- Pharo-dev - lists.pharo.org

[Pharo-project] multiple pharo images
by Francisco Garau May 3, 2012

May 3, 2012

6 5

[Pharo-project] Fwd: [squeak-dev] The Trunk: Kernel-eem.684.mcz
by StÃ©phane Ducasse May 3, 2012

May 3, 2012

Begin forwarded message: > From: pharo-project-owner(a)lists.gforge.inria.fr > Subject: Fwd: [squeak-dev] The Trunk: Kernel-eem.684.mcz > Date: April 30, 2012 6:48:24 PM GMT+02:00 > To: stephane.ducasse(a)gmail.com > > You are not allowed to post to this mailing list, and your message has > been automatically rejected. If you think that your messages are > being rejected in error, contact the mailing list owner at > pharo-project-owner(a)lists.gforge.inria.fr. > > > From: stephane ducasse <stephane.ducasse(a)gmail.com> > Subject: Fwd: [squeak-dev] The Trunk: Kernel-eem.684.mcz > Date: April 30, 2012 6:46:40 PM GMT+02:00 > To: Pharo Development <Pharo-project(a)lists.gforge.inria.fr> > > > > > Begin forwarded message: > >> From: commits(a)source.squeak.org >> Subject: [squeak-dev] The Trunk: Kernel-eem.684.mcz >> Date: April 26, 2012 6:07:56 PM GMT+02:00 >> To: squeak-dev(a)lists.squeakfoundation.org, packages(a)lists.squeakfoundation.org >> Reply-To: squeak-dev(a)lists.squeakfoundation.org >> >> Eliot Miranda uploaded a new version of Kernel to project The Trunk: >> http://source.squeak.org/trunk/Kernel-eem.684.mcz >> >> ==================== Summary ==================== >> >> Name: Kernel-eem.684 >> Author: eem >> Time: 26 April 2012, 11:07:37.737 am >> UUID: e838a325-27ec-4a62-a907-d4059451a046 >> Ancestors: Kernel-nice.683 >> >> Implement endPC numArgs and numTemps for closures >> and contexts (both block and method activations). >> Provide CompiledMethod>abstractBytecodeMessagesDo: >> et al. >> Implement BlockClosure>isClean to identify self-contained blocks. >> >> =============== Diff against Kernel-nice.683 =============== >> >> Item was added: >> + ----- Method: BlockClosure>>abstractBytecodeMessagesDo: (in category 'scanning') ----- >> + abstractBytecodeMessagesDo: aBlock >> + "Evaluate aBlock with the sequence of abstract bytecodes in the receiver." >> + self method >> + abstractBytecodeMessagesFrom: startpc >> + to: self endPC >> + do: aBlock >> + >> + "| msgs | >> + msgs := OrderedCollection new. >> + (SortedCollection sortBlock: [:a :b| a compare: b caseSensitive: false]) sortBlock >> + abstractBytecodeMessagesDo: [:msg| msgs add: msg selector]. >> + msgs"! >> >> Item was added: >> + ----- Method: BlockClosure>>blockCreationBytecodeMessage (in category 'scanning') ----- >> + blockCreationBytecodeMessage >> + "Answer the abstract bytecode message that created the receiver." >> + | blockCreationBytecodeSize | >> + ^self method abstractBytecodeMessageAt: startpc - (blockCreationBytecodeSize := 4) >> + >> + "(SortedCollection sortBlock: [:a :b| a compare: b caseSensitive: false]) sortBlock blockCreationBytecodeMessage"! >> >> Item was added: >> + ----- Method: BlockClosure>>endPC (in category 'accessing') ----- >> + endPC >> + ^self blockCreationBytecodeMessage arguments last + startpc - 1! >> >> Item was changed: >> ----- Method: BlockClosure>>hasMethodReturn (in category 'testing') ----- >> hasMethodReturn >> "Answer whether the receiver has a method-return ('^') in its code." >> + | scanner endpc | >> + scanner := InstructionStream new method: outerContext method pc: startpc. >> + endpc := self endPC. >> + scanner scanFor: [:byte | (byte between: 120 and: 124) or: [scanner pc > endpc]]. >> + ^scanner pc <= endpc! >> - | myMethod scanner preceedingBytecodeMessage end | >> - "Determine end of block from the instruction preceding it. >> - Find the instruction by using an MNU handler to capture >> - the instruction message sent by the scanner." >> - myMethod := outerContext method. >> - scanner := InstructionStream new method: myMethod pc: myMethod initialPC. >> - [scanner pc < startpc] whileTrue: >> - [[scanner interpretNextInstructionFor: nil] >> - on: MessageNotUnderstood >> - do: [:ex| preceedingBytecodeMessage := ex message]]. >> - end := preceedingBytecodeMessage arguments last + startpc - 1. >> - scanner method: myMethod pc: startpc. >> - scanner scanFor: [:byte | (byte between: 120 and: 124) or: [scanner pc > end]]. >> - ^scanner pc <= end! >> >> Item was added: >> + ----- Method: BlockClosure>>isClean (in category 'testing') ----- >> + isClean >> + "Answer if the receiver does not close-over any variables other than globals, and does >> + not ^-return (does not close over the home context). Clean blocks are amenable to >> + being created at compile-time." >> + self numCopiedValues > 0 ifTrue: >> + [^false]. >> + self abstractBytecodeMessagesDo: >> + [:msg| >> + (#( pushReceiver >> + pushReceiverVariable: popIntoReceiverVariable: storeIntoReceiverVariable: >> + methodReturnConstant: methodReturnReceiver methodReturnTop) >> + includes: msg selector) ifTrue: >> + [^false]]. >> + ^true >> + >> + "clean:" >> + "[] isClean" >> + "[:a :b| a < b] isClean" >> + "unclean" >> + "[^nil] isClean" >> + "[self class] isClean" >> + "| v | v := 0. >> + [v class] isClean"! >> >> Item was added: >> + ----- Method: BlockClosure>>numTemps (in category 'accessing') ----- >> + numTemps >> + "Answer the number of temporaries for the receiver; this includes >> + the number of arguments and the number of copied values." >> + | blockCreationBytecodeSize | >> + ^self numCopiedValues >> + + self numArgs >> + + (BlockLocalTempCounter >> + tempCountForBlockAt: startpc - (blockCreationBytecodeSize := 4) >> + in: self method)! >> >> Item was added: >> + ----- Method: CompiledMethod>>abstractBytecodeMessageAt: (in category 'scanning') ----- >> + abstractBytecodeMessageAt: pc >> + "Answer the abstract bytecode message at pc in the receiver." >> + ^[(InstructionStream new method: self pc: pc) interpretNextInstructionFor: nil] >> + on: MessageNotUnderstood >> + do: [:ex| ex message]! >> >> Item was added: >> + ----- Method: CompiledMethod>>abstractBytecodeMessagesDo: (in category 'scanning') ----- >> + abstractBytecodeMessagesDo: aBlock >> + "Evaluate aBlock with the sequence of abstract bytecodes in the receiver" >> + self abstractBytecodeMessagesFrom: self initialPC >> + to: self endPC >> + do: aBlock >> + >> + "| msgs | >> + msgs := OrderedCollection new. >> + CompiledMethod >> #abstractBytecodeMessagesFrom:to: abstractBytecodeMessagesDo: >> + [:msg| msgs add: msg selector]. >> + msgs"! >> >> Item was added: >> + ----- Method: CompiledMethod>>abstractBytecodeMessagesFrom:to:do: (in category 'scanning') ----- >> + abstractBytecodeMessagesFrom: startpc to: endpc do: aBlock >> + "Evaluate aBlock with the sequence of abstract bytecodes from startpc through endpc in the receiver" >> + | scanner | >> + scanner := InstructionStream new method: self pc: startpc. >> + [scanner pc <= endpc] whileTrue: >> + [[scanner interpretNextInstructionFor: nil] >> + on: MessageNotUnderstood >> + do: [:ex| aBlock value: ex message]] >> + >> + "| m msgs | >> + msgs := OrderedCollection new. >> + (m := CompiledMethod >> #abstractBytecodeMessagesFrom:to:) >> + abstractBytecodeMessagesFrom: m initialPC >> + to: m endPC >> + do: [:msg| msgs add: msg selector]. >> + msgs"! >> >> Item was added: >> + ----- Method: MethodContext>>endPC (in category 'private') ----- >> + endPC >> + ^closureOrNil >> + ifNil: [self method endPC] >> + ifNotNil: [closureOrNil endPC]! >> >> Item was added: >> + ----- Method: MethodContext>>numArgs (in category 'accessing') ----- >> + numArgs >> + "Answer the number of arguments for this activation." >> + ^closureOrNil >> + ifNil: [method numArgs] >> + ifNotNil: [closureOrNil numArgs]! >> >> Item was added: >> + ----- Method: MethodContext>>numTemps (in category 'accessing') ----- >> + numTemps >> + "Answer the number of temporaries for this activation; this includes >> + the number of arguments, and for blocks, the number of copied values." >> + ^closureOrNil >> + ifNil: [method numTemps] >> + ifNotNil: [closureOrNil numTemps]! >> >> > > > >

2 2

[Pharo-project] ZTimestamp
by Sven Van Caekenberghe May 3, 2012

May 3, 2012

Hi All, I don't like DateAndTime and/or TimeStamp a lot because it forces nanosecond precision and timezones on me, while in most cases I only want second precision and no timezones. Futhermore the class has some quirks that make it hard to understand. So last month I wrote my own, independent, more efficient ZTimestamp. As an experiment, I am releasing it like this: https://github.com/svenvc/ztimestamp I think Dale is really on to something with filetree, git and github. Check it out! Regards, Sven -- Sven Van Caekenberghe http://stfx.eu Smalltalk is the Red Pill

1 0

[Pharo-project] Adding hooks for well known events in morphic
by Guillermo Polito May 3, 2012

May 3, 2012

I've made up a list with all the useful morphic widget I've found in a alpha alpha alpha 2.0 image. I'd like to put in those widgets empty hooks like: Morph>>triggerOnShow "hook" for common events widgets provide usually in MVC frameworks. That means *not changing the whole interface nor refactoring everything*. *Just adding the hooks*. What we were talking with Mariano (I hate you for putting ideas in my head) today in the train was to add those empty hooks and put senders for them, what should be easy. This way we can start to homogenize morphic a bit. Most likely, in a not so far future, those empty hooks could be replaced by something like: Morph>>triggerOnShow self announcer announce: (MorphShowAnnouncement show: self) Now, here is a list of the most important widgets i've found (I could've missed some) and events they should trigger. I'm not taking into account new multitouch interfaces and stuff. For all widgets onShow onDelete onKeyUp onKeyPress onKeyDown onClick onMouseDown onMouseUp onDoubleClick onDragged onDropped onMove? onResize? For all containers onChildAdded onChildRemoved "Widgets" Check/Radio -> CheckboxButtonMorph/SimpleSwitchMorph onSelectionChanged List/Grids -> PluggableListMorph/PluggableMultiColumnListMorph/LazyListMorph/MulticolumnLazyListMorph onSelectionChanged Button -> PluggableButtonMorph/SimpleButtonMorph/MultistateButtonMorph onClick Label -> LabelMorph/FuzzyLabelMorph /MultistateLabelMorph Progress Bar-> SystemProgressBarMorph Slider -> Slider/PluggableSliderMorph onSelectionChanged Text handling -> TextMorph/PluggableTextMorph onSelectionChanged Field Handling -> PluggableTextFieldMorph/PluggableTextEditorMorph Tree -> MorphTreeMorph onSelectionChanged onExpandItem onCollapseItem Combo box -> DropListMorph/EditableDropListMorph /SearchMorph/MorphDropListMorph onSelectionChanged Menues -> MenuMorph + friends onClick (for menu item probably) Image -> ImageMorph/ImagePreviewMorph Diff -> DiffMorph Color Presenter -> ColorPresenterMorph Color chooser -> ColorChooserMorph/HSVColorSelectorMorph / SVColorSelectorMorph onSelectionChanged Task var -> TaskbarMorph "Containers" ScrollPane Windows -> SystemWindow/PluggableSystemWindow /StandardWindow Dialog window -> DialogWindow & (lots of) subclasses Panel -> PanelMorph & subclasses / PluggablePanelMorph Tabs -> TabGroupMorph (Other Tab stuff -> PluggableTabBarMorph + PluggableTabButtonMorph ) onSelectionChanged Guille

5 8

[Pharo-project] What about making the configuration browser more visible?
by Guillermo Polito May 3, 2012

May 3, 2012

What about naming it "Package Loader" or "Package center" or something like that? And putting it in the main menu instead of inside of Tools? I want people who open pharo for the first time to say "Hey, here I can download stuff!!!" Guille

9 32

[Pharo-project] Mouse handler on a morph
by Erwan Douaille May 3, 2012

May 3, 2012

Hi. I would like to know if it's possible to add a mouse handler on a specific morph ? I have a window and i would like to know the position of my mouse only in this window. For exemple i also use : self on: #mouseDown send: #captureMouse: to: self. But it give me the mouse position in pharo environment and not only in my window. Thanks :) -- Douaille Erwan <douaille.erwan(a)gmail.com>

4 4

[Pharo-project] Enable #alwaysOpenFullDebugger: and fastDragging:
by Mariano Martinez Peck May 3, 2012

May 3, 2012

Hi guys. There are 2 settings that I ALWAYS need to change and I think it would be nice to have them by default: 1) Always show debugger and avoid pre-debugger windows. It makes no sense for development. It takes you the double of time. What does the pre-debug brings you? nothing. If you want to proceed, then the debugger also has that option. If you want to abandon, you can just close the debugger. So, we should enable this. If someone wanst the pre-debugger fwindows for a reasons, then he should enable it. 2) Fast dragging. The normal dragging is slow in old machines and slow when there are lots of opened windows. The fast drag is much faster and nicer because since it is transparent you can even see what is below. Let's do a quick survey. We have 24 hrs. Do not vote if you didn't try it. I start: Debugger alwaysOpenFullDebugger: true. +1 UITheme currentSettings fastDragging: true. +1 Cheers -- Mariano http://marianopeck.wordpress.com

4 3

Re: [Pharo-project] [update 2.0] #20038
by Marcus Denker May 3, 2012

May 3, 2012

On May 3, 2012, at 2:01 PM, StÃ©phane Ducasse wrote: > > On May 2, 2012, at 2:12 PM, Marcus Denker wrote: > >> Issue 5738: do allow the same ivar in the hierarchy >> http://code.google.com/p/pharo/issues/detail?id=5738 > > Do we really want to let the user define instance variables duplicated in the class hierarchy? > I think that this is not really a good move. But I did not browse the issue because I do not have internet. > How then do you move a variable up in the hierarchy? Marcus -- Marcus Denker -- http://marcusdenker.de

3 2

[Pharo-project] [update 1.4] #14440
by Esteban Lorenzano May 3, 2012

May 3, 2012

14440 ----- 5773: Have access to the original event in PLuggableButtonMorph http://code.google.com/p/pharo/issues/detail?id=5773 5584: File Browser not recognizing cs and st if filename has more than one point http://code.google.com/p/pharo/issues/detail?id=5584 5667: Renaming a file just deletes it due to missing #fileNameFromFormattedItem: http://code.google.com/p/pharo/issues/detail?id=5667

1 0

[Pharo-project] A simple getenv primitive
by Damien Cassou May 3, 2012

May 3, 2012

Hi, I can't find any getenv primitive that would allow me to access environment an variable value from its name. This could be really helpful to get the HOME directory for example. Does one such primitive exist? If not, why? It looks trivial to implement for linux, and it is potentially also easy on Windows and Mac OS. I'm willing to try if necessary. Thank you -- Damien Cassou http://damiencassou.seasidehosting.st "Lambdas are relegated to relative obscurity until Java makes them popular by not having them." James Iry

2 2