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

Re: [Pharo-project] allowUnderscoreAsAssignment broken ?

by Igor Stasenko

On 12 August 2011 20:29, Igor Stasenko <siguctua(a)gmail.com> wrote: > On 12 August 2011 20:16, Mariano Martinez Peck <marianopeck(a)gmail.com> wrote: >> Tried in Pharo 1.4. >> >> Scanner allowUnderscoreAsAssignmentÂ Â ->> true. >> >> I then try to compile: >> >> testUnderscore >> Â Â Â | a | >> Â Â Â a _ 'a'. >> Â Â Â self assert: a equals: 'a' >> >> And I get: >> >> testUnderscore >> Â Â Â | a | >> Â Â Â a _Â Nothing more expected ->'a'. >> Â Â Â self assert: a equals: 'a' >> > > Ouch.. Someone (actually Ben) > has moved initializeTypeTable to String... > > what you are smoking Ben? :) > okay, i can ask more politely: why??? > at least, if you doing refactorings like that, you should also modify other places. There should be no two implementations of initializeTypeTable. We don't need two. >> >> thanks >> >> -- >> Mariano >> http://marianopeck.wordpress.com >> >> > > > > -- > Best regards, > Igor Stasenko AKA sig. > -- Best regards, Igor Stasenko AKA sig.

Aug. 12, 2011

Re: [Pharo-project] allowUnderscoreAsAssignment broken ?

by Igor Stasenko

On 12 August 2011 20:16, Mariano Martinez Peck <marianopeck(a)gmail.com> wrote: > Tried in Pharo 1.4. > > Scanner allowUnderscoreAsAssignmentÂ Â ->> true. > > I then try to compile: > > testUnderscore > Â Â Â | a | > Â Â Â a _ 'a'. > Â Â Â self assert: a equals: 'a' > > And I get: > > testUnderscore > Â Â Â | a | > Â Â Â a _Â Nothing more expected ->'a'. > Â Â Â self assert: a equals: 'a' > Ouch.. Someone (actually Ben) has moved initializeTypeTable to String... what you are smoking Ben? :) okay, i can ask more politely: why??? > > thanks > > -- > Mariano > http://marianopeck.wordpress.com > > -- Best regards, Igor Stasenko AKA sig.

Aug. 12, 2011

[Pharo-project] allowUnderscoreAsAssignment broken ?

by Mariano Martinez Peck

Tried in Pharo 1.4. Scanner allowUnderscoreAsAssignment ->> true. I then try to compile: testUnderscore | a | a _ 'a'. self assert: a equals: 'a' And I get: testUnderscore | a | a _ Nothing more expected ->'a'. self assert: a equals: 'a' thanks -- Mariano http://marianopeck.wordpress.com

Aug. 12, 2011

Re: [Pharo-project] #name

by Pat Maddox

I use #name all over the place and have never had problems. On Aug 12, 2011, at 9:35 AM, "Sean P. DeNigris" <sean(a)clipperadams.com> wrote: > How is #name used by the system? I often want to use it in my classes, but I > don't want to mess with whatever Object is doing with it. There are +++ > senders. > > On the flip side, should the system use a different, more specific, > intention-revealing selector to avoid this confusion? > > Thanks. > Sean > > -- > View this message in context: http://forum.world.st/name-tp3739677p3739677.html > Sent from the Pharo Smalltalk mailing list archive at Nabble.com. >

Aug. 12, 2011

Re: [Pharo-project] some explanations about what we are doing :)

