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

Format shortcut.

by Cyril Ferlicot

Hi, for now we have 2 different shortcuts in hard code on Pharo to format methods: alt + shift + u and ctrl + shift + f. I opened an issue to correct that : https://pharo.fogbugz.com/f/cases/15584/TClassDescription-sourceCodeRefacto… I propose to keep ctrl + shift + f (like in Eclipse). -- Cheers Cyril Ferlicot

May 22, 2015

[pharo-project/pharo-core]

by GitHub

Branch: refs/tags/50058 Home: https://github.com/pharo-project/pharo-core

May 22, 2015

[pharo-project/pharo-core] f3f543: 50058

by GitHub

Branch: refs/heads/5.0 Home: https://github.com/pharo-project/pharo-core Commit: f3f543ad9776f63e533fe3bbc8dd6bd5eb27fa86 https://github.com/pharo-project/pharo-core/commit/f3f543ad9776f63e533fe3bb… Author: Jenkins Build Server <board(a)pharo-project.org> Date: 2015-05-22 (Fri, 22 May 2015) Changed paths: M ConfigurationOfTxText.package/ConfigurationOfTxText.class/instance/symbolic versions/stable_.st A ConfigurationOfTxText.package/ConfigurationOfTxText.class/instance/versions/version210_.st R ScriptLoader50.package/ScriptLoader.class/instance/pharo - scripts/script50057.st A ScriptLoader50.package/ScriptLoader.class/instance/pharo - scripts/script50058.st R ScriptLoader50.package/ScriptLoader.class/instance/pharo - updates/update50057.st A ScriptLoader50.package/ScriptLoader.class/instance/pharo - updates/update50058.st M ScriptLoader50.package/ScriptLoader.class/instance/public/commentForCurrentUpdate.st M TxText-Athens.package/TxApplicationWithToolbar.class/instance/as yet unclassified/openWithSpec.st M TxText-Athens.package/TxAthensLayout.class/instance/geometry/textPositionAt_.st M TxText-Athens.package/TxAthensLayoutLine.class/instance/accessing/commandForPosition_ifOutside_.st M TxText-Athens.package/TxTextEditor.class/instance/private/recordSelectionChange_.st M TxText-Model.package/TxBasicSpan.class/instance/accessing/lineEndSpan.st M TxText-Model.package/TxBasicSpan.class/instance/operation/at_insert_.st M TxText-Model.package/TxBasicTextPosition.class/instance/converting/selectTo_.st M TxText-Model.package/TxBasicTextPosition.class/instance/private/timesToMoveRightTo_.st M TxText-Model.package/TxCharacterSpan.class/instance/operation/applyAttribute_from_to_.st M TxText-Model.package/TxCharacterStream.class/README.md R TxText-Model.package/TxModel.class/instance/accessing/characterStream.st R TxText-Model.package/TxModel.class/instance/accessing/cursor.st A TxText-Model.package/TxModel.class/instance/factory methods/characterStream.st A TxText-Model.package/TxModel.class/instance/factory methods/newCursor.st A TxText-Model.package/TxModel.class/instance/factory methods/newSelection.st R TxText-Model.package/TxModel.class/instance/factory-selection /newSelection.st R TxText-Model.package/TxModel.class/instance/factory-selection /select.st R TxText-Model.package/TxModel.class/instance/factory-selection /selectAll.st R TxText-Model.package/TxModel.class/instance/factory-selection /selection.st R TxText-Model.package/TxModel.class/instance/private/deleteFrom_to_.st R TxText-Model.package/TxModel.class/instance/private/mergeFrom_.st R TxText-Model.package/TxModel.class/instance/private/mergeFrom_to_.st R TxText-Model.package/TxModel.class/instance/private/normalize.st M TxText-Model.package/TxTextCursor.class/README.md M TxText-Model.package/TxTextPosition.class/instance/accessing/linePosition_.st M TxText-Styler.package/TxTextStyler.class/instance/visiting/visitMessageNode_.st M TxText-Styler.package/TxTextStyler.class/instance/visiting/visitSelfNode_.st M TxText-Styler.package/TxTextStyler.class/instance/visiting/visitVariableNode_.st M TxText-Styler.package/extension/OCAbstractMethodScope/instance/findVariable_ifNone_.st M TxTextTests-Model.package/TxModelTest.class/instance/running/setUp.st M TxTextTests-Model.package/TxModelTest.class/instance/test-attribute/testApplyFontWithOneSplitAtEnd.st M TxTextTests-Model.package/TxModelTest.class/instance/test-attribute/testApplyFontWithOneSplitAtStart.st Log Message: ----------- 50058 15567 TxText Cleanup for Pharo 5.0 #4 https://pharo.fogbugz.com/f/cases/15567 http://files.pharo.org/image/50/50058.zip

May 22, 2015

Re: [Pharo-dev] can someone explain me this?

by Esteban Lorenzano

> On 21 May 2015, at 23:52, Marcus Denker <marcus.denker(a)inria.fr> wrote: > > >> On 21 May 2015, at 22:14, Esteban Lorenzano <estebanlm(a)gmail.com> wrote: >> >> Smalltalk allClassesAndTraits size. "5057" >> Object withAllSubclasses size. â9900" >> >> I really do not understand :) >> > > Metaclasses. For every X there is "X classâ which is (eventually) a subclass of Object, too. rightâ¦ got it :P > > Marcus > > > >

May 22, 2015

Best practice for overwriting methods in other classes

by Kasper Osterbye

I have now twice in two different situations felt a need to "steal" a method in Nautilus and make that method an extension method in my own package. In situation A, I wanted to steal the #buildCommentPane to use a different Morph for class comments, In situation B, Sergio and I needed to redefine #getComments and #addComments:notifying: to fill out behaviour for package comments. What is the kosher way to work and contribute in such situations? I guess it will also pop up in situations outside of Nautilus. -- View this message in context: http://forum.world.st/Best-practice-for-overwriting-methods-in-other-classe… Sent from the Pharo Smalltalk Developers mailing list archive at Nabble.com.

May 22, 2015

Re: [Pharo-dev] can someone explain me this?

by Norbert Hartl

> Am 21.05.2015 um 22:14 schrieb Esteban Lorenzano <estebanlm(a)gmail.com>: > > Smalltalk allClassesAndTraits size. "5057" > Object withAllSubclasses size. â9900" > > I really do not understand :) And I enjoy those times that you do not understand something. Makes me feel closer to you :P Norbert

May 21, 2015

New Cog VMs available

by Eliot Miranda

... at http://www.mirandabanda.org/files/Cog/VM/VM.r3354 CogVM binaries as per VMMaker.oscog-eem.1318/r3354 Cogit: Fix the performance regression on x86 in r3308 VMMaker.oscog-eem.1178 through the use of XCHG instruction in CogIA32Compiler>>genPushRegisterArgsForNumArgs:. Since SendNumArgsReg is not live with small arity sends it can replace TempReg. Spur: Fix bad C bug in nextInSortedFreeListLink:given:. The result type must be usqInt, otherwise the list can be wrongly terminated and the second compaction pass can segfault. Change printFreeChunk:isNextChunk: to printFreeChunk:printAsTreeNode:, and use it to implement printSortedFreeList. Firm up the checkTraversableSortedFreeList assert routine to check that the list is traversable from lastFreeChunk, not just firstFreeChunk. Make checkTraversableSortedFreeList robust in presence of empty sorted free list. Slang: Make promoteArithmeticTypes:and: obey C99's promotion rules more closely in an effort to generate more stable sources. Types were flipping between sqInt & usqInt for variables that were assigned both types, so that one generation would produce one type and a subsequent one another (Set/Dictionary hashing?). -- best, Eliot

May 21, 2015

Re: [Pharo-dev] can someone explain me this?

by Marcus Denker

> On 21 May 2015, at 22:14, Esteban Lorenzano <estebanlm(a)gmail.com> wrote: > > Smalltalk allClassesAndTraits size. "5057" > Object withAllSubclasses size. â9900" > > I really do not understand :) > Metaclasses. For every X there is "X classâ which is (eventually) a subclass of Object, too. Marcus

May 21, 2015

Re: [Pharo-dev] Morphic menus

by Matthieu Lacaton

Yeah it is really no big deal. Just some minor things I noticed while working my way through all those morphics ;) Le 21 mai 2015 21:10, "Nicolai Hess" <nicolaihess(a)web.de> a Ã©crit : > > > 2015-05-21 16:50 GMT+02:00 Matthieu Lacaton <matthieu.lacaton(a)gmail.com>: > >> Hello everyone, >> >> I had to understand how menus work in Morphic lately and I noticed some >> things : >> >> 1) When a menu has strictly more than 1 submenu active it cannot regain >> the focus if the mouse doesn't pass over the submenus first. It is a bit >> hard for me to explain so I will give you an example : >> Open a Nautilus browser on any package and right click on a class. Now >> browse "Refactoring => Code Rewriting => Rewrite Code but don't click on >> any of these items. You will have 3 menus. Now if you exit the last menu >> without returning to the previous one (make a loop from the right for >> example if you have enough place) and go back directly to the first one >> you'll see that it doesn't respond (it is like frozen). >> >> I don't know if it is a bug or if it is intended but I find it strange as >> it is not how it is working on Windows nor Ubuntu for example. >> >> So if it is indeed a bug, what I propose as a "fix" would be this: >> >> Replace >> >> *:* >>> >>> *MenuItemMorph >>* *activateOwner Menu: evt* >>> "Activate our owner menu; e.g., pass control to it" >>> owner ifNil:[^false]. "not applicable" >>> (owner fullContainsPoint: evt position) ifFalse:[^false]. >>> owner activate: evt. >>> ^true >>> >> >> by: >> >> *MenuItemMorph >>* *activateOwner Menu: evt* >>> >> >> >>> >>> * | popUpOwner |* >>> >>> * owner ifNil: [^false].* >>> >>> * (owner fullContainsPoint: evt position) ifTrue: [* >>> >>> * owner activate: evt.* >>> >>> >>> >>> * ^true. ].* >>> >>> * popUpOwner := owner popUpOwner.* >>> >>> * [ popUpOwner ] whileNotNil: [* >>> >>> * (popUpOwner owner fullContainsPoint: evt position) ifTrue: [* >>> >>> * popUpOwner owner activate: evt.* >>> >>> >>> * ^true. ]* >>> >>> * ifFalse: [* >>> >>> >>> >>> * popUpOwner := popUpOwner owner popUpOwner. ]. ].* >>> >>> * ^false.* >>> >> >> It should work with any number of submenus but i couldn't find anywhere >> in Pharo with more than 3 menus so i couldn't test it further. >> > > > Works good. > > > >> >> 2) When you leave a submenu and quickly come back to its popUpOwner, the >> item in the submenu won't be deselected. So if you navigate a bit in the >> menu and then want to come back to this submenu, you won't be able to >> select the item by passing your mouse over it. It is not a big deal in >> 99.9% of the time as you will still be able to click on it but when this >> item is supposed to open a submenu it won't. You need to go to another item >> in the submenu and to come back. >> >> I noticed that it happened because if you are leaving the submenu too >> quickly there is no "mouseLeave" event generated so the itemMenu is not >> deselected. Thus, as it is considered as selected, if you come back to it >> later a "mouseEnter" event will not trigger anything. >> >> It is quite easy to reproduce with the same menu than for the 1) : >> Right click on a class in a Nautilus browser then browse Refactoring => >> Code Rewriting. >> Now if you quickly come back to Refactoring sometimes you'll see that >> Code Rewriting won't be deselected (it will still be blue after the 500ms >> time out). If it is the case you can go back on "Rename..." and then if you >> quickly come back to Refactoring => Code Rewriting, you won't be able to >> get the 3rd submenu. >> >> I can reproduce this very easily but it is not really a problem because >> it only happens if you are fast with your mouse. >> >> I did not try to fix it but I guess that reinitializing all the itemMenus >> to unselected when poping a menu could do the trick. >> >> 3) This one is mostly insignificant because it is quite hard to reproduce >> and doesn't trigger often at all. But sometimes when you quickly navigate >> between 3 menus (I have not been able to reproduce it with only 2 menus) >> when you want to go back to the first one it freezes and it doesn't handle >> the mouseOver events anymore (you can still click though). >> This is not the same thing than the first point because here you >> correctly pass through the previous submenus to go to the first one. >> I don't really know why this happens sometimes but I think it may be >> because the focus was not passed quickly enough between windows and one on >> the submenus gets deactivated and removed from the world wheras it still >> had the focus on him. >> >> Anyway these 3 things are not really important but I just wanted to share >> what i found with you in case you think it is valid. >> > > I find both difficult to reproduce, I can, but only on every ~ 10th > attempt. > And you can always click on the menu or open a new one. > > > >> >> Cheers, >> >> Matthieu >> >> >> > >

May 21, 2015

Re: [Pharo-dev] can someone explain me this?

by Nicolas Cellier

You should try something like (Object withAllSubclasses reject: #isMeta) size... Also maybe start with ProtoObject... In Squeak, I have ((Smalltalk allClassesAndTraits reject: #isTrait) difference: (ProtoObject withAllSubclasses reject: #isMeta)) size = 0 2015-05-21 22:14 GMT+02:00 Esteban Lorenzano <estebanlm(a)gmail.com>: > Smalltalk allClassesAndTraits size. "5057" > Object withAllSubclasses size. â9900" > > I really do not understand :) > > Esteban >

May 21, 2015