Squeak-dev June 2011

squeak-dev@lists.squeakfoundation.org

77 participants
236 discussions

The Inbox: StOMP-koba.1.mcz
by commits＠source.squeak.org 29 Jun '11

29 Jun '11

A new version of StOMP was added to project The Inbox: http://source.squeak.org/inbox/StOMP-koba.1.mcz ==================== Summary ==================== Name: StOMP-koba.1 Author: koba Time: 29 June 2011, 10:55:09.74 pm UUID: c20f39f1-2bbc-154e-9256-ad306cf5a596 Ancestors: empty log message ==================== Snapshot ====================

1 0

Re: Squeak-dev Digest, Vol 102, Issue 22
by Alberto Ferreira 29 Jun '11

29 Jun '11

Dear all ; Can i have an "HOW TO", to make a tile command/slot tiles, like a numeric value, but for suare root ? for squeak 4.1 and e-toys ? thanks in advanced Alberto Ferreira

2 1

Daily Commit Log
by commits＠source.squeak.org 28 Jun '11

28 Jun '11

Changes to Trunk (http://source.squeak.org/trunk.html) in the last 24 hours: http://lists.squeakfoundation.org/pipermail/packages/2011-June/004833.html Name: Kernel-nice.602 Ancestors: Kernel-bf.601 Minor tweaks: Use a better approximation for counting digits in base 10 Use raisedToInteger: since we now the argument 'base' is Integer Change arTanh error message because (between -1 and 1) is more inline with Smalltalk message #between:and: than (between 1 and -1). Simplify ulp since #abs matches #ulp return value for any non finite number. ============================================= http://lists.squeakfoundation.org/pipermail/packages/2011-June/004834.html Name: Graphics-nice.187 Ancestors: Graphics-ul.186 Apply Display/Character Scanner clean-ups from Cuis: - remove unused paragraph ivar from DisplayScanner - remove historical dead code from #crossedX - ifNil: simplifies code a bit ============================================= http://lists.squeakfoundation.org/pipermail/packages/2011-June/004835.html Name: Installer-Core-nice.350 Ancestors: Installer-Core-cmm.349 #anySatisfy: vs #detect:ifNone: Hint: when you ain't gonna use the element detected by #detect:ifNone:, it probably means you shall better be using one of #(anySatisfy: allStatisfy: noneSatisfy:). =============================================

1 0

The Trunk: Installer-Core-nice.350.mcz
by commits＠source.squeak.org 27 Jun '11

27 Jun '11

Nicolas Cellier uploaded a new version of Installer-Core to project The Trunk: http://source.squeak.org/trunk/Installer-Core-nice.350.mcz ==================== Summary ==================== Name: Installer-Core-nice.350 Author: nice Time: 27 June 2011, 8:44:47.278 pm UUID: 9c24a081-c6be-4425-ab27-1d8664fa16ce Ancestors: Installer-Core-cmm.349 #anySatisfy: vs #detect:ifNone: Hint: when you ain't gonna use the element detected by #detect:ifNone:, it probably means you shall better be using one of #(anySatisfy: allStatisfy: noneSatisfy:). =============== Diff against Installer-Core-cmm.349 =============== Item was changed: ----- Method: InstallerMonticello>>mcDetectFileBlock: (in category 'monticello') ----- mcDetectFileBlock: pkg pkg isString ifTrue: [ ^ [ :aFile | aFile beginsWith: pkg ] ]. (pkg isKindOf: Array) + ifTrue: [ ^ [ :aFile | pkg anySatisfy: [ :item | aFile beginsWith: item ] ] ]. - ifTrue: [ ^ [ :aFile | (pkg detect: [ :item | aFile beginsWith: item ] ifNone: [ false ]) ~= false ] ]. pkg isBlock ifTrue: [ ^ pkg ]. ! Item was changed: ----- Method: InstallerUniverse>>basicInstall (in category 'basic interface') ----- basicInstall self packages do: [ :packageName | | potentials pkg pkgAndVersion version | pkgAndVersion := self packageAndVersionFrom: packageName. pkg := pkgAndVersion first. version := pkgAndVersion last. potentials := universe packageVersionsForPackage: pkg. pkg := version isEmpty ifTrue: [ potentials last ] ifFalse: [ version := self classUVersion readFrom: version readStream. + (potentials anySatisfy: [ :p | p version = version]) ifFalse: [ ^ self error: 'version not found'] - potentials detect:[ :p | p version = version] ifNone: [ ^ self error: 'version not found'] ]. universe planToInstallPackage: pkg. ]. self uniDoInstall!

