exists? >>> Since you want to be sure to create, will you erase and re-create? >>> You see the kind of misunderstanding this could create... >>> IOW, self ensureCreateMisunderstanding >>> >> +1 I didnât like it in the first place. It is just confusing. And probably an example how hard it can be if two or more non-native speaker try to find something that makes sense. >> >> Norbert >>> >>> 2013/10/25 StÃ©phane Ducasse <stephane.ducasse(a)inria.fr> >>> Consistency >>> ensureActionObject >>> >>> ensureCreateDirectory >>> ensureCreateFile >>> >>> We discuss a lot with camillo when we did it and I think that this is good right now. >>> >>> > Out of curiosity, what was the motivation behind Issue 10924: Simplifying >>> > ensure* and create protocol? >>> > >>> > For me #ensureCreateDirectory is less intention revealing than >>> > #ensureDirectory because the "ensure" means "create only if necessary", not >>> > "definitely create" a directory. What am I missing? Thanks. >>> > >>> > >>> > >>> > ----- >>> > Cheers, >>> > Sean >>> > -- >>> > View this message in context: http://forum.world.st/ensureCreateDirectory-tp4716925.html >>> > Sent from the Pharo Smalltalk Developers mailing list archive at Nabble.com. >>> > >>> >>> >>> >> > >

Oct. 25, 2013

Re: [Pharo-dev] ensureCreateDirectory?

by Nicolas Cellier

Yes, I can understand the boring part of it - this is a detail - it's tiring to discuss small details to death - some people speak more than act (me for example :) But can we agree that - details count - critics count (convince yourself that they are not intended for bashing people) - and last one, a good comment does not redeem a bad selector (kudos to one writing the lint rule for this one) 2013/10/25 StÃ©phane Ducasse <stephane.ducasse(a)inria.fr> > We already discussed this API > it is in the bug tracker: ensureDelete, ensureDeletedâ¦. > > So do what you want I personally want to focus on more important issues. > Once the full world will agree then do not forget to update the FileSystem > Chapter > that I wrote BTW (this is for the people complaining about the lack of > documentation). > > Stef > > PERFECT is the ennemy of DONE is my motto too. > > > On Oct 25, 2013, at 3:24 PM, Nicolas Cellier < > nicolas.cellier.aka.nice(a)gmail.com> wrote: > > What are real problems exactly? > If you are going to open each and every method to read the comment or the > code instead of trusting the name, IMHO that is going to be a real problem. > If the focus shall be on real problems, then why change APIs which are not > a real problem? > > > 2013/10/25 StÃ©phane Ducasse <stephane.ducasse(a)inria.fr> > >> can we focus on real problems? >> The method has a comment. >> >> Stef >> >> On Oct 25, 2013, at 10:39 AM, Norbert Hartl <norbert(a)hartl.name> wrote: >> >> >> Am 25.10.2013 um 10:11 schrieb Nicolas Cellier < >> nicolas.cellier.aka.nice(a)gmail.com>: >> >> So action=create object=directory >> you want to be sure that you create a directory? >> What if it already exists? >> Since you want to be sure to create, will you erase and re-create? >> You see the kind of misunderstanding this could create... >> IOW, self ensureCreateMisunderstanding >> >> +1 I didnât like it in the first place. It is just confusing. And >> probably an example how hard it can be if two or more non-native speaker >> try to find something that makes sense. >> >> Norbert >> >> >> 2013/10/25 StÃ©phane Ducasse <stephane.ducasse(a)inria.fr> >> >>> Consistency >>> ensureActionObject >>> >>> ensureCreateDirectory >>> ensureCreateFile >>> >>> We discuss a lot with camillo when we did it and I think that this is >>> good right now. >>> >>> > Out of curiosity, what was the motivation behind Issue 10924: >>> Simplifying >>> > ensure* and create protocol? >>> > >>> > For me #ensureCreateDirectory is less intention revealing than >>> > #ensureDirectory because the "ensure" means "create only if >>> necessary", not >>> > "definitely create" a directory. What am I missing? Thanks. >>> > >>> > >>> > >>> > ----- >>> > Cheers, >>> > Sean >>> > -- >>> > View this message in context: >>> http://forum.world.st/ensureCreateDirectory-tp4716925.html >>> > Sent from the Pharo Smalltalk Developers mailing list archive at >>> Nabble.com <http://nabble.com/>. >>> > >>> >>> >>> >> >> >> > >

