- Pharo-dev - lists.pharo.org

Re: [Pharo-project] What are MethodDictionary faults?

by Mariano Martinez Peck

On Thu, Dec 24, 2009 at 11:46 AM, StÃ©phane Ducasse < stephane.ducasse(a)inria.fr> wrote: > > What's MW ? > > MethodWrapper. But you cannot wrap the complete system like do: (if I > remember correctly). > > ok > > Yes, I noticed that too. Is very probable that messages very general that > are used a lot will bring the class again. > > Yes the problem is to identify them. > We tried a bit not too much with mathieu but was not obvious > > yes haha > > That's the idea of the tracer. WIth this tracer you *should* be able to > see the messages chain that makes you bring the class (why it was used). But > I don't know if this stuff is working, as you said. > > which tracer? > > Suppose you evaluate this steps: Smalltalk at: #MDFaultDict put: Dictionary new. ImageSegment discoverActiveClasses .... do something here, browse, whatever .... ImageSegment activeClasses Here, you can change activeClasses to use recoverFromMDFaultWithTrace instead of recoverFromMDFault and with this you can see who used that class. It is stored in Smalltalk at: #MDFaultDict Ok...at least that is what I understood, but as this is not working, I am not sure. What I suggest: > - create a new category in the class and move all the methods that > could be removed. Tag them as deprecated. - in imageSegments just clean :) > > Ok. Perfect. Cheers Mariano Stef > _______________________________________________ > Pharo-project mailing list > Pharo-project(a)lists.gforge.inria.fr > http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project >