1 0

The Trunk: Graphics-nice.187.mcz
by commits＠source.squeak.org 27 Jun '11

27 Jun '11

Nicolas Cellier uploaded a new version of Graphics to project The Trunk: http://source.squeak.org/trunk/Graphics-nice.187.mcz ==================== Summary ==================== Name: Graphics-nice.187 Author: nice Time: 27 June 2011, 8:38:52.631 pm UUID: 3d0c74fa-7438-4139-8e2a-74979eee4a86 Ancestors: Graphics-ul.186 Apply Display/Character Scanner clean-ups from Cuis: - remove unused paragraph ivar from DisplayScanner - remove historical dead code from #crossedX - ifNil: simplifies code a bit =============== Diff against Graphics-ul.186 =============== Item was changed: ----- Method: CharacterBlockScanner>>characterBlockAtPoint:index:in: (in category 'scanning') ----- characterBlockAtPoint: aPoint index: index in: textLine "This method is the Morphic characterBlock finder. It combines MVC's characterBlockAtPoint:, -ForIndex:, and buildCharcterBlock:in:" | runLength lineStop stopCondition | line := textLine. rightMargin := line rightMargin. lastIndex := line first. self setStopConditions. "also sets font" characterIndex := index. " == nil means scanning for point" characterPoint := aPoint. (characterPoint isNil or: [characterPoint y > line bottom]) ifTrue: [characterPoint := line bottomRight]. (text isEmpty or: [(characterPoint y < line top or: [characterPoint x < line left]) or: [characterIndex notNil and: [characterIndex < line first]]]) ifTrue: [^ (CharacterBlock new stringIndex: line first text: text topLeft: line leftMargin@line top extent: 0 @ textStyle lineGrid) textLine: line]. destX := leftMargin := line leftMarginForAlignment: alignment. destY := line top. runLength := text runLengthFor: line first. + lineStop := characterIndex "scanning for index" + ifNil: [ line last ]. "scanning for point" - characterIndex - ifNotNil: [lineStop := characterIndex "scanning for index"] - ifNil: [lineStop := line last "scanning for point"]. runStopIndex := lastIndex + (runLength - 1) min: lineStop. lastCharacterExtent := 0 @ line lineHeight. spaceCount := 0. [ stopCondition := self scanCharactersFrom: lastIndex to: runStopIndex in: text string rightX: characterPoint x stopConditions: stopConditions kern: kern. "see setStopConditions for stopping conditions for character block operations." self lastCharacterExtentSetX: (specialWidth ifNil: [font widthOf: (text at: lastIndex)] ifNotNil: [specialWidth]). self perform: stopCondition ] whileFalse. characterIndex ifNil: [ "Result for characterBlockAtPoint: " (stopCondition ~~ #cr and: [ lastIndex == line last and: [ aPoint x > ((characterPoint x) + (lastCharacterExtent x / 2)) ]]) ifTrue: [ "Correct for right half of last character in line" ^ (CharacterBlock new stringIndex: lastIndex + 1 text: text topLeft: characterPoint + (lastCharacterExtent x @ 0) + (font descentKern @ 0) extent: 0 @ lastCharacterExtent y) textLine: line ]. ^ (CharacterBlock new stringIndex: lastIndex text: text topLeft: characterPoint + (font descentKern @ 0) extent: lastCharacterExtent - (font baseKern @ 0)) textLine: line] ifNotNil: ["Result for characterBlockForIndex: " ^ (CharacterBlock new stringIndex: characterIndex text: text topLeft: characterPoint + ((font descentKern) - kern @ 0) extent: lastCharacterExtent) textLine: line]! Item was changed: ----- Method: CharacterBlockScanner>>crossedX (in category 'stop conditions') ----- crossedX "Text display has wrapping. The scanner just found a character past the x location of the cursor. We know that the cursor is pointing at a character or before one." + | currentX | - | leadingTab currentX | characterIndex == nil ifFalse: [ "If the last character of the last line is a space, and it crosses the right margin, then locating the character block after it is impossible without this hack." characterIndex > text size ifTrue: [ lastIndex := characterIndex. characterPoint := (nextLeftMargin ifNil: [leftMargin]) @ (destY + line lineHeight). ^true]]. characterPoint x <= (destX + (lastCharacterExtent x // 2)) ifTrue: [lastCharacter := (text at: lastIndex). characterPoint := destX @ destY. ^true]. lastIndex >= line last ifTrue: [lastCharacter := (text at: line last). characterPoint := destX @ destY. ^true]. "Pointing past middle of a character, return the next character." lastIndex := lastIndex + 1. lastCharacter := text at: lastIndex. currentX := destX + lastCharacterExtent x + kern. self lastCharacterExtentSetX: (font widthOf: lastCharacter). characterPoint := currentX @ destY. lastCharacter = Space ifFalse: [^ true]. "Yukky if next character is space or tab." alignment = Justified ifTrue: [self lastCharacterExtentSetX: + (lastCharacterExtent x + (line justifiedPadFor: (spaceCount + 1) font: font))]. - (lastCharacterExtent x + (line justifiedPadFor: (spaceCount + 1) font: font)). - ^ true]. - true ifTrue: [^ true]. - "NOTE: I find no value to the following code, and so have defeated it - DI" - - "See tabForDisplay for illumination on the following awfulness." - leadingTab := true. - line first to: lastIndex - 1 do: - [:index | (text at: index) ~= Tab ifTrue: [leadingTab := false]]. - (alignment ~= Justified or: [leadingTab]) - ifTrue: [self lastCharacterExtentSetX: (textStyle nextTabXFrom: currentX - leftMargin: leftMargin rightMargin: rightMargin) - - currentX] - ifFalse: [self lastCharacterExtentSetX: (((currentX + (textStyle tabWidth - - (line justifiedTabDeltaFor: spaceCount))) - - currentX) max: 0)]. ^ true! Item was changed: CharacterScanner subclass: #DisplayScanner + instanceVariableNames: 'bitBlt lineY runX foregroundColor backgroundColor fillBlt lineHeight paragraphColor morphicOffset ignoreColorChanges' - instanceVariableNames: 'bitBlt lineY runX foregroundColor backgroundColor fillBlt lineHeight paragraph paragraphColor morphicOffset ignoreColorChanges' classVariableNames: '' poolDictionaries: '' category: 'Graphics-Text'! !DisplayScanner commentStamp: '<historical>' prior: 0! My instances are used to scan text and display it on the screen or in a hidden form.! Item was changed: ----- Method: DisplayScanner>>displayLines:in:clippedBy: (in category 'MVC-compatibility') ----- displayLines: linesInterval in: aParagraph clippedBy: visibleRectangle "The central display routine. The call on the primitive (scanCharactersFrom:to:in:rightX:) will be interrupted according to an array of stop conditions passed to the scanner at which time the code to handle the stop condition is run and the call on the primitive continued until a stop condition returns true (which means the line has terminated)." | leftInRun | "leftInRun is the # of characters left to scan in the current run; when 0, it is time to call 'self setStopConditions'" morphicOffset := 0@0. leftInRun := 0. self initializeFromParagraph: aParagraph clippedBy: visibleRectangle. ignoreColorChanges := false. - paragraph := aParagraph. foregroundColor := paragraphColor := aParagraph foregroundColor. backgroundColor := aParagraph backgroundColor. aParagraph backgroundColor isTransparent ifTrue: [fillBlt := nil] ifFalse: [fillBlt := bitBlt copy. "Blt to fill spaces, tabs, margins" fillBlt sourceForm: nil; sourceOrigin: 0@0. fillBlt fillColor: aParagraph backgroundColor]. rightMargin := aParagraph rightMarginForDisplay. lineY := aParagraph topAtLineIndex: linesInterval first. bitBlt destForm deferUpdatesIn: visibleRectangle while: [ linesInterval do: [:lineIndex | | string startIndex lastPos runLength stopCondition | line := aParagraph lines at: lineIndex. lastIndex := line first. self setStopConditions. " causes an assignment to inst var. alignment " leftMargin := aParagraph leftMarginForDisplayForLine: lineIndex alignment: (alignment ifNil:[textStyle alignment]). destX := (runX := leftMargin). line := aParagraph lines at: lineIndex. lineHeight := line lineHeight. fillBlt == nil ifFalse: [fillBlt destX: visibleRectangle left destY: lineY width: visibleRectangle width height: lineHeight; copyBits]. lastIndex := line first. leftInRun <= 0 ifTrue: [self setStopConditions. "also sets the font" leftInRun := text runLengthFor: line first]. destY := lineY + line baseline - font ascent. "Should have happened in setFont" runLength := leftInRun. runStopIndex := lastIndex + (runLength - 1) min: line last. leftInRun := leftInRun - (runStopIndex - lastIndex + 1). spaceCount := 0. string := text string. self handleIndentation. [ startIndex := lastIndex. lastPos := destX@destY. stopCondition := self scanCharactersFrom: lastIndex to: runStopIndex in: string rightX: rightMargin stopConditions: stopConditions kern: kern. lastIndex >= startIndex ifTrue:[ font displayString: string on: bitBlt from: startIndex to: lastIndex at: lastPos kern: kern]. "see setStopConditions for stopping conditions for displaying." self perform: stopCondition ] whileFalse. fillBlt == nil ifFalse: [fillBlt destX: destX destY: lineY width: visibleRectangle right-destX height: lineHeight; copyBits]. lineY := lineY + lineHeight]]!

1 0

The Trunk: Kernel-nice.602.mcz
by commits＠source.squeak.org 27 Jun '11

27 Jun '11

Nicolas Cellier uploaded a new version of Kernel to project The Trunk: http://source.squeak.org/trunk/Kernel-nice.602.mcz ==================== Summary ==================== Name: Kernel-nice.602 Author: nice Time: 27 June 2011, 8:27:00.479 pm UUID: 5466a1ae-e947-4881-b71f-e83278dc1e33 Ancestors: Kernel-bf.601 Minor tweaks: Use a better approximation for counting digits in base 10 Use raisedToInteger: since we now the argument 'base' is Integer Change arTanh error message because (between -1 and 1) is more inline with Smalltalk message #between:and: than (between 1 and -1). Simplify ulp since #abs matches #ulp return value for any non finite number. =============== Diff against Kernel-bf.601 =============== Item was changed: ----- Method: Float>>arTanh (in category 'mathematical functions') ----- arTanh "Answer receiver's area hyperbolic tangent. That is the inverse function of tanh." self = 0.0 ifTrue: [^self]. "Handle negativeZero" self abs = 1 ifTrue: [^self copySignTo: Float infinity]. self abs > 1 ifTrue: + [^DomainError signal: 'Receiver must be between -1.0 and 1.0']. - [^DomainError signal: 'Receiver must be between 1.0 and -1.0']. ^((1 + self) / (1 - self)) ln / 2! Item was changed: ----- Method: Float>>ulp (in category 'truncation and round off') ----- ulp "Answer the unit of least precision of self (the power of two corresponding to last bit of mantissa)" | exponent | + self isFinite ifFalse: [^self abs]. - self isFinite ifFalse: [ - self isNaN ifTrue: [^self]. - ^Float infinity]. self = 0.0 ifTrue: [^Float fmin]. exponent := self exponent. ^exponent < self class emin ifTrue: [Float fminDenormalized] ifFalse: [Float epsilon timesTwoPower: exponent]! Item was changed: ----- Method: Integer>>numberOfDigitsInBase: (in category 'printing') ----- numberOfDigitsInBase: b "Return how many digits are necessary to print this number in base b. This does not count any place for minus sign, radix prefix or whatever. Note that this algorithm may cost a few operations on LargeInteger." | nDigits q total | self negative ifTrue: [^self negated numberOfDigitsInBase: b]. self < b ifTrue: [^1]. b isPowerOfTwo ifTrue: [^self highBit + b highBit - 2 quo: b highBit - 1]. "A conversion from base 2 to base b has to be performed. This algorithm avoids Float computations like (self log: b) floor + 1, 1) because they are inexact 2) because LargeInteger might overflow 3) because this algorithm might be cheaper than conversion" q := self. total := 0. ["Make an initial nDigits guess that is lower than or equal to required number of digits" nDigits := b = 10 ifTrue: [((q highBit - 1) * 1233 >> 12) + 1. "This is because (2 log)/(10 log)*4096 is slightly greater than 1233"] ifFalse: [q highBit quo: b highBit]. total := total + nDigits. "See how many digits remains above these first nDigits guess" + (q := q quo: (b raisedToInteger: nDigits)) < b] whileFalse. - (q := q quo: (b raisedTo: nDigits)) < b] whileFalse. ^q = 0 ifTrue: [total] ifFalse: [total + 1]! Item was changed: ----- Method: LargePositiveInteger>>printOn:base: (in category 'printing') ----- printOn: aStream base: b "Append a representation of this number in base b on aStream. In order to reduce cost of LargePositiveInteger ops, split the number in approximately two equal parts in number of digits." | halfDigits halfPower head tail nDigitsUnderestimate | "Don't engage any arithmetic if not normalized" (self digitLength = 0 or: [(self digitAt: self digitLength) = 0]) ifTrue: [^self normalize printOn: aStream base: b]. nDigitsUnderestimate := b = 10 + ifTrue: [((self highBit - 1) * 1233 >> 12) + 1. "This is because (2 log)/(10 log)*4096 is slightly greater than 1233"] - ifTrue: [((self highBit - 1) * 3 quo: 10) + 1 "because 1024 is almost a kilo"] ifFalse: [self highBit quo: b highBit]. "splitting digits with a whole power of two is more efficient" halfDigits := 1 bitShift: nDigitsUnderestimate highBit - 2. halfDigits <= 1 ifTrue: ["Hmmm, this could happen only in case of a huge base b... Let lower level fail" ^self printOn: aStream base: b nDigits: (self numberOfDigitsInBase: b)]. "Separate in two halves, head and tail" halfPower := b raisedToInteger: halfDigits. head := self quo: halfPower. tail := self - (head * halfPower). "print head" head printOn: aStream base: b. "print tail without the overhead to count the digits" tail printOn: aStream base: b nDigits: halfDigits!

1 0

Re: [squeak-dev] The Inbox: Tools-fbs.364.mcz
by Frank Shearar 27 Jun '11

27 Jun '11

On 27 June 2011 12:56, <commits(a)source.squeak.org> wrote: > A new version of Tools was added to project The Inbox: > http://source.squeak.org/inbox/Tools-fbs.364.mcz > > ==================== Summary ==================== > > Name: Tools-fbs.364 > Author: fbs > Time: 27 June 2011, 12:56:26.402 pm > UUID: c881a91c-eaac-4a40-acfb-79fbe119558f > Ancestors: Tools-fbs.363 > > Mantis #7648: when deselecting a class after a class-side method's been viewed, the selectedMessageName must be nil: you can't be viewing a method at this point! I tried writing a test for this, and could never quite succeed. Maybe it's just because of the algebraic data type stuff I'm working with at the moment, but I run into this bug all the time. frank

1 0

The Inbox: Tools-fbs.364.mcz
by commits＠source.squeak.org 27 Jun '11

27 Jun '11

A new version of Tools was added to project The Inbox: http://source.squeak.org/inbox/Tools-fbs.364.mcz ==================== Summary ==================== Name: Tools-fbs.364 Author: fbs Time: 27 June 2011, 12:56:26.402 pm UUID: c881a91c-eaac-4a40-acfb-79fbe119558f Ancestors: Tools-fbs.363 Mantis #7648: when deselecting a class after a class-side method's been viewed, the selectedMessageName must be nil: you can't be viewing a method at this point! =============== Diff against Tools-fbs.363 =============== Item was changed: ----- Method: Browser>>selectClassNamed: (in category 'class list') ----- selectClassNamed: aSymbolOrString | className currentMessageCategoryName currentMessageName | + currentMessageCategoryName := [self selectedMessageCategoryName] on: Error do: [:ex| ex return: nil]. currentMessageName := [self selectedMessageName] on: Error do: [:ex| ex return: nil]. selectedClassName := aSymbolOrString ifNotNil: [ aSymbolOrString asSymbol ]. self setClassOrganizer. "Try to reselect the category and/or selector if the new class has them." selectedMessageCategoryName :=(self messageCategoryList includes: currentMessageCategoryName) ifTrue: [currentMessageCategoryName] ifFalse: [nil]. + selectedMessageName := (self messageList includes: currentMessageName) + ifTrue: [currentMessageName] + ifFalse: [nil]. - selectedMessageName := currentMessageName. self hasMessageSelected ifTrue: [self editSelection: #editMessage] ifFalse: [self hasMessageCategorySelected ifTrue: [self editSelection: #newMessage] ifFalse: [self classCommentIndicated ifTrue: [self editSelection: #editComment] ifFalse: [self editSelection: (self hasClassSelected not ifTrue: [(metaClassIndicated or: [ self hasSystemCategorySelected not ]) ifTrue: [#none] ifFalse: [#newClass]] ifFalse: [#editClass])]]]. contents := nil. self selectedClass isNil ifFalse: [className := self selectedClass name. (RecentClasses includes: className) ifTrue: [RecentClasses remove: className]. RecentClasses addFirst: className. RecentClasses size > 16 ifTrue: [RecentClasses removeLast]]. self changed: #classSelectionChanged. self changed: #classCommentText. self changed: #classListIndex. "update my selection" self changed: #messageCategoryList. self changed: #messageList. self changed: #relabel. self contentsChanged!

1 0

The Inbox: ToolsTests-fbs.48.mcz
by commits＠source.squeak.org 27 Jun '11

27 Jun '11

A new version of ToolsTests was added to project The Inbox: http://source.squeak.org/inbox/ToolsTests-fbs.48.mcz ==================== Summary ==================== Name: ToolsTests-fbs.48 Author: fbs Time: 27 June 2011, 12:51:07.632 pm UUID: 47511e5c-e42a-4f13-a270-fe4f6f64a38c Ancestors: ToolsTests-eem.47 Broken test: the commonSelector was not common within the commonCategory! =============== Diff against ToolsTests-eem.47 =============== Item was changed: ----- Method: BrowserTest>>testSelectClassNamedPreservesPlace (in category 'as yet unclassified') ----- testSelectClassNamedPreservesPlace | commonCategory commonSelector otherClass | browser selectSystemCategory: browser class category. otherClass := HierarchyBrowser. "Unlikely to move out of Browser's package!!" browser selectClassNamed: browser class name. commonCategory := (browser class organization categories intersection: otherClass organization categories) first. + commonSelector := #classList. - commonSelector := (otherClass organization allMethodSelectors intersection: browser class organization allMethodSelectors) first. browser selectMessageCategoryNamed: commonCategory. browser selectMessageNamed: commonSelector. browser selectClass: otherClass. self assert: browser selectedClassName = otherClass name. self assert: browser selectedMessageCategoryName = commonCategory. self assert: browser selectedMessageName = commonSelector.!

1 0

User interrupt does not always break UI process
by Bert Freudenberg 27 Jun '11

27 Jun '11

I often get the Finalization process when I press Cmd-. Usually I want the UI process. But it should be possible to break into a higher-level one too, if it takes much time. Not sure if this could be made more intelligent? - Bert -

4 5

← Newer
1
2
3
4
5
6
7
...
24
Older →

Jump to page: