June 2012 - Pharo-dev - lists.pharo.org

Re: [Pharo-project] ClassBoxes... bitrotted, or usable?

by StÃ©phane Ducasse

On Jun 5, 2012, at 9:52 AM, Jan Vrany wrote: > On 04/06/12 22:56, Alexandre Bergel wrote: >> This was the implementation in Modular Smalltalk if I remember correctly. >> >> In plain Smalltalk you send a message named #msg and the lookup gives you a compile method. >> In Modular Smalltalk, there is an intermediary step: #msg -> #Namespace.msg -> compile method >> > > I see. What I do not see is how to achieve this without a cost at runtime. Given that it should support imports. Without imports, SNs are > pretty much useless :-) I guess that you are saying that you need a look up between selectornamespace because you can have A>String>>foo ^ 666 A>String>>bar ^ self foo => compiled into A>String>>bar ^ self #A.foo B import A B>String>>foo ^ 42 in B 'ababab' foo -> 42 'ababab' bar =>compiled in ababab' #B.bar should still execute 'ababab' #A.bar since it is not redefined Yes I agree with you. Stef

June 5, 2012

Re: [Pharo-project] ClassBoxes... bitrotted, or usable?

by StÃ©phane Ducasse

> >> Jan did you have a look at http://timelessrepo.com/refinements-in-ruby >> Because I would like to know if there are more selector or classbox oriented or something else? >> > Briefly. That post is more a "wow-that's-cool" sort of post rather than in-depth technical report. Looks like classboxes without local > rebinding - that's what they say, after all. Yes I just read it and it looks more like selectorNamespaces but without any explanations.

June 5, 2012

Re: [Pharo-project] Next step in cleaning DataStream

by Mariano Martinez Peck

Guys..the fact of having to make everything deprecated instead of removing it is really a pain. Now ReferenceStream is not removed but deprecated..meaning that it *should* work. So instead of removing everything I have to change it a lot so that it "still" works but deprecated. Can we just assume that ReferenceStream will be removed in 2.0 please? On Tue, Jun 5, 2012 at 1:32 PM, Mariano Martinez Peck <marianopeck(a)gmail.com > wrote: > Ok. Perfect. So let's assume DataStream is used only for Motnicello. I > will rename it to MCDataStream, move it to Monticello Package and see all > what I can remove/clean. > I will let you know. > > Cheers > > > On Tue, Jun 5, 2012 at 12:51 PM, Esteban Lorenzano <estebanlm(a)gmail.com>wrote: > >> +1 >> >> On Jun 4, 2012, at 6:15 PM, Damien Cassou wrote: >> >> > On Mon, Jun 4, 2012 at 6:09 PM, Pavel Krivanek < >> pavel.krivanek(a)gmail.com> wrote: >> >> I think it is a good proposal. Then it can be in separate package or >> as part >> >> of MC package. >> > >> > I would put DataStream and its friends within the monticello package, >> > and prepend MC to their names. Then, you can clean the resulting >> > classes if you want to. >> > >> > -- >> > Damien Cassou >> > http://damiencassou.seasidehosting.st >> > >> > "Lambdas are relegated to relative obscurity until Java makes them >> > popular by not having them." James Iry >> > >> >> >> > > > -- > Mariano > http://marianopeck.wordpress.com > > -- Mariano http://marianopeck.wordpress.com

June 5, 2012

Re: [Pharo-project] About announcements

by Igor Stasenko

On 5 June 2012 06:17, Santiago Bragagnolo <santiagobragagnolo(a)gmail.com> wrote: > Â Hi All! > > Â Im adding an announcer in my project, 'cause, in order to build a call > graph, ill need to know about inferencer-stack movements.Â I was thinking, > the meaning of it is to create an announcement object with it configuration > for any push or pop of a stack that's similar to the runtime stack, these > are a lots lots lots of objects proably with lesser optimization than > BlockClosure. I was thinking in add extension methods that allow me > something like > i am not sure you using right thing for your task. an announcement is an event which you broadcast to subscribers. you can announce any object as announcement.. just make sure your subscription handles it.. i.e. you can use: announcer on: BlockAnnouncement send: #doBlock: to: self if you write: BlockAnnouncement class>>handles: announcement ^ announcement class == BlockClosure Myclass>>doBlock: closure .... do anything you like... x := closure value .. etc So, you can announce like: announcer announce: [ Stack push ] > self announce: [ StackPush new ] ifAnyIsInteresetedIn: StackPush. > > > Well, i have two questions: > > a) Am i right? > b) subscriptions variable at SubscriptionRegistry is an OrderedCollection, > but is commonly accessed to get the subscriptors related with some concrete > type, why not a Dictionary? is not faster than an OrderCollection in the > tipical case of use? Ok, it could complicate calls like 'remove: > subscriptor' but, commonly if you have the subscriptor, you have the > Announcement type. Maybe the common ammount of subscriptors in a announcer > is too little? > > Thanks! -- Best regards, Igor Stasenko.