Oct. 25, 2013

Re: [Pharo-dev] [pharo-project/pharo-core] 4021f3: 30523

by Goubier Thierry

Le 25/10/2013 15:27, Marcus Denker a Ã©crit : > > ... even with the package tags there is too much in the left pane. Yes, I've seen that as well. Either go into the six pane browser of S. Travis Pope, or use the tree to build proper categories of packages like I do in the AltBrowser. I.e. it's a GUI issue, not a package issue. Thierry > On 25 Oct 2013 at 15:19:02, Christophe Demarey (christophe.demarey(a)inria.fr) wrote: >> >> Why do we need to remove configurations of projects loaded into the image? >> I would say the opposite: all projects in the image should have their configuration loaded. It is part of the project. >> >> Le 25 oct. 2013 Ã 14:00, GitHub a Ã©crit : >>> 12005 unload BaselineOfFileTree, BaselineOfMetacello >>> https://pharo.fogbugz.com/f/cases/12005 >> >> >> - smime.p7s, 6 KB > > > > -- Thierry Goubier CEA list Laboratoire des Fondations des SystÃ¨mes Temps RÃ©el EmbarquÃ©s 91191 Gif sur Yvette Cedex France Phone/Fax: +33 (0) 1 69 08 32 92 / 83 95

Oct. 25, 2013

Re: [Pharo-dev] ensureCreateDirectory?

by StÃ©phane Ducasse

We already discussed this API it is in the bug tracker: ensureDelete, ensureDeletedâ¦. So do what you want I personally want to focus on more important issues. Once the full world will agree then do not forget to update the FileSystem Chapter that I wrote BTW (this is for the people complaining about the lack of documentation). Stef PERFECT is the ennemy of DONE is my motto too. On Oct 25, 2013, at 3:24 PM, Nicolas Cellier <nicolas.cellier.aka.nice(a)gmail.com> wrote: > What are real problems exactly? > If you are going to open each and every method to read the comment or the code instead of trusting the name, IMHO that is going to be a real problem. > If the focus shall be on real problems, then why change APIs which are not a real problem? > > > 2013/10/25 StÃ©phane Ducasse <stephane.ducasse(a)inria.fr> > can we focus on real problems? > The method has a comment. > > Stef > > On Oct 25, 2013, at 10:39 AM, Norbert Hartl <norbert(a)hartl.name> wrote: > >> >> Am 25.10.2013 um 10:11 schrieb Nicolas Cellier <nicolas.cellier.aka.nice(a)gmail.com>: >> >>> So action=create object=directory >>> you want to be sure that you create a directory? >>> What if it already exists? >>> Since you want to be sure to create, will you erase and re-create? >>> You see the kind of misunderstanding this could create... >>> IOW, self ensureCreateMisunderstanding >>> >> +1 I didnât like it in the first place. It is just confusing. And probably an example how hard it can be if two or more non-native speaker try to find something that makes sense. >> >> Norbert >>> >>> 2013/10/25 StÃ©phane Ducasse <stephane.ducasse(a)inria.fr> >>> Consistency >>> ensureActionObject >>> >>> ensureCreateDirectory >>> ensureCreateFile >>> >>> We discuss a lot with camillo when we did it and I think that this is good right now. >>> >>> > Out of curiosity, what was the motivation behind Issue 10924: Simplifying >>> > ensure* and create protocol? >>> > >>> > For me #ensureCreateDirectory is less intention revealing than >>> > #ensureDirectory because the "ensure" means "create only if necessary", not >>> > "definitely create" a directory. What am I missing? Thanks. >>> > >>> > >>> > >>> > ----- >>> > Cheers, >>> > Sean >>> > -- >>> > View this message in context: http://forum.world.st/ensureCreateDirectory-tp4716925.html >>> > Sent from the Pharo Smalltalk Developers mailing list archive at Nabble.com. >>> > >>> >>> >>> >> > >

Oct. 25, 2013

Re: [Pharo-dev] [pharo-project/pharo-core] 4021f3: 30523

by Tudor Girba

I agree with this. In particular, the configuration classes offer extra information that can be used by smart browsers. We need them. Cheers, Doru On Fri, Oct 25, 2013 at 3:18 PM, Christophe Demarey < Christophe.Demarey(a)inria.fr> wrote: > Why do we need to remove configurations of projects loaded into the image? > I would say the opposite: all projects in the image should have their > configuration loaded. It is part of the project. > > Le 25 oct. 2013 Ã 14:00, GitHub a Ã©crit : > > 12005 unload BaselineOfFileTree, BaselineOfMetacello > > https://pharo.fogbugz.com/f/cases/12005 > > > -- www.tudorgirba.com "Every thing has its own flow"

Oct. 25, 2013