February 2021 - Pharo-users

Problem with Dictionary and Associations

by David＠totallyobjects.com

I am using STON to objects out to disk. Up to two days ago, I was reading them in as Dictionaries and converting to objects from there. All of a sudden yesterday morning, I got an error saying that the association is only indexable with integers. Even so, I don't seem to be able to access the contents. Fistly, any ideas why this has changed and secondly, any ideas how to fix it? David Totally Objects Sent from my Huawei tablet

Feb. 10, 2021

Hack to show timestamp and author on versions of methods

by Davide Varvello

Hi Guys In my opinion it is really uncomfortable working without the timestamp of a specific version of a method, I read the explanation of Esteban in the previous thread (http://forum.world.st/pharo-8-method-versions-how-to-see-author-timestamp-t…) but probably it will take some time before to see a new VersionBrowser and the history based on git is not really a solution to me. So it took a day of debugging but I discovered the flow of "versions" in Pharo 8 and I added the logic of the timestamp and the author of a method. Basically I created a subclass of MessageList I named it VersionMessageList, in this way the original MessageList is preserved for other goals I can safely ignore. Now 1) You can file-in VersionMessageList.st (see below) and 2) You have to change VersionBrowser >> initializePresenters using VersionMessageList this way: VersionBrowser >> initializePresenters messageList := self instantiate: VersionMessageList . .... Be careful, it's a hack and it's working on Pharo8, I don't know if it works on Pharo9 Cheers Davide VersionMessageList.st <http://forum.world.st/file/t77843/VersionMessageList.st> -- Sent from: http://forum.world.st/Pharo-Smalltalk-Users-f1310670.html

Feb. 10, 2021

[ANN] Willow v14.1.0 [v14.1.0] released!

by Buenos Aires Smalltalk

Feb. 10, 2021

Re: RB - New introduced refactorings

by StÃ©phane Ducasse

Thanks evelyn for the announce. I will use it for the blog :) > On 8 Feb 2021, at 09:40, Evelyn Cusi Lopez <cusi.evelyn(a)gmail.com> wrote: > > Hello all, > > These weeks I added new refactorings, following I will give some examples of them. > > 1. Create accessors with lazy initialization > > This refactoring create accessors for variables with lazy initialization, for example: > - we have a class > > Object subclass: #SomeClass > instanceVariableNames: 'stringVar' > classVariableNames: '' > package: 'Example' > > - we can create accessors with lazy initialization executing this script: > > (RBCreateAccessorsWithLazyInitializationForVariableRefactoring > variable: #stringVar > class: SomeClass > classVariable: false > defaultValue: '''example''') execute > > the result of this refactoring is: > > SomeClass >> stringVar > ^ stringVar ifNil: [ stringVar := '' ] > > SomeClass >> stringVar: anObject > stringVar := anObject > > 2. Extract setUp refactoring (is the first version by the moment, it need to be improved) > > This refactoring create a setUp method from a code fragment. > You can select an interval of some code in a test method and call this refactoring to create a setUp method implementing that code and replace the code by nothing. The selected class need to be a subclass of TestCase. > > For example: > > - given a class > > TestCase subclass: #SomeTest > instanceVariableNames: '' > classVariableNames: '' > package: 'Example' > > SomeTest >>testFoo > self x. > self assert: true > > SomeTest >> self x > self doSomething > > - To extract setUp method you can execute this script: > > (RBExtractSetUpMethodRefactoring extract: (10 to: 16) > from: #testFoo > in: SomeTest) execute > > the result of this refactoring is: > > SomeTest >> setUp > super setUp. > self x. > > SomeTest >> testFoo > self assert: true > > 3. Move method to class side refactoring (This refactoring is still awaiting review) > > This refactoring move a method to class side. > For example: > - given a method > > ClassX >> annotatedBlockMarkup > ^ '@@' > > - we can execute the refactoring with this script: > > (RBMoveMethodToClassSideRefactoring > method: (ClassX >> #annotatedBlockMarkup) > class: ClassX) execute. > > the result of this refactoring is: > > ClassX >> annotatedBlockMarkup > ^ self class annotatedBlockMarkup > > ClassX class >> annotatedBlockMarkup > ^ '@@' > > Any suggestions or comment of these refactorings are welcome. I hope these new refactorings are useful to you . > > Regards, > Evelyn C. -------------------------------------------- StÃ©phane Ducasse http://stephane.ducasse.free.fr / http://www.pharo.org 03 59 35 87 52 Assistant: Aurore Dalle FAX 03 59 57 78 50 TEL 03 59 35 86 16 S. Ducasse - Inria 40, avenue Halley, Parc Scientifique de la Haute Borne, BÃ¢t.A, Park Plaza Villeneuve d'Ascq 59650 France

Feb. 9, 2021

Magritte+Seaside 3.4.4+Pharo issues

by Sanjay Minni

Hi I am posting these here (rather in desperation) as I seem to be stuck very erratically in Magritte. Magritte appeared to me the next best thing after sliced bread but somehow i just cant get it to toast. trying the simplest of the example - exercise 1 - from the booklet "Meta Data at work with Magritte" - item 4.3 - which is a form rendering CMPerson from the code example library Magritte-ContactManager - i now get errors / walkbacks in the Classes. Yet it seemed working earlier. (PS I have only loaded Pharo 8) - MAInternalEditorComponent - MAFileModelHandler MessageNotUnderstood__receiver_of__requiresMultipartForm__is_nil.pdf <http://forum.world.st/file/t368721/MessageNotUnderstood__receiver_of__requi…> MessageNotUnderstood__MAFileModelHandler___#key_.pdf <http://forum.world.st/file/t368721/MessageNotUnderstood__MAFileModelHandler…> I see it being used in desktop and probably in web (seaside) as well but somehow I am drawing a blank So can someone pls just help me and put any fixes as required (I will do a sprint over 2 weeks to get an app up) - - its a make or break for me ----- cheers, Sanjay -- Sent from: http://forum.world.st/Pharo-Smalltalk-Users-f1310670.html