June 5, 2012

[Pharo-project] [update 2.0] #20112

by Marcus Denker

20112 ----- Issue 6009: Improved ProcessTest http://code.google.com/p/pharo/issues/detail?id=6009 Issue 6023: Enhance MCHttpRepository>>readStreamForFileNamed:do: to require binary data http://code.google.com/p/pharo/issues/detail?id=6023 Issue 6022: Nuke Stream>>isTypeHTTP and Stream>>sleep http://code.google.com/p/pharo/issues/detail?id=6022 -- Marcus Denker -- http://marcusdenker.de

June 5, 2012

[Pharo-project] [update 2.0] #20111

by Marcus Denker

20111 ----- Issue 6027: PackageServices is unused and undocumented http://code.google.com/p/pharo/issues/detail?id=6027 Issue 6028: fix typo in #printPrimitiveOn: http://code.google.com/p/pharo/issues/detail?id=6028 Issue 6026: New version of Spec 0.4 http://code.google.com/p/pharo/issues/detail?id=6026 -- Marcus Denker -- http://marcusdenker.de

June 5, 2012

Re: [Pharo-project] About announcements

by Santiago Bragagnolo

yes yes, the dispatching is done :3, i used it at least in 3 times, and works great! the meaning of the self announce: [ StackPush new ] ifAnyIsInteresetedIn: StackPush. is something like, self announce: aBlock ifAnyIsInteresetedIn: anAnnouncementSubtype. then sub := subscriptors interestedIn: anAnnouncementSubype sub size > 0 ifTrue: [ realAnnounce := aBlock value. ] The thing is, with stack announcements i'll have at least the double of announcement instantiation than message sents. (push & pop). There are a lot of announcements in a simple expression. It could be useful to get lazy instantiation of the announcement? Yep, announcement -> subscriptions 2012/6/5 Esteban Lorenzano <estebanlm(a)gmail.com> > Hi, > > On Jun 5, 2012, at 6:17 AM, Santiago Bragagnolo wrote: > > > Hi All! > > > > Im adding an announcer in my project, 'cause, in order to build a call > graph, ill need to know about inferencer-stack movements. I was thinking, > the meaning of it is to create an announcement object with it configuration > for any push or pop of a stack that's similar to the runtime stack, these > are a lots lots lots of objects proably with lesser optimization than > BlockClosure. I was thinking in add extension methods that allow me > something like > > > > self announce: [ StackPush new ] ifAnyIsInteresetedIn: StackPush. > > > > > > Well, i have two questions: > > > > a) Am i right? > > No.. dispatch of events to those who are interested is already done. > when you suscribe using #on: anAnnouncement do: ..., or #on: > anAnnouncement send: ... to: ... you are already declaring "I'm interested > on anAnnouncement", and the dispatcher delivers appropriately. > > > b) subscriptions variable at SubscriptionRegistry is an > OrderedCollection, but is commonly accessed to get the subscriptors related > with some concrete type, why not a Dictionary? is not faster than an > OrderCollection in the tipical case of use? Ok, it could complicate calls > like 'remove: subscriptor' but, commonly if you have the subscriptor, you > have the Announcement type. Maybe the common ammount of subscriptors in a > announcer is too little? > > a dictionary of what? announcement->suscriptions? > > Esteban > > > > > Thanks! > > > > > > > > >

June 5, 2012

Re: [Pharo-project] [update 2.0] #20110

by Nicolas Cellier

2012/6/5 Sven Van Caekenberghe <sven(a)beta9.be>: > Haha, very funny. > > It's a good thing there are so many jokes hidden inside the image. > > We could add a couple/many more: > > self deny: 1 = 2. > self deny: 1 = -1. > self deny: (4 / 2) = 3 > self shouldnt: [ 1 / 2 ] raise: ZeroDivide > > ;-) > Yes it's laughable. However, I recently hacked Intger/Fraction to speed up mixed operations and guess what... once you know the miserable error's we can make... Nicolas > On 05 Jun 2012, at 12:10, Serge Stinckwich wrote: > >> On Tue, Jun 5, 2012 at 5:04 PM, Damien Cassou <damien.cassou(a)gmail.com> wrote: >>> On Tue, Jun 5, 2012 at 11:55 AM, Camillo Bruni <camillobruni(a)gmail.com> wrote: >>>>> testGeneralInquiries >>>>> Â Â Â | now d t dt | >>>>> >>>>> Â Â Â now Â := self timeClass dateAndTimeNow. >>>>> Â Â Â self >>>>> Â Â Â Â Â Â Â assert: now size = 2; >>>>> Â Â Â Â Â Â Â assert: now last <= self timeClass now. >>>>> >>>>> Â Â Â self should: [ self timeClass timeWords ] raise: MessageNotUnderstood. >>>>> >>>>> Â Â Â d := '2 June 1973' asDate. >>>>> Â Â Â t := '4:02:47 am' asTime. >>>>> Â Â Â dt := self timeClass dateAndTimeFromSeconds: (2285280000 + 14567). >>>>> Â Â Â self >>>>> Â Â Â Â Â Â Â assert: dt = {d. t.}. >>>>> >>>>> >>>>> The middle part is a GREAT example of a useful unit testâ¦ >>>> >>>> that's called a negative test :D, assume what possible could fail in the universe and test for it :D >>> >>> I will write a paper on automatic negative test generation. Who wants >>> to join? :-) >> >> You can even write a program that write a paper about negative tests >> generation ;-) >> >> -- >> Serge Stinckwich >> UMI UMMISCO 209 (IRD/UPMC), Hanoi, Vietnam >> Every DSL ends up being Smalltalk >> http://doesnotunderstand.org/ >> > >

