May 2015 - Pharo-dev - lists.pharo.org

New CogVMs available
by Eliot Miranda May 19, 2015

May 19, 2015

at http://www.mirandabanda.org/files/Cog/VM/VM.r3350/. CogVM binaries as per VMMaker.oscog-eem.1315/r3350 General: Make the bounds check in primitiveObject:perform:... more restrictive and fix bugs in the commentary. Make stack pages a reasonable size. My math was wrong and the size was half as big as intended. Extend the fastPrimTrace scheme to trace stack overflow, prim failure and prim retry to help debug the phantom stack frame bug in Cog Spur. Cogit: Fix old bug in ceMNUFromPICMNUMethod:receiver: that has been masked by cacheing of MNU method in MNU PIC entries. ceMNUFromPICMNUMethod:receiver: must set up lkupClass before calling handleMNU:... Revamp icache flushing in the Cogit for ARM. Move icache flush for generated methods/pics to the fillIn*Header: routines. Make sure the cache is flushed in generateCaptureCStackPointers:. Flush the entire PIC when extending it. Fix ARM-only regression in StackToRegisterMappingCogit's PIC aborts when the SimpleStackBasedCogit was fixed. Make ARM rotated 8bit constants actually be correct(er) ie 16r4000001 (SmallInteger 16r2000000) work. Spur: Fix ceActivateFailingPrimitiveMethod: to return properly after retrying a primitive after checkForAndFollowForwardedPrimitiveState. e.g. fixes | s | 1 to: 5 do: [:i| s := 'bar'. s becomeForward: 'bzzt' copy. 'foo' <= s] Fix bug in followForwardedFrameContents:stackPointer: so that arguments are actually followed. N.B. drives the above bug underground ;-). so that named constants (defines) are not elided. Make accessor depth calculation more accurate by not considering fetchInteger:ofObject: and arraySize: et al as object accessors. Newspeak: Remove landmine left from restructing implicit receiver and outer sends as clean sends. Slang: Do a more thorough job of transforming complex to:[by:]do: loops so that the limit expression is only evaluated once, not on each iteration of the loop. Make the read-before-written initializer safer, only initialzing simple types. Plugins: Regenerated because of changes in accessorDepth calculation, constant elimination, and read-before-use initialization. -- best, Eliot

1 0

[SURVEY] Pillar input/output files.
by Cyril Ferlicot May 18, 2015

May 18, 2015

Hi, I'm currently working on the way we manage the input files and the output files. The way we manage the files now makes hard some improvement of Pillar. That's why with Damien we though about makes some changes. The changes are big so I wanted to talk about that with you. We propose to have always 1 file with all the work inside. That can use the inputFile tag. For example: "${inputFile:Chapter1.pillar}$ ${inputFile:Chapter2.pillar}$ ${inputFile:Chapter3.pillar}$ ${inputFile:Chapter4.pillar}$" And the option "separateOutputFiles" will now create 1 file each time we have a new Chapter. (!Level 1 header). That will make simpler the way to number everything and to create anchor because those 2 doesn't work well with the "separateOutputFile" option. What do you thing? That would be good with you? -- Cheers Cyril Ferlicot

5 6

Loading Pillar in 4.0
by Sean P. DeNigris May 18, 2015

May 18, 2015

Loading from the Configuration Browser gives: This package depends on the following classes: PRWarning You must resolve these dependencies before you will be able to load these definitions: PRParameterWarning ----- Cheers, Sean -- View this message in context: http://forum.world.st/Loading-Pillar-in-4-0-tp4825915.html Sent from the Pharo Smalltalk Developers mailing list archive at Nabble.com.