Feb. 9, 2021

Re: RB - New introduced refactorings

by Tim Mackinnon

Evelyn - this is brillian stuff, I appreciate the work going into this to make it all better. Keep pushing on it. Tim On Mon, 8 Feb 2021, at 7:20 PM, Evelyn Cusi Lopez wrote: > Hi, > > The current move method to class side refactoring, doesn't correct the senders of this method, neither checks correctly references to instance variables or method calls (on instance side). At the moment this improvement only covers that method's senders won't break, creating a method on instance side with the same name referring to the method on side class, as shown in the example. > > - given a method > > ClassX >> annotatedBlockMarkup > ^ '@@' > > > the result of this refactoring is: > > ClassX >> annotatedBlockMarkup > ^ self class annotatedBlockMarkup > > ClassX class >> annotatedBlockMarkup > ^ '@@' > > I still need to improve, but it's a small advance that I wanted to show you to get feedback. > > Regards, > > Evelyn C. > > El lun, 8 feb 2021 a las 14:57, Kasper Osterbye (<kasper.osterbye(a)gmail.com>) escribiÃ³: >> Hi >> >> How does your "move method to class side" differ from the one already there? >> >> image.png >> >> Best, >> >> Kasper > > *Attachments:* > * image.png

Feb. 9, 2021

Re: RB - New introduced refactorings

by Evelyn Cusi Lopez

Hi, The current move method to class side refactoring, doesn't correct the senders of this method, neither checks correctly references to instance variables or method calls (on instance side). At the moment this improvement only covers that method's senders won't break, creating a method on instance side with the same name referring to the method on side class, as shown in the example. - given a method ClassX >> annotatedBlockMarkup ^ '@@' the result of this refactoring is: ClassX >> annotatedBlockMarkup ^ self class annotatedBlockMarkup ClassX class >> annotatedBlockMarkup ^ '@@' I still need to improve, but it's a small advance that I wanted to show you to get feedback. Regards, Evelyn C. El lun, 8 feb 2021 a las 14:57, Kasper Osterbye (<kasper.osterbye(a)gmail.com>) escribiÃ³: > Hi > > How does your "move method to class side" differ from the one already > there? > > [image: image.png] > > Best, > > Kasper >

Feb. 8, 2021

Re: RB - New introduced refactorings

by Kasper Osterbye

Hi How does your "move method to class side" differ from the one already there? [image: image.png] Best, Kasper

Feb. 8, 2021

Re: pharo 8 - method versions - how to see author/timestamp?

by Davide Varvello

Esteban, Is there a hack I can do to put the timestamp in the VersionBrowser? Thanks Davide Esteban Lorenzano wrote > versions are going to stay there, we never even thought on removing that > ;) > they will come from Epicea instead the changes file (because yes, we want > to remove the changes file... but this one is a hard one ;) ). > and yes, they will have a timestamp (which is what the in-image versions > is lacking now). > > Esteban > On Feb 5 2021, at 4:47 pm, Noury Bouraqadi < > bouraqadi@ > > wrote: >> So, will this remain in the image, and the timestamp will be eventually >> fixed? >> >> Thanks -- Sent from: http://forum.world.st/Pharo-Smalltalk-Users-f1310670.html

Feb. 8, 2021

RB - New introduced refactorings

by Evelyn Cusi Lopez

Hello all, These weeks I added new refactorings, following I will give some examples of them. *1. Create accessors with lazy initialization* This refactoring create accessors for variables with lazy initialization, for example: - we have a class Object subclass: #SomeClass instanceVariableNames: 'stringVar' classVariableNames: '' package: 'Example' - we can create accessors with lazy initialization executing this script: (RBCreateAccessorsWithLazyInitializationForVariableRefactoring variable: #stringVar class: SomeClass classVariable: false defaultValue: '''example''') execute the result of this refactoring is: SomeClass >> stringVar ^ stringVar ifNil: [ stringVar := '' ] SomeClass >> stringVar: anObject stringVar := anObject *2. Extract setUp refactoring* (is the first version by the moment, it need to be improved) This refactoring create a setUp method from a code fragment. You can select an interval of some code in a test method and call this refactoring to create a setUp method implementing that code and replace the code by nothing. The selected class need to be a subclass of TestCase. For example: - given a class TestCase subclass: #SomeTest instanceVariableNames: '' classVariableNames: '' package: 'Example' SomeTest >>testFoo self x. self assert: true SomeTest >> self x self doSomething - To extract setUp method you can execute this script: (RBExtractSetUpMethodRefactoring extract: (10 to: 16) from: #testFoo in: SomeTest) execute the result of this refactoring is: SomeTest >> setUp super setUp. self x. SomeTest >> testFoo self assert: true *3. Move method to class side refactoring* (This refactoring is still awaiting review) This refactoring move a method to class side. For example: - given a method ClassX >> annotatedBlockMarkup ^ '@@' - we can execute the refactoring with this script: (RBMoveMethodToClassSideRefactoring method: (ClassX >> #annotatedBlockMarkup) class: ClassX) execute. the result of this refactoring is: ClassX >> annotatedBlockMarkup ^ self class annotatedBlockMarkup ClassX class >> annotatedBlockMarkup ^ '@@' Any suggestions or comment of these refactorings are welcome. I hope these new refactorings are useful to you . Regards, Evelyn C.

Feb. 8, 2021