- Pharo-dev - lists.pharo.org

Re: [Pharo-project] Problem with Traits, Ring, RPcakage and Obsoletes

by Marcus Denker

On Apr 19, 2012, at 7:02 PM, Eliot Miranda wrote: > > > On Wed, Apr 18, 2012 at 11:27 PM, Marcus Denker <marcus.denker(a)inria.fr> wrote: > > On Apr 18, 2012, at 11:49 PM, Mariano Martinez Peck wrote: > > > Hi guys. I am running some tests with Fuel. In such test I create classes and traits. The problem is that I get a: "MessageNotUnderstood: receiver of "isProvided" is nil" > > Everything starts (it seems) when I do #addToComposition: and I receive a MethodRecategorized event and RPackage calls #systemMethodRecategorizedActionFrom: > > There, the problem is with -> method isProvidedByATrait > > the class of the method is 'AnObsoleteClassForTestToBeDeleted1' > > And the error finally arrives to #providedLocatedMethod where it does "each method isProvided" the problem is that "each" is a RGMethodDefinition with everything in nil except 'name' and 'annotations'. > > I attach PharoDebug.log. > > Any help is really appreaciated. > > we should fix this bug: > > http://code.google.com/p/pharo/issues/detail?id=4750 > > it used to be that methods did not know their name, thus MethodReference had to be used for many things. > This combined with not sharing compiledMethods means that traits can use methods > directly. > > I'm very worried by this. This misses the fact that a MethodReference and a CompiledMethod are two different things *because* methods can be redefined. So if one holds onto a method that gets redefined one can have more than one method with the same name. However, a MethodReference always refers to the method that is currently installed. In this way a MethodReference is a pointer, a coordinate, but a method is a referent. Personally I think the distinction is important; browser tools should use references, not referents. > But the tools use Ring... somehow I thin it's odd that the inner workings of Traits used MethodReference (it was just because methods have no selector...) Marcus -- Marcus Denker -- http://marcusdenker.de

April 19, 2012

Re: [Pharo-project] The video about Athens

by Esteban Lorenzano

hi, On Apr 19, 2012, at 8:27 PM, phil(a)highoctane.be wrote: > I loaded the config and it went fine. > > Got the NBCogVM from Jenkins. > > But then, got issues with the 1.4 image. > > FSFileSystem isn't in. What to do? try just FileSystem :) > And then, after commenting out the offending part, I end up with Error: function unavailable in NBFFICallout - This by running: > > | m | > m:= AthensDemoMorph new openInWorld. > m nextFigure. > m delete. > > This is on Windows XP SP3. Do I need something extra? Is it not supported on Windows? > > > Philippe > > > 2012/4/19 Denis Kudriashov <dionisiydk(a)gmail.com> > It's so great stuff!!! :)) > > Is cairo used as athens backend for this demo? > Is it should work with Bitblt backend? > > In demo I see many usage of angel parameter for transformations. Do you think about "angel unit objects" like "90 degrees" or "2 radians" instead of raw numbers ? > > How #scaledBy: works with Point argument (self extend) and Float like (frame / 100 sin abs) at the same time? What' logic? > > Best regards, > Denis > > 2012/4/19 Igor Stasenko <siguctua(a)gmail.com> > Hello, all > > I recorded a small introductory video about Athens. > Here it is: > > http://vimeo.com/40653315 > > (about 21 minutes). > > I explain some basic stuff like drawing things using Athens canvas, > building paths, use of transformation matrices etc. > > Please note, this is a first video i ever recorded in such context in > my life, so don't be too harsh :) > Also, note that the animations are jagged because of video recording > frame rate, not because it is that slow in reality. > > There are still a blog post pending about it.. which i plan to finish > when i will finish ;) > > -- > Best regards, > Igor Stasenko. > > > > > > -- > Philippe Back > "Helping you hit the top 3 outcomes you really want to achieve" > > Mob: +32(0) 478 650 140 | Fax: +32 (0) 70 408 027 Mail: phil(a)highoctane.be | Web: http://philippeback.eu | Blog: > http://philippeback.be > > High Octane SPRL > rue cour Boisacq 101 > 1301 Bierges >