by Igor Stasenko

On 12 August 2011 16:52, Henrik Johansen <henrik.s.johansen(a)veloxit.no> wrote: > > > On 12. aug. 2011, at 15:15, StÃ©phane Ducasse <stephane.ducasse(a)inria.fr> wrote: > >> hi guys >> >> Context: >> So let us face it our attempt to manage a core and another image did not work well. After RPackage I will focus on a process (based on metacello) where >> we take a core + a spec + fixes and it produces a new core + a new spec + a new set of packages. But I need time to do that. >> >> Actions: >> So now we are including several packages in the image so that we do not have two images. >> Â Â FS >> Â Â RB engine >> Â Â Shout >> Â Â OCompletion >> Â Â Collection-Arithmetics >> Â Â SystemAnnouncement >> Â Â Â Â this one is special and some of you may not know it. >> Â Â Â Â It is a package that raise announcement for system changes. >> Â Â Â Â The idea is that we will then >> Â Â Â Â Â Â 1- change all the tools to register to SystemAnnouncement >> Â Â Â Â Â Â 2- replace all the SystemChangeNotification hooks by SystemAnnouncement >> Â Â Â Â Â Â 3- remove our old friend SystemChangeNotifier. >> >> The idea is then to be able to integrate/enhance these packages deeper into the system >> Â Â FS for example to replace the old file package >> >> Stef, Marcus, Igor, Luc, Nick sprinting on a sunday day (yes it happens here too). >> > >> > Ugh. > Collection-Aritmetics was extracted exactly -because- it's nothing that belongs in image by default... I have no idea why it required.. hopefully we could remove users of it and unload it again :) > As for RB, my impression through your conversations with Lukas is that it's not enough to include it, you need someone who understands the architecture for bug fixes/changes that preserve architectural integrity rather than end up as ugly hacks. > > I like the functionality of OCompletion, but from what little I've read of Shout, only it's mother could love it. (same applies to texteditor etc. though, so I guess it's not a very good argument against integrating it) > So true. But first we need to integrate this stuff, and then start working on cleaning it out. > FS + SystemAnnouncements, I'm all for :) > > Cheers, > Henry > -- Best regards, Igor Stasenko AKA sig.

Aug. 12, 2011

Re: [Pharo-project] #name

by Max Leske

+ 1 On 12.08.2011, at 17:35, Sean P. DeNigris wrote: > How is #name used by the system? I often want to use it in my classes, but I > don't want to mess with whatever Object is doing with it. There are +++ > senders. > > On the flip side, should the system use a different, more specific, > intention-revealing selector to avoid this confusion? > > Thanks. > Sean > > -- > View this message in context: http://forum.world.st/name-tp3739677p3739677.html > Sent from the Pharo Smalltalk mailing list archive at Nabble.com. >

Aug. 12, 2011

[Pharo-project] [update 1.4] #14088

by Marcus Denker

14088 ----- Issue 4625: Fix execution of code in Implementor/sender http://code.google.com/p/pharo/issues/detail?id=4625 Issue 4628: Nuking ImageSegment (Part I) http://code.google.com/p/pharo/issues/detail?id=4628 -- Marcus Denker -- http://marcusdenker.de

Aug. 12, 2011

Re: [Pharo-project] I am petrified. I don't know what to say ........... (just look at that)