Dec. 24, 2009

[Pharo-project] Fwd: [squeak-dev] The Trunk: Traits-nice.248.mcz

by Nicolas Cellier

For Trait experts only. ---------- Forwarded message ---------- From: <commits(a)source.squeak.org> Date: 2009/12/24 Subject: [squeak-dev] The Trunk: Traits-nice.248.mcz To: squeak-dev(a)lists.squeakfoundation.org, packages(a)lists.squeakfoundation.org Nicolas Cellier uploaded a new version of Traits to project The Trunk: http://source.squeak.org/trunk/Traits-nice.248.mcz ==================== Summary ==================== Name: Traits-nice.248 Author: nice Time: 24 December 2009, 11:49:10 am UUID: 557593d7-1db7-2347-bf40-2d40e3b91f55 Ancestors: Traits-nice.247 Move FixedIdentitySet off the Array hierarchy. Provide a fast implementation using MethodDictionary tricks - handles collisions (instead of blindly ignoring the entry) - eventually grow. I did not understand previous design decision... The conflict just did happen (I put a halt: and caught one in Object...) According to my own scale, make it work > make it fast. Rationale about the new design: #grow costs, but I think it is user responsibility to fix a reasonnable capacity. collisions handling should not cost much (except above 4096 entries) If any expert knowing the reasons for this class and knowing how to fire the profiling tests could have a look, thanks... =============== Diff against Traits-nice.247 =============== Item was changed: + Collection variableSubclass: #FixedIdentitySet + Â Â Â instanceVariableNames: 'tally capacity hashShift' - Array variableSubclass: #FixedIdentitySet - Â Â Â instanceVariableNames: 'tally capacity' Â Â Â Â classVariableNames: '' Â Â Â Â poolDictionaries: '' Â Â Â Â category: 'Traits-Requires'! + !FixedIdentitySet commentStamp: 'nice 12/24/2009 11:46' prior: 0! + This is a fast implementation of fixed size identity sets. + Same algorithm as MethodDictionary are used, and thus FixedIdentitySet is to IdentitySet what MethodDictionary is to IdentityDictionary. + The main features are: + 1) do not use an array instance variable so as to fast-up creation and every access + 2) due to the fixed allocated size, growing costs an expensive #become: operation. Preallocate me with care. + 3) my size is a power of two so the the hashing algorithm be most efficient. + 4) for maximum random access efficiency, at least half the storage area is always kept empty - !FixedIdentitySet commentStamp: 'NS 5/26/2005 13:00' prior: 0! - This is a fast but lazy implementation of fixed size identity sets. The two main difference to regular identity sets are: - - 1) These identity sets have a fixed size. If they are full, adding another element doesn't have any effect. - 2) No rehashing. If two elements were to be stored on the same position in the underlying array, one of them is simply discarded. + Unlike MethodDictionary, this class will scale a bit better over the 4096 basicSize limit inherent to identityHash, thanks to a proper bitShift.! - As a consequence of (1) and (2), these identity sets are very fast!! Note that this class inherits form Array. This is not clean but reduces memory overhead when instances are created.! Item was added: + ----- Method: FixedIdentitySet>>removeAll (in category 'removing') ----- + removeAll + Â Â Â tally = 0 ifTrue: [^self]. + Â Â Â 1 to: self basicSize do: [:i | self basicAt: i put: nil]. + Â Â Â tally := 0! Item was changed: + ----- Method: FixedIdentitySet>>add: (in category 'adding') ----- - ----- Method: FixedIdentitySet>>add: (in category 'accessing') ----- Â add: anObject + Â Â Â | index | + Â Â Â index := self scanFor: anObject. + Â Â Â (self basicAt: index) + Â Â Â Â Â Â Â ifNil: [ + Â Â Â Â Â Â Â Â Â Â Â self basicAt: index put: anObject. + Â Â Â Â Â Â Â Â Â Â Â tally := tally + 1. + Â Â Â Â Â Â Â Â Â Â Â self isFull ifTrue: [ self grow ]] + Â Â Â Â Â Â Â "ifNotNil: [] already inside". + Â Â Â ^anObject! - Â Â Â | index old | - Â Â Â self isFull ifTrue: [^ false]. - Â Â Â index := self indexOf: anObject. - Â Â Â old := self basicAt: index. - Â Â Â old == anObject ifTrue: [^ true]. - Â Â Â old ifNotNil: [^ false]. - Â Â Â self basicAt: index put: anObject. - Â Â Â tally := tally + 1. - Â Â Â ^ true! Item was changed: + ----- Method: FixedIdentitySet>>addAll:notIn: (in category 'adding') ----- - ----- Method: FixedIdentitySet>>addAll:notIn: (in category 'accessing') ----- Â addAll: aCollection notIn: notCollection Â Â Â Â aCollection do: [:each | - Â Â Â Â Â Â Â self isFull ifTrue: [^ self]. Â Â Â Â Â Â Â Â (notCollection includes: each) ifFalse: [self add: each]. Â Â Â Â ].! Item was changed: Â ----- Method: FixedIdentitySet class>>with:with:with:with:with: (in category 'instance creation') ----- Â with: firstObject with: secondObject with: thirdObject with: fourthObject with: fifthObject Â Â Â Â "Answer an instance of me, containing the five arguments as the elements." + Â Â Â ^ (self new: 5) - Â Â Â ^ self new Â Â Â Â Â Â Â Â add: firstObject; Â Â Â Â Â Â Â Â add: secondObject; Â Â Â Â Â Â Â Â add: thirdObject; Â Â Â Â Â Â Â Â add: fourthObject; Â Â Â Â Â Â Â Â add: fifthObject; Â Â Â Â Â Â Â Â yourself! Item was changed: Â ----- Method: FixedIdentitySet class>>with:with:with:with:with:with: (in category 'instance creation') ----- Â with: firstObject with: secondObject with: thirdObject with: fourthObject with: fifthObject with: sixthObject Â Â Â Â "Answer an instance of me, containing the six arguments as the elements." + Â Â Â ^ (self new: 6) - Â Â Â ^ self new Â Â Â Â Â Â Â Â add: firstObject; Â Â Â Â Â Â Â Â add: secondObject; Â Â Â Â Â Â Â Â add: thirdObject; Â Â Â Â Â Â Â Â add: fourthObject; Â Â Â Â Â Â Â Â add: fifthObject; Â Â Â Â Â Â Â Â add: sixthObject; Â Â Â Â Â Â Â Â yourself! Item was changed: + ----- Method: FixedIdentitySet class>>new (in