- Pharo-dev - lists.pharo.org

[Pharo-project] pharo.st should be a permanent domain, I think

by Esteban Lorenzano

Hi, I tried pharo.st to reach pharo. I liked. It's small, concise, beautiful (and it has a .st!) One problem is that as soon as you type pharo.st, you loose the url. I would like to stay in the url I entered to pharo-project. I think this is easily doable with an apache configuration. I'm not an apache guy, but I'm sure if there are some problem with this, and all of you agree this is desirable, someone from the list (cough, cough, DiÃ³genes, cough, cough) can do the job :) Cheers, Esteban

Jan. 31, 2011

Re: [Pharo-project] Issue 3234 in pharo: fold: reduce: Merge

by pharo＠googlecode.com

Updates: Status: Fixed Comment #2 on issue 3234 by stephane...(a)gmail.com: fold: reduce: Merge http://code.google.com/p/pharo/issues/detail?id=3234 (No comment was entered for this change.)

Jan. 31, 2011

[Pharo-project] what do we do for OmniBrowser?

by StÃ©phane Ducasse

Hi guys Lukas will not maintain OB for Pharo 1.2 and personnally I will not do it because I do not have the knowledge and the time. So now this is time to know what do we do for OmniBrowser. I have all my ears open. The first step I would do - host OB for 1.1 in a repository where we have write access. - create a configurationOf for OB and save it there Stef

Jan. 31, 2011

Re: [Pharo-project] About KeyBindings

by StÃ©phane Ducasse

I have the impression that Guillermo work is not the same as GlobalKeyBinding but I may be wrong Anyway fernando could I have a look at your code? Because I really want that. Stef On Jan 31, 2011, at 4:18 PM, Fernando Olivero wrote: > Hi, recently there were some discussion on implementing KeyBindings for our IDE. > > There are at least 3 frameworks which address this functionality, the > last one mentioned is the work done by Guillermo. > I've implemented another one, for Gaucho. > > From what i understand, the previous work focuses on implementing > GLOBAL keybindings . > Basically adding behavior to HandMorph or PasteUpMorph or the toolbar > ( docking bar) present in the World. > > My implementation is a generic keybinding for any Morph, which can be > applied to any morph. Special cases for the docking bar or the paste > up Morph can be easily implemented following this uniform scheme. > > I didn't have the time to integrate it into Pharo yet, but will do in > one week from now. > > I like to bring up the discussion on which one we should adopt and > integrate into Morphic, points in favor of my approach: > 1) clean > 2) customizable > 3) generic: no special cases > > I believe the other Global keybindings should be a subset of this > generic keybinding mechanism. > > Fernando > > pd: From a previous mail: > > "... > So for example, if you want to delete any Morph from the system > by pressing cmd-w, you have to add the following binding: > > condition := GMCondition > compositeWith: > #( #understandsKeyBindingCommandCondition > #uneditedCondition ) ). > binding := GMKeyBinding > actingOn: $w asciiValue > modifiedBy: #(#command) > satisfying: condition > applying: #close. > m := Morph new. > m addKeyBinding: binding . > ... > " > > pd2: Useful and related comment from Stef > ".. > A classVar Binding and an instance var binding. > **All** the methods only access binding > binding get initialized with the default table defined in Binding > => we can have table binding sharing > => we can have instance based customization. >

Jan. 31, 2011

Re: [Pharo-project] Text Selection problem in 1.2 dev

by StÃ©phane Ducasse

So alain should I do something? Or just ProfStef should be fixed? Stef > yes, it doesn't work because the fix has been overwritten with old code. > I think that only SmalltalkEditor>>#evaluateSelection has been touched. > Just browse this method, then open versions and set #evaluateSelection back > to its previous version (12/21/2010). > > > > Le 31/01/2011 17:17, StÃ©phane Ducasse a Ã©crit : >> Alain in >> 13033 >> >> Gofer >> ... metacelloRepository >> >> >> (ConfigurationOfProfStef project version: '1.6') load. >> >> ProfStef go >> >> >> >> >>> It seems to me that it has already been reported and fixed a time ago. >>> see http://code.google.com/p/pharo/issues/detail?id=3472 >>> take >>> >>> SLICE-Issue-3472-Text-selection-problem-in-Pharo-dev-12 and >>> SLICE-Issue-3472-Text-selection-problem-in-Pharo-dev-12-Part-2 >>> >>> Cheers >>> Alain >>> >>> Le 31/01/2011 14:23, StÃ©phane Ducasse a Ã©crit : >>>> hi guys >>>> >>>> in 1.2 dev when I play with ProfStef go I get an error after a couple of screens normally 3. >>>> >>>> I do not understand why but it looks really a problem with text selection >>>> >>>> https://pharo-ic.lille.inria.fr/hudson/view/Pharo/job/Pharo%201.2/ >>>> >>>> Stef >>>> >>>> >>>> >>> >> >> > >

Jan. 31, 2011

Re: [Pharo-project] Text Selection problem in 1.2 dev

by StÃ©phane Ducasse

Ok! Laurent can you fix profStef? Stef On Jan 31, 2011, at 6:41 PM, Alain Plantec wrote: > >>> I do not understand why but it looks really a problem with text selection > It's not a problem with text selection. > The problem is that profStef next changes the contents of the morph (with setText:). > Nothing anormal here but, while doing so it kills the editor from where the selection is evaluated. > See the portion of #evaluateSelection given below: > -------------------- > evaluateSelection > ... > result := [ > rcvr class evaluatorClass new > evaluate: self selectionForDoitAsStream > in: ctxt > to: rcvr > notifying: self <------***** the problem is here ***** > ifFail: [FakeClassPool adopt: nil. ^ #failedDoit] > logged: true. > ] > ... > ----------------- > The receiver is passed as the requestor and it is bad because after the selection has been evaluated, > the receiver is destroyed by the #setText: and a new editor is built. > After having evaluated the selection, the compiler sends several messages to the requestor. > Then it crashes because the requestor is destroyed by the evaluation ... :) > So the solution is to set the editor morph as the requestor (which is stable), not the editor. > > Cheers > Alain > >

Jan. 31, 2011

Re: [Pharo-project] would be nice if configuration comment always contain how to load them :(

by StÃ©phane Ducasse

Why not put this comment by default in all the configurationOfXXX I spent 10 min to find how to load Profstef to check a bug and I wrote the chapter on metacello with mariano so there is something wrong. Stef > Don't forget that the aim is to have all the configurations updated to > the new symbolic version capabilities of the Metacello. > > Given that, then the standard (official) way to load (either by hand or > by tools like the browser) *any* configuration will be something similar > to: > > Gofer it "or new" > squeaksource: "MetacelloRepository"; > project: "XXX"; > load. > > ((Smalltalk globals at: #ConfigurationOfXXX) project version: #stable) > load > > This of course will be hidden from users by the existence of the > user-tools and will install the stable version tested for the image > version metacello it is running on. > > Of course it will be way to load a specific version (either because a > bug fix hunting or simply by personal preference) of a given > configuration. Something like: > > ((Smalltalk globals at: #ConfigurationOfXXX) project version: > '1.4-something') load > > But this will be only for specific users and for specific reasons. Most > users only want the stable version for its platform and metacello's > symbolic versions give them that. > > Cheers. > > > El lun, 31-01-2011 a las 17:23 +0100, Luc Fabresse escribiÃ³: >> +1 >> >> >> #Luc >> >> >> 2011/1/31 StÃ©phane Ducasse <stephane.ducasse(a)inria.fr> >> >> >> >> > > -- > Miguel CobÃ¡ > http://twitter.com/MiguelCobaMtz > http://miguel.leugim.com.mx > > > >

Jan. 31, 2011

[Pharo-project] Smalltalk Stammtisch ZÃ¼rich Februar: 09.02.

by Philippe Marschall

Hallo, Der Stammtisch findet statt am - Mittwoch, 09.02.2011, 18h - Steinfels ZÃ¼rich , http://www.steinfels-zuerich.ch/ - http://maps.google.com/maps?ie=UTF8&q=restaurant+steinfels+z%C3%BCrich&fb=1… Wer kommen will soll sich bitte per EMail melden damit wir wissen fÃ¼r wie viele Leute wir reservieren mÃ¼ssen. Nein, es gibt keine RÃ¶sti. Philippe ============================================ Smalltalk Beer-trinking event: - Wednesday, 09.02.2011, 18h - Steinfels ZÃ¼rich , http://www.steinfels-zuerich.ch/ - http://maps.google.com/maps?ie=UTF8&q=restaurant+steinfels+z%C3%BCrich&fb=1… If you plan to attend please send a mail. No, there won't be any RÃ¶sti. Philippe

Jan. 31, 2011

Re: [Pharo-project] Hudson now Jenkins

by Philippe Marschall

On 31.01.2011 15:08, Marcus Denker wrote: > > On Jan 31, 2011, at 3:04 PM, Geert Claes wrote: > >> >> Since we are using Hudson I was wondering if anyone has been following >> Hudson's issues with Oracle at all? >> >> This is how I understand it; the creator of Hudson (Kohsuke Kawaguchi) and >> founder of infradna.com (which has now joined forces with Cloudbees - >> offering Hudson support) has had some disagreements with Oracle mainly >> regarding the ownership of the "Hudson" name. The Hudson community has >> recently voted for the project to be renamed to Jenkins, so hudson-labs.org >> is now jenkins-ci.org >> >> ps. I actually started looking at Hudson because I was baffled at the fact >> that the success icon is "blue" rather than a more intuitive "green" and >> hey, there is a plugin for that: >> http://wiki.hudson-ci.org/display/HUDSON/Green+Balls >> > Yes, but sadly the installed version of hudson is too old for this plugin. > (yes, we need to update. I will ask again) Hudson/Jenkins release once a week, pretty amazing. Cheers Philippe

Jan. 31, 2011

Re: [Pharo-project] Hudson now Jenkins

by Philippe Marschall

On 31.01.2011 15:04, Geert Claes wrote: > > Since we are using Hudson I was wondering if anyone has been following > Hudson's issues with Oracle at all? > > This is how I understand it; the creator of Hudson (Kohsuke Kawaguchi) and > founder of infradna.com (which has now joined forces with Cloudbees - > offering Hudson support) has had some disagreements with Oracle mainly > regarding the ownership of the "Hudson" name. The Hudson community has > recently voted for the project to be renamed to Jenkins, so hudson-labs.org > is now jenkins-ci.org > > ps. I actually started looking at Hudson because I was baffled at the fact > that the success icon is "blue" rather than a more intuitive "green" and > hey, there is a plugin for that: > http://wiki.hudson-ci.org/display/HUDSON/Green+Balls My understanding is that blue in Japanse culture has the same meaning as green in "western" culture. Cheers Philippe

Jan. 31, 2011