by Nicolas Cellier

2011/8/12 Nicolas Cellier <nicolas.cellier.aka.nice(a)gmail.com>: > 2011/8/12 StÃ©phane Ducasse <stephane.ducasse(a)inria.fr>: >> You are really lucky. >> One the last dinosaurs.... there are rarer and rarer. >> May be you can improve with a timesRepeat: >> > > Yes, good idea, refactoring with an even worse hack is a good joke, > For example (thisContext asManyTimes: 6) repeat sender > > Object>>asManyTimes: anInteger > Â ^ManyTimesProxy target: self times: anInteger > > ManyTimesProxy>>repeat > Â Â ^RepeatManyTimesProxy target: target times: count > > RepeatManyTimesProxy>>doesNotUnderstand: aMessage > Â count <= 0 ifTrue: [^target]. > Â count := count - 1. > Â target := target perform: aMessage selector with: aMessage arguments. > Â ^self perform: Â aMessage selector with: aMessage arguments > > Nicolas > Oh, maybe we should not just stop at RE-factoring but also use factoring and write it ((thisContext asManyTimes: 3) repeat asManyTimes: 2) repeat sender but I'm unsure of the result... Nicolas >> >> Stef >> >> On Aug 12, 2011, at 5:12 PM, Igor Stasenko wrote: >> >>> DiskProxy>>comeFullyUpOnReload: smartRefStream >>> Â Â Â "Internalize myself into a fully alive object after raw loading from a >>> Â Â Â DataStream. (See my class comment.) DataStream will substitute the >>> Â Â Â object from this eval for the DiskProxy." >>> >>> Â Â Â | globalObj symbol arrayIndex | >>> Â Â Â symbol := globalObjectName. Â Â "See if class is mapped to another name" >>> Â Â Â (smartRefStream respondsTo: #renamed) >>> Â Â Â Â Â Â Â ifTrue: [ >>> Â Â Â Â Â Â Â Â Â Â Â "If in outPointers in an ImageSegment, remember original class >>> Â Â Â Â Â Â Â Â Â Â Â name. >>> Â Â Â Â Â Â Â Â Â Â Â See mapClass:installIn:. Would be lost otherwise." >>>>>>>> Â Â Â Â Â Â Â Â Â (thisContext sender sender sender sender sender sender sender sender receiver class == ImageSegment >>>>>>>> Â Â Â Â Â Â Â Â Â Â Â Â Â and: [ thisContext sender sender sender sender method == (DataStream compiledMethodAt: #readArray) ]) >>> Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â ifTrue: [ >>> Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â arrayIndex := thisContext sender sender sender sender tempAt: >>> 4. Â Â "index var in readArray. Later safer to find i on stack >>> Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â of context." >>> Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â smartRefStream renamedConv at: arrayIndex put: symbol ]. Â Â Â Â "save >>> original name" >>> Â Â Â Â Â Â Â Â Â Â Â symbol := smartRefStream renamed at: symbol ifAbsent: [ symbol ] ]. Â Â "map" >>> Â Â Â globalObj := Smalltalk globals >>> Â Â Â Â Â Â Â at: symbol >>> Â Â Â Â Â Â Â ifAbsent: [ >>> Â Â Â Â Â Â Â Â Â Â Â preSelector == nil & (constructorSelector = #yourself) >>> Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â ifTrue: [ >>> Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Transcript >>> Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â cr; >>> Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â show: symbol , ' is undeclared.'. >>> Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â (Undeclared includesKey: symbol) >>> Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â ifTrue: [ ^ Undeclared at: symbol ]. >>> Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Undeclared at: symbol put: nil. >>> Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â ^ nil ]. >>> Â Â Â Â Â Â Â Â Â Â Â ^ self error: 'Global "' , symbol , '" not found' ]. >>> Â Â Â preSelector >>> Â Â Â Â Â Â Â ifNotNil: [ >>> Â Â Â Â Â Â Â Â Â Â Â Symbol >>> Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â hasInterned: preSelector >>> Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â ifTrue: [ :selector | >>> Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â [ globalObj := globalObj perform: selector ] >>> Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â on: Error >>> Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â do: [ :ex | >>> Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â ex messageText = 'key not found' >>> Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â ifTrue: [ ^ nil ]. >>> Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â ^ ex signal ] ] ]. Â Â Â "keep the Proxy if Project does not exist" >>> Â Â Â constructorSelector ifNil: [ ^ globalObj ]. >>> Â Â Â Symbol >>> Â Â Â Â Â Â Â hasInterned: constructorSelector >>> Â Â Â Â Â Â Â ifTrue: [ :selector | >>> Â Â Â Â Â Â Â Â Â Â Â [ ^ globalObj perform: selector withArguments: constructorArgs ] >>> Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â on: Error >>> Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â do: [ :ex | >>> Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â ex messageText = 'key not found' >>> Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â ifTrue: [ ^ nil ]. >>> Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â ^ ex signal ] ]. Â Â Â Â "args not checked against Renamed" >>> Â Â Â ^ nil >>> >>> -- >>> Best regards, >>> Igor Stasenko AKA sig. >>> >> >> >> >

Aug. 12, 2011

Re: [Pharo-project] I am petrified. I don't know what to say ........... (just look at that)

by StÃ©phane Ducasse

> 2011/8/12 StÃ©phane Ducasse <stephane.ducasse(a)inria.fr>: >> You are really lucky. >> One the last dinosaurs.... there are rarer and rarer. >> May be you can improve with a timesRepeat: >> > > Yes, good idea, refactoring with an even worse hack is a good joke, > For example (thisContext asManyTimes: 6) repeat sender Excellent idea. We will get more compact code :) We had a really good laugh. Pharo will become much more serious in the future so we should chase these dinosaurs now :) > Object>>asManyTimes: anInteger > ^ManyTimesProxy target: self times: anInteger > > ManyTimesProxy>>repeat > ^RepeatManyTimesProxy target: target times: count > > RepeatManyTimesProxy>>doesNotUnderstand: aMessage > count <= 0 ifTrue: [^target]. > count := count - 1. > target := target perform: aMessage selector with: aMessage arguments. > ^self perform: aMessage selector with: aMessage arguments > Nicolas

Aug. 12, 2011

Re: [Pharo-project] Ring question

by Veronica Isabel Uquillas Gomez

On 12 Aug 2011, at 17:12, StÃ©phane Ducasse wrote: > Hi veronica > > How do I create a RGMethodDefinition for a SmalltalkClass > > I used class:selector: but this is wrong because for example for a class > > RGMethodDefinition > class: Point class > selector: #new > > but then I get error when querying className that 'Point class' is not in the SystemDictionary or something like that. indeed that only works for ring classes.. The other option is using the factory RGFactory current createMethodNamed: aSymbol realParent: aClass > > I suggest to add a creation method for realClass: selector: > > what do you think? > it will be handy for tools ( so far i did not need it) gonna add it > Stef

Aug. 12, 2011