Squeak-dev November 2010

squeak-dev@lists.squeakfoundation.org

77 participants
254 discussions

The Inbox: Monticello-bp.411.mcz
by commits＠source.squeak.org 21 Nov '10

21 Nov '10

A new version of Monticello was added to project The Inbox: http://source.squeak.org/inbox/Monticello-bp.411.mcz ==================== Summary ==================== Name: Monticello-bp.411 Author: bp Time: 21 November 2010, 6:38:42.624 pm UUID: cb55688c-2329-45c5-9d98-8df274e13a43 Ancestors: Monticello-bp.410 - Changed MCWorkingCopy description so that the package name is not duplicated when, e.g. Kernel (dtl.517) instead of Kernel (Kernel-dtl.517). This makes the Monticello Browser easier to use. =============== Diff against Monticello-bp.410 =============== Item was added: + ----- Method: MCAncestry>>ancestorStringWithout: (in category 'ancestry') ----- + ancestorStringWithout: packageName + ^ String streamContents: + [:s | self ancestors do: [:ea | s nextPutAll: (ea nameWithout: packageName)] separatedBy: [s nextPutAll: ', ']]! Item was added: + ----- Method: MCVersionInfo>>nameWithout: (in category 'accessing') ----- + nameWithout: packageName + | result | + result := self name. + (result startsWith: packageName , '-') ifTrue: [ + result := result copyFrom: packageName size + 2 to: result size]. + ^result! Item was changed: ----- Method: MCWorkingCopy>>description (in category 'accessing') ----- description + ^ self packageNameWithStar, ' (', (ancestry ancestorStringWithout: self packageName), ')'! - ^ self packageNameWithStar, ' (', ancestry ancestorString, ')'!

1 0

The Inbox: Monticello-bp.410.mcz
by commits＠source.squeak.org 21 Nov '10

21 Nov '10

A new version of Monticello was added to project The Inbox: http://source.squeak.org/inbox/Monticello-bp.410.mcz ==================== Summary ==================== Name: Monticello-bp.410 Author: bp Time: 21 November 2010, 5:14:50.468 pm UUID: d1fd1b50-24dc-4eb1-8938-ea167b731a79 Ancestors: Monticello-ul.409 - printOn: for MCWorkingCopy prints the package name =============== Diff against Monticello-ul.409 =============== Item was added: + ----- Method: MCWorkingCopy>>printOn: (in category 'accessing') ----- + printOn: stream + super printOn: stream. + stream + nextPut: $(; + nextPutAll: self packageName; + nextPut: $)!

1 0

KD-Tree
by Udo Schneider 21 Nov '10

21 Nov '10

All, is there a KD-Tree implementation for Squeak/Pharo (or Smalltalk in general) anywhere? I tried SqueakSource w/o success. Best Regards, Udo

1 0

ComplexCondition on squeaksource
by Enrico Spinielli 21 Nov '10

21 Nov '10

Hi all, I have created an open for all project with ComplexCondition as described in Andres Valloud's book "A Mentoring Course on Smalltalk". http://www.squeaksource.com/ComplexCondition This class will allow to write very intention-revealing code. Have a look at the tests: a pearl of design IMHO. Hope it helps Bye -- Enrico Spinielli "Do Androids dream of electric sheep?"— Philip K. Dick "Hear and forget; see and remember;do and understand."—Mitchel Resnick

1 0

OSX squeak crash: maximal size of image?
by Michal 21 Nov '10

21 Nov '10

My image has grown to 480MB and needs to grow still at least to the double of that (using squeak for research on language, involving huge language samples). I am now starting to get low-space notifiers, and squeak crashing into thin air. I am on OSX with 4G RAM, squakVM 4.2.4b1, squeak 4.2b -- is there a hard limit on the size of the image? Is there a way for me to get beyond that limit? thanks, Michal