April 19, 2012

Re: [Pharo-project] Long click on pluggeable button

by StÃ©phane Ducasse

In 2.0 we will integrate the event refactorings we did to start to be able to think about event dispatch. Fernando built state machine on top of our enhanced events. Fernando state machine idea for handling events is probably the way to go. Stef On Apr 19, 2012, at 8:19 PM, phil(a)highoctane.be wrote: > Interacting with tablets and touch is indeed a different experience (just inputting numbers properly is big - gestures would be great for example). > > Longpress or double touch is also to be discussed. For editing a point, you'd be better off with two fingers or a double touch. > > I am experimenting with this with my iPad app (not in Smalltalk but for the end user it is the same). > > Also, swiping gestures require some recording and effects over time. > > Best is to have a low level thing that picks up the touch events and a set of event handlers that do get the notification when all the low level messy thinking is done. Otherwise the app code gets cluttered as hell. Well, that's my 0.02 EUR. > > I am *very* interested in prototyping such things, beta testing and so on. > > Phil > > 2012/4/19 StÃ©phane Ducasse <stephane.ducasse(a)inria.fr> > Hilaire > > we should just continue and slowly but steadily fix and simplify. > We have been doing it and we will continue. Now if you have concrete with examples suggestions > please let us know. > > Stef > > > Responding to myself as it may be useful to other: > > > > > > btn showSelectionFeedback: false. > > btn changed. > > > > but well. > > > > Hilaire > > > > Le 18/04/2012 11:32, Hilaire Fernandes a Ã©crit : > >> Hello, > >> > >> Don't know if any of you tried to use long click event with pluggable > >> button. > >> > >> At first it looks pretty easy, as: > >> > >> btn on: #mouseStillDown send: #hello to: self. > >> > >> does the job. > >> > >> The hell is comming because btn aspect is changed but it states is not > >> changed. This is what I want but the aspect changed, but not the state, > >> so it get unsynchronized. > >> > >> I cannot find how to get the btn back to its aspect reflected by its > >> sate (#getModelState). I look at #update: but did not find. > >> > >> Of course it will be even better if its aspect is not changed when > >> #mouseStillDown event is triggered, but only at mouseUp. > >> > >> > >> Thanks > >> > >> Hilaire > >> > >> > >> PS: > >> - I found state to be misleading, should not it be selected? > >> - Also the use of symbol to describe the morph is a hell > >> - the morph extension add a layer of obscurity > >> - update/change is catastrophic in term of readability to maintain large > >> application > >> > >> Yes, I fell Morph does not scale when you get large application. At > >> least I got lost in my own code only when dealing with this specific part. > >> > >> > > > > > > -- > > Dr. Geo -- http://www.drgeo.eu > > > > > > > > > > -- > Philippe Back > "Helping you hit the top 3 outcomes you really want to achieve" > > Mob: +32(0) 478 650 140 | Fax: +32 (0) 70 408 027 Mail: phil(a)highoctane.be | Web: http://philippeback.eu | Blog: > http://philippeback.be > > High Octane SPRL > rue cour Boisacq 101 > 1301 Bierges >

April 19, 2012

Re: [Pharo-project] The video about Athens

by phil＠highoctane.be

I loaded the config and it went fine. Got the NBCogVM from Jenkins. But then, got issues with the 1.4 image. FSFileSystem isn't in. What to do? And then, after commenting out the offending part, I end up with Error: function unavailable in NBFFICallout - This by running: | m | m:= AthensDemoMorph new openInWorld. m nextFigure. m delete. This is on Windows XP SP3. Do I need something extra? Is it not supported on Windows? Philippe 2012/4/19 Denis Kudriashov <dionisiydk(a)gmail.com> > It's so great stuff!!! :)) > > Is cairo used as athens backend for this demo? > Is it should work with Bitblt backend? > > In demo I see many usage of angel parameter for transformations. Do you > think about "angel unit objects" like "90 degrees" or "2 radians" instead > of raw numbers ? > > How #scaledBy: works with Point argument (self extend) and Float like > (frame / 100 sin abs) at the same time? What' logic? > > Best regards, > Denis > > 2012/4/19 Igor Stasenko <siguctua(a)gmail.com> > >> Hello, all >> >> I recorded a small introductory video about Athens. >> Here it is: >> >> http://vimeo.com/40653315 >> >> (about 21 minutes). >> >> I explain some basic stuff like drawing things using Athens canvas, >> building paths, use of transformation matrices etc. >> >> Please note, this is a first video i ever recorded in such context in >> my life, so don't be too harsh :) >> Also, note that the animations are jagged because of video recording >> frame rate, not because it is that slow in reality. >> >> There are still a blog post pending about it.. which i plan to finish >> when i will finish ;) >> >> -- >> Best regards, >> Igor Stasenko. >> >> > -- Philippe Back "Helping you hit the top 3 outcomes you really want to achieve" Mob: +32(0) 478 650 140 | Fax: +32 (0) 70 408 027 Mail: phil(a)highoctane.be| Web: http://philippeback.eu | Blog: http://philippeback.be High Octane SPRL rue cour Boisacq 101 1301 Bierges

April 19, 2012

Re: [Pharo-project] Stupid question about Fuel

by StÃ©phane Ducasse

On Apr 19, 2012, at 6:12 PM, Sean P. DeNigris wrote: > > Sean P. DeNigris wrote >> >> Steph, would you be happy integrating such a feature? If so, I can have it >> ready quickly and bring back needed (at least for Edgar and I) >> functionality, turned off by default, so there would be no difference felt >> for those that don't want it... >> > > Steph, did you see this? No I'm sick since a week and running between meetingsâ¦ Now if you have the code I can have a look and let you know because when I still how some parts of moprhic are writtenâ¦. I prefer to have less feature to focus on the right ones. > > Thanks, > Sean > > -- > View this message in context: http://forum.world.st/Stupid-question-about-Fuel-tp4558930p4571402.html > Sent from the Pharo Smalltalk mailing list archive at Nabble.com. >

April 19, 2012

Re: [Pharo-project] Long click on pluggeable button

by Hilaire Fernandes

Le 19/04/2012 20:04, StÃ©phane Ducasse a Ã©crit : > We have been doing it and we will continue. Now if you have concrete with examples suggestions > please let us know. It is not easy for me, it required long though on the matter. One think I really really don't like is the use of symbol, because I fell like lost, and no tool to discover what's going on. And symbol are largely used in the UI framework. May be it is not avoidable, don't really know. Hilaire -- Dr. Geo -- http://www.drgeo.eu

April 19, 2012

Re: [Pharo-project] Ehh??? cannot fileout with Nautilus and RPackage?

by StÃ©phane Ducasse

On Apr 19, 2012, at 5:40 PM, Mariano Martinez Peck wrote: > fileOut > > self error: 'I should be implemented somehow' > > mmmmmmm Yes a good soul should sit and add the magic invocation. > > > -- > Mariano > http://marianopeck.wordpress.com >

April 19, 2012

Re: [Pharo-project] Long click on pluggeable button

by phil＠highoctane.be

Interacting with tablets and touch is indeed a different experience (just inputting numbers properly is big - gestures would be great for example). Longpress or double touch is also to be discussed. For editing a point, you'd be better off with two fingers or a double touch. I am experimenting with this with my iPad app (not in Smalltalk but for the end user it is the same). Also, swiping gestures require some recording and effects over time. Best is to have a low level thing that picks up the touch events and a set of event handlers that do get the notification when all the low level messy thinking is done. Otherwise the app code gets cluttered as hell. Well, that's my 0.02 EUR. I am *very* interested in prototyping such things, beta testing and so on. Phil 2012/4/19 StÃ©phane Ducasse <stephane.ducasse(a)inria.fr> > Hilaire > > we should just continue and slowly but steadily fix and simplify. > We have been doing it and we will continue. Now if you have concrete with > examples suggestions > please let us know. > > Stef > > > Responding to myself as it may be useful to other: > > > > > > btn showSelectionFeedback: false. > > btn changed. > > > > but well. > > > > Hilaire > > > > Le 18/04/2012 11:32, Hilaire Fernandes a Ã©crit : > >> Hello, > >> > >> Don't know if any of you tried to use long click event with pluggable > >> button. > >> > >> At first it looks pretty easy, as: > >> > >> btn on: #mouseStillDown send: #hello to: self. > >> > >> does the job. > >> > >> The hell is comming because btn aspect is changed but it states is not > >> changed. This is what I want but the aspect changed, but not the state, > >> so it get unsynchronized. > >> > >> I cannot find how to get the btn back to its aspect reflected by its > >> sate (#getModelState). I look at #update: but did not find. > >> > >> Of course it will be even better if its aspect is not changed when > >> #mouseStillDown event is triggered, but only at mouseUp. > >> > >> > >> Thanks > >> > >> Hilaire > >> > >> > >> PS: > >> - I found state to be misleading, should not it be selected? > >> - Also the use of symbol to describe the morph is a hell > >> - the morph extension add a layer of obscurity > >> - update/change is catastrophic in term of readability to maintain large > >> application > >> > >> Yes, I fell Morph does not scale when you get large application. At > >> least I got lost in my own code only when dealing with this specific > part. > >> > >> > > > > > > -- > > Dr. Geo -- http://www.drgeo.eu > > > > > > > -- Philippe Back "Helping you hit the top 3 outcomes you really want to achieve" Mob: +32(0) 478 650 140 | Fax: +32 (0) 70 408 027 Mail: phil(a)highoctane.be| Web: http://philippeback.eu | Blog: http://philippeback.be High Octane SPRL rue cour Boisacq 101 1301 Bierges

April 19, 2012

Re: [Pharo-project] Problem with Traits, Ring, RPcakage and Obsoletes

by StÃ©phane Ducasse

Please please please pay attention! RPackageOrganizer needs another pass because - not because of RPackageOrganizer but because of the state of the system Then notice that RPackageOrganizer has been designed in a way that the tests are ***NOT*** run against the default RPackageOrganizer to make sure that if you write a bad tests you do not break the system. Have a look at RPackageOrganize API and how we use it in RPackageOrganizer tests Now if may happen that you keep a debugger open when you debug a test of RPackageOrganizer and it may confuse you because you will see propagation of events handled in the RPackageOrganizer instance and not the default one. So pay attention to close the debugger so that notification are passed to the default organizer. Now I need somebody good and with time to have a look at the integration of RPackage into the system. But right now I do not have 3 hours to me to get concentrated. Stef

April 19, 2012

Re: [Pharo-project] OB, Refactoring, and OCompletion in 1.4

by Esteban Lorenzano

...and btw, you shouldn't install newocompletion. It is integrated with pharocore now. nd the versions in repository are not updated (my fault, completely forget about that :P). If you want to re-load it, you can install packages in pharo repository (NECompletion and NOCompletion, in that order) best, Esteban On Apr 19, 2012, at 7:58 PM, StÃ©phane Ducasse wrote: > Yes people should fix the configurations. > We will start to work on the process described in the PharoVision so that all the Configurations get tested. > > Stef > > > On Apr 18, 2012, at 6:35 PM, Sean P. DeNigris wrote: > >> The Welcome window points to the config browser, which loads basic OB (no >> refactoring). I was expecting to have refactoring installed, which is done >> via (per >> http://forum.world.st/ANN-new-stable-version-of-omnibrowser-td4525968.html#…) >> (ConfigurationOfOmniBrowser >> project version: #stable) load: #( 'Dev' 'Dev Tests'). >> >> This scripts installs OmniBrowser + OCompletion + Refactoring Browser >> >> The above works, but maybe this is a problem because it's using the old >> OCompletion? >> >> Then I tried to load NewOCompletion to see if it played nicely with OB, and >> got: "Error: Could not resolve: ConfigurationOfNewECompletion >> [ConfigurationOfNewECompletion] in /.../Pharo-1.4/package-cache". >> >> So I loaded ConfigurationOfNewECompletion manually in the MC browser, and >> then ran into "MessageNotUnderstood: NOCCompletionTable >> class>>methodChanged:" >> >> So there are some problems with the configs for the new completion work... >> and if these new tools work with OB, maybe the config for OB should be >> updated to use them, and then maybe the default group of the OB config >> should load 'Dev'... >> >> Sean >> >> Sean >> >> -- >> View this message in context: http://forum.world.st/OB-Refactoring-and-OCompletion-in-1-4-tp4568254p45682… >> Sent from the Pharo Smalltalk mailing list archive at Nabble.com. >> > >

April 19, 2012