3 3

Performance of GT-Inspector on large collections
by Alejandro Infante May 18, 2015

May 18, 2015

Hello, I have noticed that working with large collections in the new inspector is noticeable slower than with the old one :( I executed a tiny benchmark that is just inspecting a 1.000.000 objects collection: âââââââââ Smalltalk tools inspectorTool: GTInspector. Transcript show: 'Exp with GT-Inspector: '. Transcript show: [ ((1 to: 1000000) collect: [ :i | Object new ]) inspect ] timeToRun. Transcript cr. Smalltalk tools inspectorTool: EyeInspector. Transcript show: 'Exp with EyeInspector: '. Transcript show: [ ((1 to: 1000000) collect: [ :i | Object new ]) inspect ] timeToRun. âââââââââ And the results were: Exp with GT-Inspector: 0:00:01:04.979 Exp with EyeInspector: 0:00:00:00.381 Are you facing the same issue? Alejandro

3 2

Re: [Pharo-dev] Trying to use Rubric in Nautilus ....
by Franck Warlouzet May 18, 2015

May 18, 2015

Ok I was not looking the solution in Nautilus, but in Rubric ... Thanks a lot ! My eyes are alive again Franck To: pharo-dev(a)lists.pharo.org From: pharo-dev(a)lists.pharo.org CC: alain.plantec(a)yahoo.com Subject: Re: [Pharo-dev] Trying to use Rubric in Nautilus .... Date: Mon, 18 May 2015 07:38:15 -0700 --PiÃ¨ce jointe du message transmise-- Subject: Re: [Pharo-dev] Trying to use Rubric in Nautilus .... From: alain.plantec(a)yahoo.com Date: Mon, 18 May 2015 16:37:59 +0200 To: pharo-dev(a)lists.pharo.org also send #withCodeSizeFeedback to have almost the same look as with PluggableTextMorph. AbstractNautilusUI >>buildNewSourceTextArea sourceTextArea := self textMorphClass on: self text: #sourceCodeFrom: accept: #compileSource:notifying: readSelection: #contentsSelectionFrom: menu: #sourceCodeMenu:shifted:. sourceTextArea askBeforeDiscardingEdits: true; beForSmalltalkCode; withLineNumbers; withCodeSizeFeedback; <==== here ====== beWrapped; vResizing: #spaceFill; hResizing: #spaceFill; font: StandardFonts codeFont; spaceFillWeight: 3; on: #keyStroke send: #keyStroke:fromSourceCodeMorph: to: self. self setSourceCodeShorcutsTo: sourceTextArea. ^ sourceTextArea Alain On 18 May 2015, at 16:17, Alain Plantec <alain.plantec(a)yahoo.com> wrote:Hello Franck, just changed shoutAboutToStyle: as follow:AbstractNautilusUI>>shoutAboutToStyle: aPluggableShoutMorphOrView âThe comments are not styled" ^ aPluggableShoutMorphOrView ~= commentTextArea and it seems to work. Now, pay attention that, with Rubric, this is the editing mode that is passed as argument (for aPluggableShoutMorphOrView). Here, a RubSmalltalkCodeMode. if you plan to use a Rubric also for the comments, then, normally shoutAboutToStyle: will not be sent because I guess a RubPlainTextMode will be used for the comment pane. This mode do not use shout.In that case the final version should be: AbstractNautilusUI>>shoutAboutToStyle: aPluggableShoutMorphOrView ^ true CheersAlain On 18 May 2015, at 14:41, Franck Warlouzet <franck.warlouzet(a)hotmail.fr> wrote:Hi, With Alain's last commit of Rubric (205) we made a few changes to put Rubric in Nautilus. You can see thoses changes in the attached file. As you can see there is no colors in the browser... Is RubScrolledTextMorph>>beForSmalltalkCode the good option ? I thought there would be just an option like aboutToStyle: true or something like that but there is not. I do not know if I missed something obvious or if I have to add the connection to RubShoutStylerDecorator. Any idea everyone ? I am losing my eyes with uncolored code but I guess it is worth it. Thanks, Franck To: pharo-dev(a)lists.pharo.org From: pharo-dev(a)lists.pharo.org CC: alain.plantec(a)yahoo.com Subject: Re: [Pharo-dev] Trying to use Rubric in Nautilus .... Date: Thu, 14 May 2015 05:16:30 -0700 --PiÃ¨ce jointe du message transmise-- Subject: Re: [Pharo-dev] Trying to use Rubric in Nautilus .... From: alain.plantec(a)yahoo.com Date: Thu, 14 May 2015 14:16:16 +0200 To: pharo-dev(a)lists.pharo.org done & committedCheersAlain On 14 May 2015, at 12:06, Alain Plantec <alain.plantec(a)yahoo.com> wrote:ahh, good remark :)the menu stuff is not finished.I do it nowthanks Alain On 13 May 2015, at 18:09, Franck Warlouzet <franck.warlouzet(a)hotmail.fr> wrote:Hi alain thanks a lot! This helps us a lot. We got nearly something working. We will produce a list of methods and changes we had to do. see in the attached cs. Now lookupMenu "default implementation of the algorithm that lookup the menu" | default | default := [ self editingMode menu ]. ^ self model ifNil: [ default value ] ifNotNil: [ :m | m menu ifNil: [ default value ] ] We add a new method in nautilus that creates a menu as follow because m is our NautilusModel menu ^ self sourceCodeMenu: MenuMorph new shifted: false. But I was wondering why m menu does not take into account the information we passed when we created the rubPluggable sourceTextArea := self textMorphClass on: self text: #sourceCodeFrom: accept: #compileSource:notifying: readSelection: #contentsSelectionFrom: menu: #sourceCodeMenu:shifted:. I see that RubEditingArea has a policy but we could find how we can improve RubPluggable to use the menu: To: pharo-dev(a)lists.pharo.org From: pharo-dev(a)lists.pharo.org CC: alain.plantec(a)yahoo.com Subject: Re: [Pharo-dev] Trying to use Rubric in Nautilus .... Date: Wed, 13 May 2015 07:26:15 -0700 --PiÃ¨ce jointe du message transmise-- Subject: Re: [Pharo-dev] Trying to use Rubric in Nautilus .... From: alain.plantec(a)yahoo.com Date: Wed, 13 May 2015 16:25:34 +0200 To: pharo-dev(a)lists.pharo.org see Rubric-AlainPlantec.196 in the Rubric repo. Iâve added RubPluggableTextMorph and RubPluggableTextMorphExample.Need to be polished certainly but it should facilitate the integration of Rubric in Nautilus. cheersAlain On 12 May 2015, at 22:45, stepharo <stepharo(a)free.fr> wrote:For fun I put the following trace in sourceCodeFrom: aTextMorph Transcript show: '.'. ^ self sourceCodeAnd in general this method is invoked three times instead of one :) Before asking we read all the examples and I saw that selector: is used to communicate and update the code pane. (UIManager default newAutoAcceptTextEntryFor: self get: #selector set: #selector: class: String getEnabled: #haveClassName help: 'Enter a selector' translated entryCompletion: entryCompletion) acceptOnCR: false; ghostText: 'Selector'; withDropListButton; I think this is only for the drop down list. What I meant is that I could use set: and pass the selector to be invoked by the dropdown when the drop down is selected. And it will notify self with the selector: selector. So we could sepcify any selector (ie I can register to the drop box and say call me once you are set). In RubMethodEditingExample>>#selector: it calls the update explicit self updateCodeWith: (cls sourceCodeAt: selector) and updateCodeWith: sets the text on the text model. Now my problem (and may be there is not solution) is how can I connect from a pluggableListMorph to a RubScrollText when the list does not generate announcement. This is why I tried to pass via dependents or something like that. I will try to generate announcements but this is a lot more work and I do not know if it will work. Stef <glueRubNautilus.2.cs> <Unnamed.cs>

1 0

Re: [Pharo-dev] Trying to use Rubric in Nautilus ....
by Franck Warlouzet May 18, 2015

May 18, 2015

Hi, With Alain's last commit of Rubric (205) we made a few changes to put Rubric in Nautilus. You can see thoses changes in the attached file. As you can see there is no colors in the browser... Is RubScrolledTextMorph>>beForSmalltalkCode the good option ? I thought there would be just an option like aboutToStyle: true or something like that but there is not. I do not know if I missed something obvious or if I have to add the connection to RubShoutStylerDecorator. Any idea everyone ? I am losing my eyes with uncolored code but I guess it is worth it. Thanks, Franck To: pharo-dev(a)lists.pharo.org From: pharo-dev(a)lists.pharo.org CC: alain.plantec(a)yahoo.com Subject: Re: [Pharo-dev] Trying to use Rubric in Nautilus .... Date: Thu, 14 May 2015 05:16:30 -0700 --PiÃ¨ce jointe du message transmise-- Subject: Re: [Pharo-dev] Trying to use Rubric in Nautilus .... From: alain.plantec(a)yahoo.com Date: Thu, 14 May 2015 14:16:16 +0200 To: pharo-dev(a)lists.pharo.org done & committedCheersAlain On 14 May 2015, at 12:06, Alain Plantec <alain.plantec(a)yahoo.com> wrote:ahh, good remark :)the menu stuff is not finished.I do it nowthanks Alain On 13 May 2015, at 18:09, Franck Warlouzet <franck.warlouzet(a)hotmail.fr> wrote:Hi alain thanks a lot! This helps us a lot. We got nearly something working. We will produce a list of methods and changes we had to do. see in the attached cs. Now lookupMenu "default implementation of the algorithm that lookup the menu" | default | default := [ self editingMode menu ]. ^ self model ifNil: [ default value ] ifNotNil: [ :m | m menu ifNil: [ default value ] ] We add a new method in nautilus that creates a menu as follow because m is our NautilusModel menu ^ self sourceCodeMenu: MenuMorph new shifted: false. But I was wondering why m menu does not take into account the information we passed when we created the rubPluggable sourceTextArea := self textMorphClass on: self text: #sourceCodeFrom: accept: #compileSource:notifying: readSelection: #contentsSelectionFrom: menu: #sourceCodeMenu:shifted:. I see that RubEditingArea has a policy but we could find how we can improve RubPluggable to use the menu: To: pharo-dev(a)lists.pharo.org From: pharo-dev(a)lists.pharo.org CC: alain.plantec(a)yahoo.com Subject: Re: [Pharo-dev] Trying to use Rubric in Nautilus .... Date: Wed, 13 May 2015 07:26:15 -0700 --PiÃ¨ce jointe du message transmise-- Subject: Re: [Pharo-dev] Trying to use Rubric in Nautilus .... From: alain.plantec(a)yahoo.com Date: Wed, 13 May 2015 16:25:34 +0200 To: pharo-dev(a)lists.pharo.org see Rubric-AlainPlantec.196 in the Rubric repo. Iâve added RubPluggableTextMorph and RubPluggableTextMorphExample.Need to be polished certainly but it should facilitate the integration of Rubric in Nautilus. cheersAlain On 12 May 2015, at 22:45, stepharo <stepharo(a)free.fr> wrote:For fun I put the following trace in sourceCodeFrom: aTextMorph Transcript show: '.'. ^ self sourceCodeAnd in general this method is invoked three times instead of one :) Before asking we read all the examples and I saw that selector: is used to communicate and update the code pane. (UIManager default newAutoAcceptTextEntryFor: self get: #selector set: #selector: class: String getEnabled: #haveClassName help: 'Enter a selector' translated entryCompletion: entryCompletion) acceptOnCR: false; ghostText: 'Selector'; withDropListButton; I think this is only for the drop down list. What I meant is that I could use set: and pass the selector to be invoked by the dropdown when the drop down is selected. And it will notify self with the selector: selector. So we could sepcify any selector (ie I can register to the drop box and say call me once you are set). In RubMethodEditingExample>>#selector: it calls the update explicit self updateCodeWith: (cls sourceCodeAt: selector) and updateCodeWith: sets the text on the text model. Now my problem (and may be there is not solution) is how can I connect from a pluggableListMorph to a RubScrollText when the list does not generate announcement. This is why I tried to pass via dependents or something like that. I will try to generate announcements but this is a lot more work and I do not know if it will work. Stef <glueRubNautilus.2.cs>

2 2

[pharo-project/pharo-core]
by GitHub May 18, 2015

May 18, 2015

Branch: refs/tags/50049 Home: https://github.com/pharo-project/pharo-core

1 0

[pharo-project/pharo-core] 0c77aa: 50049
by GitHub May 18, 2015

May 18, 2015

Branch: refs/heads/5.0 Home: https://github.com/pharo-project/pharo-core Commit: 0c77aac8e97adf1d258ac772125041bcd6628980 https://github.com/pharo-project/pharo-core/commit/0c77aac8e97adf1d258ac772… Author: Jenkins Build Server <board(a)pharo-project.org> Date: 2015-05-18 (Mon, 18 May 2015) Changed paths: R Athens-Cairo.package/AthensCairoCanvas.class/instance/drawing text/setFont_.st R Athens-Cairo.package/AthensCairoPDFSurface.class/instance/private/newCanvas.st A Athens-Core.package/extension/BitmapFillStyle/instance/asAthensPaintOn_.st M Athens-Morphic.package/AthensParagraph.class/instance/as yet unclassified/secondarySelection_.st R Athens-Morphic.package/AthensWorldMorph.class/instance/world state/viewBox_.st R Athens-Morphic.package/AthensWrapWorldMorph.class/instance/drawing/fullDrawOnAthensCanvas_.st R Athens-Morphic.package/AthensWrapWorldMorph.class/instance/drawing/render.st M Athens-Morphic.package/extension/EllipseMorph/instance/asAthensShapeOn_.st M Athens-Morphic.package/extension/EllipseMorph/instance/drawOnAthensCanvas_.st A Athens-Morphic.package/extension/PolygonMorph/instance/adjustSegmentPointsForArrows.st A Athens-Morphic.package/extension/PolygonMorph/instance/asAthensCurvedOpenPathOn_.st A Athens-Morphic.package/extension/PolygonMorph/instance/asAthensCurvedPathOn_.st A Athens-Morphic.package/extension/PolygonMorph/instance/asAthensLinePathOn_.st A Athens-Morphic.package/extension/PolygonMorph/instance/asAthensPathOn_.st A Athens-Morphic.package/extension/PolygonMorph/instance/computeCurvedSegmentCtrlPoints.st A Athens-Morphic.package/extension/PolygonMorph/instance/drawArrowOnAthensCanvas_at_from_.st A Athens-Morphic.package/extension/PolygonMorph/instance/drawArrowsOnAthensCanvas_.st A Athens-Morphic.package/extension/PolygonMorph/instance/drawOnAthensCanvas_.st A Athens-Morphic.package/extension/PolygonMorph/instance/hasTwoColorDash.st A Athens-Morphic.package/extension/PolygonMorph/instance/setDashOnStroke_.st A Athens-Morphic.package/extension/PolygonMorph/instance/setSecondColorDashOnStroke_.st A ConfigurationOfAthens.package/ConfigurationOfAthens.class/instance/versions/version35_.st M Manifest-Core.package/extension/RPackage/instance/packageManifest.st R ScriptLoader50.package/ScriptLoader.class/instance/pharo - scripts/script50048.st A ScriptLoader50.package/ScriptLoader.class/instance/pharo - scripts/script50049.st R ScriptLoader50.package/ScriptLoader.class/instance/pharo - updates/update50048.st A ScriptLoader50.package/ScriptLoader.class/instance/pharo - updates/update50049.st M ScriptLoader50.package/ScriptLoader.class/instance/public/commentForCurrentUpdate.st Log Message: ----------- 50049 15557 New PackageManifests should be classes, not instances https://pharo.fogbugz.com/f/cases/15557 15556 use new athens configuration 3.5 https://pharo.fogbugz.com/f/cases/15556 http://files.pharo.org/image/50/50049.zip

1 0

Project set up
by valmy roi May 18, 2015

May 18, 2015

Hi, Is there a way to export a project from pharo? I explain: create a set up of a project (game for example) so that i can give it to someone to play, and he will not need to import it in pharo. If it doesn't exist, it should de created -- Kenfack Valmy-Roi IngÃ©nieur des travaux en Informatique de Gestion/ Analyste Programmeur ElÃ¨ve ingÃ©nieur de conception en gÃ©nie logiciel Institut SupÃ©rieur du Sahel Email : roykenvalmy(a)gmail.com Tel: +237 22 11 01 87/ +237 676 94 23 01 / +237 699 04 0612

4 4

Please integrate 15523 and 15468
by Sean P. DeNigris May 18, 2015

May 18, 2015

If there is no objection, please integrate: Issue 15523: Code Cruft Rule Only Matches One-Liners https://pharo.fogbugz.com/default.asp?15523 Issue 15468: Enhance the Welcome window https://pharo.fogbugz.com/default.asp?15468 Especially the second one, which was aimed at supporting our new users for the 4.0 release. Thanks! ----- Cheers, Sean -- View this message in context: http://forum.world.st/Please-integrate-15523-and-15468-tp4826949.html Sent from the Pharo Smalltalk Developers mailing list archive at Nabble.com.

2 2