category 'instance creation') ----- - ----- Method: FixedIdentitySet class>>new (in category 'constants') ----- Â new Â Â Â Â ^ self new: self defaultSize! Item was added: + ----- Method: FixedIdentitySet>>grow (in category 'private') ----- + grow + Â Â Â | newSelf | + Â Â Â newSelf := self species new: capacity * 2. Â "This will double the capacity" + Â Â Â self do: [ :anObject | newSelf add: anObject ]. + Â Â Â self become: newSelf! Item was changed: + ----- Method: FixedIdentitySet>>includes: (in category 'testing') ----- + includes: aSymbol + Â Â Â "This override assumes that pointsTo is a fast primitive" + + Â Â Â aSymbol ifNil: [^ false]. + Â Â Â ^ self pointsTo: aSymbol! - ----- Method: FixedIdentitySet>>includes: (in category 'accessing') ----- - includes: anObject - Â Â Â ^ (self basicAt: (self indexOf: anObject)) == anObject! Item was added: + ----- Method: FixedIdentitySet>>fixCollisionsFrom: (in category 'private') ----- + fixCollisionsFrom: start + Â Â Â "The element at start has been removed and replaced by nil. + Â Â Â This method moves forward from there, relocating any entries + Â Â Â that had been placed below due to collisions with this one." + + Â Â Â | key index mask | + Â Â Â index := start. + Â Â Â mask := self basicSize - 1. + Â Â Â [ (key := self basicAt: (index := (index bitAnd: mask) + 1)) == nil ] whileFalse: [ + Â Â Â Â Â Â Â | newIndex | + Â Â Â Â Â Â Â (newIndex := self scanFor: key) = index ifFalse: [ + Â Â Â Â Â Â Â Â Â Â Â | element | + Â Â Â Â Â Â Â Â Â Â Â element := self basicAt: index. + Â Â Â Â Â Â Â Â Â Â Â self basicAt: index put: (self basicAt: newIndex). + Â Â Â Â Â Â Â Â Â Â Â self basicAt: newIndex put: element.] ]! Item was changed: + ----- Method: FixedIdentitySet class>>new: (in category 'instance creation') ----- - ----- Method: FixedIdentitySet class>>new: (in category 'constants') ----- Â new: anInteger + Â Â Â ^ (self basicNew: (self arraySizeForCapacity: anInteger)) initializeCapacity: anInteger! - Â Â Â ^ (super new: (self arraySizeForCapacity: anInteger)) initializeCapacity: anInteger! Item was changed: + ----- Method: FixedIdentitySet>>remove:ifAbsent: (in category 'removing') ----- - ----- Method: FixedIdentitySet>>remove:ifAbsent: (in category 'accessing') ----- Â remove: anObject ifAbsent: aBlock + Â Â Â | index element | + Â Â Â index := self scanFor: anObject. + Â Â Â (element := self basicAt: index) ifNil: [ ^aBlock value ]. + Â Â Â self basicAt: index put: nil. + Â Â Â tally := tally - 1. + Â Â Â self fixCollisionsFrom: index. + Â Â Â ^element! - Â Â Â | index | - Â Â Â index := self indexOf: anObject. - Â Â Â ^ (self basicAt: index) == anObject - Â Â Â Â Â Â Â ifTrue: [self basicAt: index put: nil. tally := tally - 1. anObject] - Â Â Â Â Â Â Â ifFalse: [aBlock value].! Item was added: + ----- Method: FixedIdentitySet>>rehash (in category 'private') ----- + rehash + Â Â Â | newSelf | + Â Â Â newSelf := self species new: self size. + Â Â Â self do: [ :anObject | newSelf add: anObject ]. + Â Â Â ^newSelf! Item was changed: Â ----- Method: FixedIdentitySet>>initializeCapacity: (in category 'initialize-release') ----- Â initializeCapacity: anInteger Â Â Â Â tally := 0. + Â Â Â capacity := anInteger. + Â Â Â hashShift := self basicSize highBit - 4096 highBit max: 0! - Â Â Â capacity := anInteger.! Item was changed: + ----- Method: FixedIdentitySet class>>arraySizeForCapacity: (in category 'private') ----- - ----- Method: FixedIdentitySet class>>arraySizeForCapacity: (in category 'constants') ----- Â arraySizeForCapacity: anInteger Â Â Â Â "Because of the hash performance, the array size is always a power of 2 Â Â Â Â and at least twice as big as the capacity anInteger" Â Â Â Â ^ anInteger <= 0 Â Â Â Â Â Â Â Â ifTrue: [0] Â Â Â Â Â Â Â Â ifFalse: [1 << (anInteger << 1 - 1) highBit].! Item was added: + ----- Method: FixedIdentitySet>>scanFor: (in category 'private') ----- + scanFor: anObject + Â Â Â "Scan the key array for the first slot containing either a nil (indicating an empty slot) or an element that matches anObject. Answer the index of that slot or raise an error if no slot is found. This method will be overridden in various subclasses that have different interpretations for matching elements." + + Â Â Â | index start mask | + Â Â Â anObject ifNil: [self error: 'This class collection cannot handle nil as an element']. + Â Â Â mask := self basicSize - 1. + Â Â Â index := start := ((anObject identityHash bitShift: hashShift) bitAnd: mask) + 1. + Â Â Â [ + Â Â Â Â Â Â Â | element | + Â Â Â Â Â Â Â ((element := self basicAt: index) == nil or: [ element == anObject ]) + Â Â Â Â Â Â Â Â Â Â Â ifTrue: [ ^index ]. + Â Â Â Â Â Â Â (index := (index bitAnd: mask) + 1) = start ] whileFalse. + Â Â Â self errorNoFreeSpace! Item was removed: - ----- Method: FixedIdentitySet>>destructiveAdd: (in category 'accessing') ----- - destructiveAdd: anObject - Â Â Â | index old | - Â Â Â self isFull ifTrue: [^ false]. - Â Â Â index := self indexOf: anObject. - Â Â Â old := self basicAt: index. - Â Â Â self basicAt: index put: anObject. - Â Â Â old ifNil: [tally := tally + 1]. - Â Â Â ^ true! Item was removed: - ----- Method: FixedIdentitySet>>notFull (in category 'testing') ----- - notFull - Â Â Â ^ tally < capacity! Item was removed: - ----- Method: FixedIdentitySet>>addAll: (in category 'accessing') ----- - addAll: aCollection - Â Â Â aCollection do: [:each | - Â Â Â Â Â Â Â self isFull ifTrue: [^ self]. - Â Â Â Â Â Â Â self add: each. - Â Â Â ].! Item was removed: - ----- Method: FixedIdentitySet>>indexOf: (in category 'private') ----- - indexOf: anObject - Â Â Â anObject isNil ifTrue: [self error: 'This class collection cannot handle nil as an element']. - Â Â Â ^ (anObject identityHash bitAnd: self basicSize - 1) + 1!

Dec. 24, 2009

Re: [Pharo-project] What are MethodDictionary faults?

by StÃ©phane Ducasse

> What's MW ? MethodWrapper. But you cannot wrap the complete system like do: (if I remember correctly). > Yes, I noticed that too. Is very probable that messages very general that are used a lot will bring the class again. Yes the problem is to identify them. We tried a bit not too much with mathieu but was not obvious > That's the idea of the tracer. WIth this tracer you *should* be able to see the messages chain that makes you bring the class (why it was used). But I don't know if this stuff is working, as you said. which tracer? What I suggest: - create a new category in the class and move all the methods that could be removed. Tag them as deprecated. - in imageSegments just clean :) Stef

Dec. 24, 2009

Re: [Pharo-project] New pharo images

by Alain Plantec

StÃ©phane Ducasse a Ã©crit : >> 2009/12/24 StÃ©phane Ducasse <stephane.ducasse(a)inria.fr>: >> >>> Damien in 1.1 may be we should put the nice little font of Cuis on :) >>> >> Tell me what I should do. >> > > We should check the code of settings to see how the preferences is done > when we press the launch button on the side of the install bitmap DejaVu fonts > I've checked, we have "bitmap dejavu sans" with sizes 7, 9 and 12. Alain

Dec. 24, 2009

Re: [Pharo-project] What are MethodDictionary faults?

by Mariano Martinez Peck

Adrian: Thanks, indeed cannotInterpret: was what I was looking for :) On Wed, Dec 23, 2009 at 5:19 PM, StÃ©phane Ducasse <stephane.ducasse(a)inria.fr > wrote: > Ok this is in the image I did not remember (getting old) :) but the vm pay > attention > not to crash totally when you put nil in a methodDict. > > ok > Now what I know is that a while back when we activate > Smalltalk at: #MDFaultDict put: Dictionary new. > We kind of either crashed or got all the classes of the system in. > > I have the impression that this is just a hack and I would remove it from > the system > because we can do with MW or reflectivity (even if we probably cannot trace > the tracing code). > What's MW ? Now...the question is, suppose I remove all the methods like discoverUnusedCLasses, activeClasses, and so on, can I remove also the methods induceMDFault, recoverFromMDFault, etc ??? Or there are other users? As I understood, there aren't. In addition, I don't see the relation between cannotInterpret: and those methods, because cannotInerpret: does not call this methods to recover. > Also the granularity is really the class so as soon as one method is > invoked you get all the class back. > Yes, I noticed that too. Is very probable that messages very general that are used a lot will bring the class again. > So with mathieu we simply something could not understand why all the > classes would be touched > may be browser was scanning all the image. No idea. but at the end it was > not useful. > That's the idea of the tracer. WIth this tracer you *should* be able to see the messages chain that makes you bring the class (why it was used). But I don't know if this stuff is working, as you said. > > Stef > > On Dec 23, 2009, at 4:51 PM, Adrian Kuhn wrote: > > > Mariano Martinez Peck <marianopeck@...> writes: > > > >> So, it seems to HACK (it forces a fault) this MD fault recover to > discover > >> inactive classes and other things. Ok, forget that hack for a moment. > >> > >> What questions is, what are these faults? Why a method dictionary can > be in > >> nil ? How is this produced ? And how is it fixed? > >> > >> I saw for example this: > >> > >> Behaviour>>methodDict > >> methodDict == nil ifTrue: [self recoverFromMDFaultWithTrace]. > >> ^ methodDict > >> > >> But again...how can that be nil ??? > > > > I guess your missing link is > > > > ProtoObject >> #cannotInterpret: > > > > which is executed by the VM whenever you try to send a message to an > object > > whose class does not have a method dictionary. Further methods that you > should > > look at are > > > > ClassDescription >> #induceMDFault > > ClassDescription >> #recoverFromMDFault > > ClassDescription >> #recoverFromMDFaultWithTrace > > > > and their senders. Except for the call to #cannotInterpret by the VM, all > the > > rest is done in the image (stef got that wrong) and you can go and read > the > > code. All in all, it is a highly sophisticated (read *ugly*) hack and I'd > say > > an evil use of superpowers. > > > > --AA > > > > > > > > > > > > > > _______________________________________________ > > Pharo-project mailing list > > Pharo-project(a)lists.gforge.inria.fr > > http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project > > > _______________________________________________ > Pharo-project mailing list > Pharo-project(a)lists.gforge.inria.fr > http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project >

Dec. 24, 2009

Re: [Pharo-project] SUnit forked, gofer it

by Mariano Martinez Peck

hahahah did I convinced you ? Thank you very much. This was my little christmas gift. I put the issue as closed. Cheers Mariano On Thu, Dec 24, 2009 at 10:56 AM, Lukas Renggli <renggli(a)gmail.com> wrote: > Changed in > > Name: OB-SUnitIntegration-lr.16 > Author: lr > Time: 24 December 2009, 10:54:56 am > UUID: 661b0ef3-03df-4193-84ed-e482c9baaec6 > Ancestors: OB-SUnitIntegration-AdrianKuhn.15 > > - run tests depending on selection > > 2009/12/24 Mariano Martinez Peck <marianopeck(a)gmail.com>: > > > > > > On Thu, Dec 24, 2009 at 9:52 AM, Lukas Renggli <renggli(a)gmail.com> > wrote: > >> > >> Indeed, it aways runs all tests of a class. Should that be changed? I > > > > Yes!!!! If you have selected a particular method, the idea is to run > THAT > > method, not all. Why you need to select then ? > > If you want all methods go to the class or even to the tests runner. > > > >> > >> found it more useful to run them all, instead of just one. > >> > >> Lukas > >> > >> 2009/12/24 Adrian Kuhn <akuhn(a)iam.unibe.ch>: > >> > Tudor Girba <tudor.girba@...> writes: > >> > > >> >> It should work if your mouse is over the methods pane. > >> > > >> > Not true in OB. > >> > > >> > > >> > -- > >> > 4th Workshop on Dynamic Languages and Applications > >> > Submit papers by March 31, 2010. http://bit.ly/dyla2010 > >> > > >> > > >> > _______________________________________________ > >> > Pharo-project mailing list > >> > Pharo-project(a)lists.gforge.inria.fr > >> > http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project > >> > > >> > >> > >> > >> -- > >> Lukas Renggli > >> http://www.lukas-renggli.ch > >> > >> _______________________________________________ > >> Pharo-project mailing list > >> Pharo-project(a)lists.gforge.inria.fr > >> http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project > > > > > > _______________________________________________ > > Pharo-project mailing list > > Pharo-project(a)lists.gforge.inria.fr > > http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project > > > > > > -- > Lukas Renggli > http://www.lukas-renggli.ch > > _______________________________________________ > Pharo-project mailing list > Pharo-project(a)lists.gforge.inria.fr > http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project >

Dec. 24, 2009

Re: [Pharo-project] SUnit forked, gofer it

by Lukas Renggli

Changed in Name: OB-SUnitIntegration-lr.16 Author: lr Time: 24 December 2009, 10:54:56 am UUID: 661b0ef3-03df-4193-84ed-e482c9baaec6 Ancestors: OB-SUnitIntegration-AdrianKuhn.15 - run tests depending on selection 2009/12/24 Mariano Martinez Peck <marianopeck(a)gmail.com>: > > > On Thu, Dec 24, 2009 at 9:52 AM, Lukas Renggli <renggli(a)gmail.com> wrote: >> >> Indeed, it aways runs all tests of a class. Should that be changed? I > > Yes!!!!Â Â If you have selected a particular method, the idea is to run THAT > method, not all. Why you need to select then ? > If you want all methods go to the class or even to the tests runner. > >> >> found it more useful to run them all, instead of just one. >> >> Lukas >> >> 2009/12/24 Adrian Kuhn <akuhn(a)iam.unibe.ch>: >> > Tudor Girba <tudor.girba@...> writes: >> > >> >> It should work if your mouse is over the methods pane. >> > >> > Not true in OB. >> > >> > >> > -- >> > 4th Workshop on Dynamic Languages and Applications >> > Submit papers by March 31, 2010. http://bit.ly/dyla2010 >> > >> > >> > _______________________________________________ >> > Pharo-project mailing list >> > Pharo-project(a)lists.gforge.inria.fr >> > http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project >> > >> >> >> >> -- >> Lukas Renggli >> http://www.lukas-renggli.ch >> >> _______________________________________________ >> Pharo-project mailing list >> Pharo-project(a)lists.gforge.inria.fr >> http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project > > > _______________________________________________ > Pharo-project mailing list > Pharo-project(a)lists.gforge.inria.fr > http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project > -- Lukas Renggli http://www.lukas-renggli.ch

Dec. 24, 2009

Re: [Pharo-project] SUnit forked, gofer it

by Mariano Martinez Peck

On Thu, Dec 24, 2009 at 9:52 AM, Lukas Renggli <renggli(a)gmail.com> wrote: > Indeed, it aways runs all tests of a class. Should that be changed? I > Yes!!!! If you have selected a particular method, the idea is to run THAT method, not all. Why you need to select then ? If you want all methods go to the class or even to the tests runner. > found it more useful to run them all, instead of just one. > > Lukas > > 2009/12/24 Adrian Kuhn <akuhn(a)iam.unibe.ch>: > > Tudor Girba <tudor.girba@...> writes: > > > >> It should work if your mouse is over the methods pane. > > > > Not true in OB. > > > > > > -- > > 4th Workshop on Dynamic Languages and Applications > > Submit papers by March 31, 2010. http://bit.ly/dyla2010 > > > > > > _______________________________________________ > > Pharo-project mailing list > > Pharo-project(a)lists.gforge.inria.fr > > http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project > > > > > > -- > Lukas Renggli > http://www.lukas-renggli.ch > > _______________________________________________ > Pharo-project mailing list > Pharo-project(a)lists.gforge.inria.fr > http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project >

Dec. 24, 2009

Re: [Pharo-project] Fwd: sort: and sortBlock:

by StÃ©phane Ducasse

> > sorted is bad name Now given the situation and the result of the discussion we could have it in addition to sort sorted -> always a copy sort -> inplace when possible on SortedCollection, OrderedCollection and others. Stef

Dec. 24, 2009

Re: [Pharo-project] SUnit forked, gofer it

by StÃ©phane Ducasse

Normally it works to click on the class or package to run all the test of the packages now may be the binding does not do it but pressing on the icon was working like that On Dec 24, 2009, at 10:16 AM, Adrian Kuhn wrote: > Lukas Renggli <renggli@...> writes: > >> Indeed, it always runs all tests of a class. Should that be changed? I >> found it more useful to run them all, instead of just one. > > Fixed it in my image (and attached on google). > > Its awesome when you work on one test in particular.It is somehow annoying > when you browse a package ... click click click ... dive into one method > and want to run the whole test case. But one might get used to it quickly. > > Maybe different action for Cmd+T in the source editor and the method list? > (or is that too confusing?) > > --AA > > PS: also broken was that when you ran one method the history of all other > methods was lost. Fixed in akuhn/SUnit now. > > > _______________________________________________ > Pharo-project mailing list > Pharo-project(a)lists.gforge.inria.fr > http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project

Dec. 24, 2009