3 4

New Tutorial on Events in Squeak, Pharo and Cuis
by Chris Cunnington 21 Nov '10

21 Nov '10

That was a very useful tutorial and it has increased my knowledge about event management in Squeak. There's an interesting kind of Catch-22 about this kind of learning: you need to look at it once clearly; and then you spend the rest of your career avoiding it. It's sort of like Assembly. The mean age of those selectors is the year 2000. In the decade since then abstractions have risen over these details. Tweak has events attached to instvars to handle dependency. Etoys has scripting tiles that can be used productively with StackMorph to create HyperCard. (If you don't like HyperCard, I think the scripting tools in Etoys could be used quite powerfully for things not linked to education.) The last ten years has seen abstractions for scripting or UI developments that would pull me away from actually developing in the manner of the tutorial. If more tutorials like this are made, I'll work through them. But I don't think I'll develop that way. Chris

2 1

Daily Commit Log
by commits＠source.squeak.org 20 Nov '10

20 Nov '10

Changes to Trunk (http://source.squeak.org/trunk.html) in the last 24 hours: http://lists.squeakfoundation.org/pipermail/packages/2010-November/003937.h… Name: Compiler-nice.181 Ancestors: Compiler-eem.180 Let (Scanner isLiteralSymbol: #<) be true ============================================= http://lists.squeakfoundation.org/pipermail/packages/2010-November/003938.h… Name: Graphics-tbn.157 Ancestors: Graphics-ul.156 Fix http://code.google.com/p/pharo/issues/detail?id=3225 also in Squeak Thx to jbaptiste.arnaud: =============================================