June 5, 2012

[Pharo-project] IWST'12 Call for papers

by Alain Plantec

Dear Smalltalkers, We apologize for multiple reception of this announcement. The International Workshop on Smalltalk Technologies, is a European Smalltalk User Group (ESUG) Conference joint event, that focuses on research activities - namely academic creative work undertaken on smalltalk use, and more generally on object technologies. IWST was launched in 2009, in Brest, during the 17th ESUG Conference. The second edition took place in Barcelona, and the third edition was held in 2011 at the heart of historic Edinburgh. Next edition will be in Gent, Begium this summer (28th of august). ESUG gathers groups of professionals and hobbyists who share an interest in the Smalltalk programming languages and related technologies. The goal of the workshop is to create a forum around advances or experience in Smalltalk. IWST contributes to triggering discussions and exchanges of ideas. As always, we make a point to have a very constructive paper review process that aims to provide in-depth comments and suggestions to all of them (no one-liners!) This makes it ideal to disseminate new research directions, and is a good venue for early stage PhD students to receive feedback on their work. In addition, we intent to invite best papers to a journal special issue as we did last year. Looking forward to your submissions, and to meeting you in Gent! Best regards, LoÃ¯c Lagadec& Alain Plantec Co-chairs of IWST International Workshop on Smalltalk Technologies 2012 August 28th, 2012 Gent, Belgium. ESUG 2012 Smalltalk joint event Important dates Submission deadline: June 15, 2012 Notification deadline: July 15, 2012 All accepted papers will be published in ACM DL, and the authors of the best papers will be invited to submit an extended version to a journal special issue (To Be Confirmed). Goals and scopes The goals of the workshop is to create a forum around advances or experience in Smalltalk and to trigger discussions and exchanges of ideas. Participants are invited to submit research articles. We will not enforce any length restriction. However we expect papers of two kinds: Short position papers describing emerging ideas. Long research papers with deeper description of experiments and of research results. Topics We welcome contributions on all aspects, theoretical as well as practical, of Smalltalk related topics such as: Aspect-oriented programming, Design patterns, Experience reports, Frameworks, Implementation, new dialects or languages implemented in Smalltalk, Interaction with other languages, Meta-programming and Meta-modeling, Tools Links home page:http://www.esug.org/wiki/pier/Conferences/2012/International-Workshop-… submission:http://www.easychair.org/conferences/?conf=iwst2012 Publication Both submissions and final papers must be prepared using the ACM SIGPLAN 10 point format. Templates for Word and LaTeX are available at http://www.acm.org/sigs/sigplan/authorInformation.htm. This site also contains links to useful informations on how to write effective submissions. Program chairs LoÃ¯c Lagadec and Alain Plantec (LabSticc CACS/CNRS, University of Brest, France) Program committee Gabriela Arevalo Universidad Nacional de Quilmes, Argentina Alexandre Bergel University of Chile Andrew P. Black Portland State University, US Marcus Denker Rmod, INRIA Lille - Nord Europe, France Luc Fabresse Ecole des Mines de Douai, France, Tudor Girba CompuGroup Medical Schweiz, Switzerland Andy Kellens Software Languages Lab, Vrije Universiteit Brussel, Belgium MickaÃ«l Kerboeuf LabSticc, University of Brest, France Jannik Laval LaBRI, University of Bordeaux, France Mariano Martinez Peck Ecole des Mines de Douai, France, Lukas Renggli Google, Switzerland Jorge Ressia Software Composition Group, University of Bern, Switzerland Bastian Steinert HPI, Software Architecture Group, Germany Hernan Wilkinson 10Pines, IT consultancy, Buenos Aires, Argentina Roel Wuyts IMEC Leuven, Belgium

June 5, 2012

[Pharo-project] About Seed, Hazelnut and related

by Edgar J. De Cleene

>From the excellent blog http://playingwithobjects.wordpress.com/ Guille was doing we read The results of the project so far are: * It can create an Smalltalk image living inside another image. * This inner/guest image can be written in a new .image file. * With this approach a small kernel of 1.1MB has been reached. * SpaceTally runs and prints reports to understand how the space is distributed among objects. * A tool to detect every uninitialized class variable/class instance variable and references to unexistant globals was developed. I like to know exact steps for doing this, so maybe can feedback some.. Guille: Buenisimo lo que estas haciendo. Y como ya los tenemos a Esteban, Igor, Santiago , te esperamos http://ar.groups.yahoo.com/group/squeakRos/ Edgar

June 5, 2012