1 0

[BUG] Autocompletition in OmniBrowser
by Seckler, Eric 19 Nov '10

19 Nov '10

Hiho everyone, first of all, I'm quite a newbie to Squeak - I beg your pardon for any mistakes on my account :-) I think I have spotted a bug in OmniBrowser's autocompletition feature in the current Squeak trunk image. In one of my classes, whenever I type any instance variable in a method and hit the space key followed by any other key (in order to type the message to be send to that instance variable), the attached error rises up. It seems like SqueakInstvarInterfaceExtractor>>extractInterfacesFrom:addTo: is broken. It's called with the following method: RRBoard>>createFields fields := Matrix new: RRConstants boardSize tabulate: [ :y :x | self newFieldAt: x at: y ] The method nativeSend:numArgs: in the same class finally crashes when it is called with this method and na=2: RRBoard>>newFieldAt: x at: y | field | field := RRField newForBoard: self. self addMorph: field. field position: ((x - 1) * RRConstants fieldSize) @ ((y - 1) * RRConstants fieldSize); borderOfClass: (self fieldBorderClassAt: x at: y). ^ field The method SqueakTypeCollector>>handleAssignment:forTmp:in: is called with the same method and val=#temp->1 and index=0. The array which is causing the exception contains only one element: a ExtractedTypeForTmp instance with the following instance variables: ivarClass: RRBoard ivarName: 1 tag: nil nbOfCorrectTypes: nil nbOfIncorrectTypes: nil interface: an IdentitySet() assignments: an OrderedCollection() extractedTypes: nil links: a Set() compiledMethod: (RRBoard>>#createFields "a CompiledMethod(1220)") I'm using a 4.2 trunk image prepared by the Software Architecture Group at our university. If you need any more information or source code please contact me. Thanks for your help Eric ----- Eric Seckler Rembrandtstr. 8 14612 Falkensee phone: +49-3322-236795 mobile: +49-151-22948869 email: eric.seckler(a)student.hpi.uni-potsdam.de email2: e.seckler(a)forge2.de -------------------------------------------------------------------------------- 19 November 2010 9:13:07.475 pm VM: Win32 - Smalltalk Image: Squeak4.2alpha [latest update: #7] SecurityManager state: Restricted: false FileAccess: true SocketAccess: true Working Dir P:\docs\Studium\HPI\Lehrveranstaltungen\S1011\Softwarearchitektur I\Squeak\SWA-2010.app\Contents\Resources Trusted Dir P:\docs\Studium\HPI\Lehrveranstaltungen\S1011\Softwarearchitektur I\Squeak\SWA-2010.app\Contents\Resources\betasheet Untrusted Dir C:\Users\betasheet\Documents\My Squeak Array(Object)>>error: Receiver: {ExtractedType: Sends: {} Assignments: {} } Arguments and temporary variables: aString: 'subscript is out of bounds: 2' Receiver's instance variables: {ExtractedType: Sends: {} Assignments: {} } Array(Object)>>errorSubscriptBounds: Receiver: {ExtractedType: Sends: {} Assignments: {} } Arguments and temporary variables: index: 2 Receiver's instance variables: {ExtractedType: Sends: {} Assignments: {} } Array(Object)>>at: Receiver: {ExtractedType: Sends: {} Assignments: {} } Arguments and temporary variables: index: 2 Receiver's instance variables: {ExtractedType: Sends: {} Assignments: {} } SqueakTypeCollector(TypeCollector)>>handleAssignment:forTmp:in: Receiver: a SqueakTypeCollector Arguments and temporary variables: val: #temp->1 index: 0 aCompiledMethod: (RRBoard>>#newFieldAt:at: "a CompiledMethod(3951)") Receiver's instance variables: theClass: RRBoard instVars: #(#bounds #owner #submorphs #fullBounds #color #extension #borderWidt...etc... typingResults: {ExtractedType: Sends: {} Assignments: {} . ExtractedType: Send...etc... currentExtractedMethod: (RRBoard>>#createFields "a CompiledMethod(1220)") localTypingResults: an IdentityDictionary((RRBoard>>#adjustSize "a CompiledMeth...etc... --- The full stack --- Array(Object)>>error: Array(Object)>>errorSubscriptBounds: Array(Object)>>at: SqueakTypeCollector(TypeCollector)>>handleAssignment:forTmp:in: - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - [] in SqueakInstvarInterfaceExtractor(InstvarInterfaceExtractor)>>nativeSend:numArgs: Array(SequenceableCollection)>>withIndexDo: Array(SequenceableCollection)>>doWithIndex: SqueakInstvarInterfaceExtractor(InstvarInterfaceExtractor)>>nativeSend:numArgs: SqueakInstvarInterfaceExtractor>>send:numArgs: SqueakInstvarInterfaceExtractor>>send:super:numArgs: InstructionStream>>interpretNextInstructionFor: SqueakInstvarInterfaceExtractor(InstvarInterfaceExtractor)>>extractInterfacesFrom:addTo: [] in SqueakTypeCollector(TypeCollector)>>typeInstvarsOfClass: MethodDictionary>>keysAndValuesDo: RRBoard class(Behavior)>>selectorsAndMethodsDo: SqueakTypeCollector(TypeCollector)>>typeInstvarsOfClass: TypeCollector class>>typeInstvarsOfClass: OContext(ECContext)>>typeInfosFor: OContext(ECContext)>>guessWithRoelTyper:class: OContext(ECContext)>>guessInstVarClass: OContext(ECContext)>>guessTypeOf: OContext(ECContext)>>computeReceiverClass OContext(ECContext)>>compute OContext(ECContext)>>setController:class:source:position: OContext class(ECContext class)>>controller:class:source:position: OController>>openMenuFor: OController>>handleKeystrokeBefore:editor: ECToolSet class>>codeCompletionAround:textMorph:keyStroke: ToolSet class>>codeCompletionAround:textMorph:keyStroke: OBTextMorphWithShout(TextMorph)>>keyStroke: OBTextMorphWithShout(TextMorphForEditView)>>keyStroke: OBTextMorphWithShout(TextMorph)>>handleKeystroke: KeyboardEvent>>sentTo: OBTextMorphWithShout(Morph)>>handleEvent: OBTextMorphWithShout(Morph)>>handleFocusEvent: [] in HandMorph>>sendFocusEvent:to:clear: [] in PasteUpMorph>>becomeActiveDuring: BlockClosure>>on:do: PasteUpMorph>>becomeActiveDuring: HandMorph>>sendFocusEvent:to:clear: HandMorph>>sendEvent:focus:clear: HandMorph>>sendKeyboardEvent: HandMorph>>handleEvent: HandMorph>>processEvents [] in WorldState>>doOneCycleNowFor: Array(SequenceableCollection)>>do: WorldState>>handsDo: WorldState>>doOneCycleNowFor: WorldState>>doOneCycleFor: PasteUpMorph>>doOneCycle [] in Project class>>spawnNewProcess [] in BlockClosure>>newProcess

1 0

The Trunk: Graphics-tbn.157.mcz
by commits＠source.squeak.org 19 Nov '10

19 Nov '10

Torsten Bergmann uploaded a new version of Graphics to project The Trunk: http://source.squeak.org/trunk/Graphics-tbn.157.mcz ==================== Summary ==================== Name: Graphics-tbn.157 Author: tbn Time: 19 November 2010, 9:24:53.858 pm UUID: 0e6ea246-6d40-ce4c-a609-ae1e5b4a82c4 Ancestors: Graphics-ul.156 Fix http://code.google.com/p/pharo/issues/detail?id=3225 also in Squeak Thx to jbaptiste.arnaud: =============== Diff against Graphics-ul.156 =============== Item was changed: ----- Method: Form>>fadeImage:at:indexAndMaskDo: (in category 'transitions') ----- fadeImage: otherImage at: topLeft indexAndMaskDo: indexAndMaskBlock + "This fade uses halftones as a blending hack. Zeros in the halftone produce the original image (self), and ones in the halftone produce the 'otherImage'. IndexAndMaskBlock gets evaluated prior to each cycle, and the resulting boolean determines whether to continue cycling." + | index imageRect maskForm resultForm pix returnPix| - | index imageRect maskForm resultForm | imageRect := otherImage boundingBox. resultForm := self copy: (topLeft extent: imageRect extent). maskForm := Form extent: 32@32. index := 0. + [indexAndMaskBlock value: (index := index+1) value: maskForm] whileTrue: [maskForm reverse. resultForm copyBits: imageRect from: resultForm at: 0@0 clippingBox: imageRect rule: Form over fillColor: maskForm. maskForm reverse. resultForm copyBits: imageRect from: otherImage at: 0@0 clippingBox: imageRect rule: Form under fillColor: maskForm. self copyBits: imageRect from: resultForm at: topLeft clippingBox: self boundingBox rule: Form over fillColor: nil. Display forceDisplayUpdate]! Item was changed: ----- Method: Form>>fadeImageCoarse:at: (in category 'transitions') ----- fadeImageCoarse: otherImage at: topLeft "Display fadeImageCoarse: (Form fromDisplay: (40@40 extent: 300@300)) reverse at: 40@40" + | d pix| - | d | d := self depth. ^ self fadeImage: otherImage at: topLeft indexAndMaskDo: + [:i :mask | | j | - [:i :mask | - | pix j | i=1 ifTrue: [pix := (1 bitShift: d) - 1. + 1 to: 8//d-1 do: [:q | pix := pix bitOr: (pix bitShift: d*4)]] . + i <= 16 ifTrue:[ + j := i-1//4+1. - 1 to: 8//d-1 do: [:q | pix := pix bitOr: (pix bitShift: d*4)]]. - i <= 16 ifTrue: - [j := i-1//4+1. (0 to: 28 by: 4) do: [:k | mask bits at: j+k + put: (pix bitOr: (mask bits at: j+k))]. + true] - put: ((mask bits at: j+k) bitOr: (pix bitShift: i-1\\4*d))]. - "mask display." true] ifFalse: [false]]! Item was changed: ----- Method: Form>>fadeImageFine:at: (in category 'transitions') ----- fadeImageFine: otherImage at: topLeft "Display fadeImageFine: (Form fromDisplay: (40@40 extent: 300@300)) reverse at: 40@40" + | d pix| - | d | d := self depth. ^ self fadeImage: otherImage at: topLeft indexAndMaskDo: + [:i :mask | | j ii | - [:i :mask | - | pix j ii | i=1 ifTrue: [pix := (1 bitShift: d) - 1. 1 to: 8//d-1 do: [:q | pix := pix bitOr: (pix bitShift: d*4)]]. i <= 16 ifTrue: [ii := #(0 10 2 8 7 13 5 15 1 11 3 9 6 12 4 14) at: i. j := ii//4+1. (0 to: 28 by: 4) do: [:k | mask bits at: j+k put: + ((mask bits at: j+k) bitOr: (pix))]. - ((mask bits at: j+k) bitOr: (pix bitShift: ii\\4*d))]. true] ifFalse: [false]]! Item was changed: ----- Method: Form>>fadeImageSquares:at: (in category 'transitions') ----- fadeImageSquares: otherImage at: topLeft "Display fadeImageSquares: (Form fromDisplay: (40@40 extent: 300@300)) reverse at: 40@40" ^ self fadeImage: otherImage at: topLeft indexAndMaskDo: [:i :mask | mask fill: ((16-i) asPoint extent: (i*2) asPoint) fillColor: Color black. i <= 16]! Item was changed: ----- Method: Form>>fadeImageVert:at: (in category 'transitions') ----- fadeImageVert: otherImage at: topLeft "Display fadeImageVert: (Form fromDisplay: (10@10 extent: 300@300)) reverse at: 10@10" | d | d := self depth. ^ self fadeImage: otherImage at: topLeft indexAndMaskDo: [:i :mask | mask fill: ((mask width//2//d-i*d)@0 extent: i*2*d@mask height) fillColor: Color black. i <= (mask width//d)]!

1 0

Re: [squeak-dev] The Trunk: Compiler-nice.181.mcz
by Eliot Miranda 19 Nov '10

19 Nov '10

Thanks, Nicholas! On Fri, Nov 19, 2010 at 8:32 AM, <commits(a)source.squeak.org> wrote: > Nicolas Cellier uploaded a new version of Compiler to project The Trunk: > http://source.squeak.org/trunk/Compiler-nice.181.mcz > > ==================== Summary ==================== > > Name: Compiler-nice.181 > Author: nice > Time: 19 November 2010, 5:32:20.688 pm > UUID: 86d69779-755f-0548-8a95-11173c3b8370 > Ancestors: Compiler-eem.180 > > Let (Scanner isLiteralSymbol: #<) be true > > =============== Diff against Compiler-eem.180 =============== > > Item was changed: > ----- Method: Scanner class>>isLiteralSymbol: (in category 'testing') > ----- > isLiteralSymbol: aSymbol > "Test whether a symbol can be stored as # followed by its > characters. > Symbols created internally with asSymbol may not have this property, > e.g. '3' asSymbol." > > | i ascii type next last | > i := aSymbol size. > i = 0 ifTrue: [^ false]. > > "TypeTable should have been origined at 0 rather than 1 ..." > ascii := (aSymbol at: 1) asciiValue. > type := TypeTable at: ascii ifAbsent: [^false]. > type == #xLetter ifTrue: [ > next := last := nil. > [i > 1] > whileTrue: > [ascii := (aSymbol at: i) > asciiValue. > type := TypeTable at: ascii > ifAbsent: [^false]. > (type == #xLetter or: [type == > #xDigit or: [type == #xColon > and: [ > next == nil > > ifTrue: [last := #xColon. true] > > ifFalse: [last == #xColon and: [next ~~ #xDigit and: [next ~~ > #xColon]]]]]]) > ifFalse: [^ false]. > next := type. > i := i - 1]. > + ^ true]. > + type == #xBinary ifTrue: > + [^(2 to: i) allSatisfy: [:j | > - ^ true]. > - type == #xBinary ifTrue: [^i = 1]. "Here we could extend to > - ^(2 to: i) allSatisfy: [:j | > ascii := (aSymbol at: j) asciiValue. > + (TypeTable at: ascii ifAbsent: []) == #xBinary]]. > - (TypeTable at: ascii ifAbsent: []) == #xBinary]" > type == #verticalBar ifTrue: [^i = 1]. > ^false! > > >

1 0

← Newer
1
...
7
8
9
10
11
12
13
...
26
Older →

Jump to page: