Squeak-dev May 2011

squeak-dev@lists.squeakfoundation.org

84 participants
303 discussions

The Trunk: ToolsTests-eem.47.mcz
by commits＠source.squeak.org 24 May '11

24 May '11

Eliot Miranda uploaded a new version of ToolsTests to project The Trunk: http://source.squeak.org/trunk/ToolsTests-eem.47.mcz ==================== Summary ==================== Name: ToolsTests-eem.47 Author: eem Time: 23 May 2011, 6:12:10.584 pm UUID: b9c22a05-2c13-4351-8540-d49384b0916e Ancestors: ToolsTests-ul.46 Add test for allCallsOn: et al discovering references to Float bindingOf: #NaN. =============== Diff against ToolsTests-ul.46 =============== Item was added: + ----- Method: BrowseTest>>testFindReferencesToFloatNaN (in category 'testing') ----- + testFindReferencesToFloatNaN + self assert: (SystemNavigation new allCallsOn: (Float bindingOf: #NaN)) notEmpty!

1 0

The Trunk: Kernel-eem.590.mcz
by commits＠source.squeak.org 24 May '11

24 May '11

Eliot Miranda uploaded a new version of Kernel to project The Trunk: http://source.squeak.org/trunk/Kernel-eem.590.mcz ==================== Summary ==================== Name: Kernel-eem.590 Author: eem Time: 23 May 2011, 6:10:15.256 pm UUID: bbcacb44-5bd6-40fe-965a-697b4b126ffd Ancestors: Kernel-nice.589 Fix whichSelectorsReferTo:special:byte:thorough: for Float bindingOf: #NaN, so that one can browse references to the NaN class var. Use literaslsDo: instead of literals do: in spaceUsed. =============== Diff against Kernel-nice.589 =============== Item was changed: ----- Method: Behavior>>spaceUsed (in category 'private') ----- spaceUsed "Answer a rough estimate of number of bytes used by this class and its metaclass. Does not include space used by class variables." | space | space := 0. self methodsDo: [:method | space := space + 16. "dict and org'n space" space := space + (method size + 6 "hdr + avg pad"). + method literalsDo: [:lit | - method literals do: [:lit | (lit isMemberOf: Array) ifTrue: [space := space + ((lit size + 1) * 4)]. (lit isMemberOf: Float) ifTrue: [space := space + 12]. (lit isMemberOf: ByteString) ifTrue: [space := space + (lit size + 6)]. (lit isMemberOf: LargeNegativeInteger) ifTrue: [space := space + ((lit size + 1) * 4)]. (lit isMemberOf: LargePositiveInteger) ifTrue: [space := space + ((lit size + 1) * 4)]]]. ^ space! Item was changed: ----- Method: Behavior>>whichSelectorsReferTo:special:byte:thorough: (in category 'testing method dictionary') ----- whichSelectorsReferTo: literal special: specialFlag byte: specialByte thorough: thorough "Answer a set of selectors whose methods access the argument as a literal. If thorough is true, then dives into the compact literal notation, making it slow but thorough " | who | who := IdentitySet new. self selectorsAndMethodsDo: [ :selector :method | (((thorough ifFalse: [ method hasLiteral: literal ] ifTrue: [ method hasLiteralThorough: literal ]) or: [ specialFlag and: [ method scanFor: specialByte ] ]) and: [ literal isVariableBinding not or: [ + "N.B. (method indexOfLiteral: literal) < method numLiterals copes with l;ooking for + Float bindingOf: #NaN, since (Float bindingOf: #NaN) ~= (Float bindingOf: #NaN)." + (method indexOfLiteral: literal) ~= 0] ]) ifTrue: [ - method literals allButLast includes: literal] ]) ifTrue: [ who add: selector ] ]. ^who!

1 0

The Trunk: KernelTests-eem.192.mcz
by commits＠source.squeak.org 24 May '11

24 May '11

Eliot Miranda uploaded a new version of KernelTests to project The Trunk: http://source.squeak.org/trunk/KernelTests-eem.192.mcz ==================== Summary ==================== Name: KernelTests-eem.192 Author: eem Time: 23 May 2011, 6:03:03.434 pm UUID: d45cfe25-d091-4b7f-a9c7-09cb01098771 Ancestors: KernelTests-dtl.191 Use cm literalAt: cm numLiterals instead of cm literals last. Arguambly should use cm methodClassAssociation =============== Diff against KernelTests-dtl.191 =============== Item was changed: ----- Method: MethodPropertiesTest>>testAllMethodsHaveMethodClass (in category 'testing') ----- testAllMethodsHaveMethodClass Smalltalk garbageCollect. self assert: (CompiledMethod allInstances + reject: [:cm | | lastLiteral | + lastLiteral := cm literalAt: cm numLiterals. + lastLiteral isVariableBinding + and: [lastLiteral value isBehavior + or: [lastLiteral value isTrait]]]) isEmpty - reject: [:cm | cm literals last isVariableBinding - and: [cm literals last value isBehavior - or: [cm literals last value isTrait]]]) isEmpty description: 'CompiledMethods must have methodClass literal'!

1 0

The Inbox: Graphics-wiz.183.mcz
by commits＠source.squeak.org 24 May '11

24 May '11

A new version of Graphics was added to project The Inbox: http://source.squeak.org/inbox/Graphics-wiz.183.mcz ==================== Summary ==================== Name: Graphics-wiz.183 Author: wiz Time: 23 May 2011, 8:55:06.627 pm UUID: de71490d-a20e-4413-b8b0-8bc9ae3bfdb9 Ancestors: Graphics-nice.182 The last of three pieces fixing M7635. adjustOrigins and adjustExtent assure parameters are defined and not in class Fraction form. clipRange has had lazy initialization factored out by useing the adjust methods. copyBits now tries the two adjust methods before trying cliprange. It also now tries clipRange without the horrible roundVariables method. As a last resort it sends and error method before proceeding to try the roundVariables method. That should do it. Oh yeah be sure to load the other two parts first. Everything will depend on the presense of asNonFraction. =============== Diff against Graphics-nice.182 =============== Item was added: + ----- Method: BitBlt>>adjustExtents (in category 'accessing') ----- + adjustExtents + "Answer if any of the extent numbers were class fraction or undefined. + Assure all extent numbers are defined and non-fractional." + + | answer | + (answer := width isFraction | height isFraction | clipWidth isFraction | clipHeight isFraction + | width isNil | height isNil | clipWidth isNil | clipHeight isNil) ifFalse: [ ^answer ] . + width := ( width ifNil: [ destForm width ] ) asNonFraction . + height := ( height ifNil: [ destForm height ] ) asNonFraction . + clipWidth := (clipWidth ifNil:[clipWidth := destForm width]) asNonFraction . + clipHeight := (clipHeight ifNil:[clipHeight := destForm height]) asNonFraction . + ^answer ! Item was added: + ----- Method: BitBlt>>adjustOrigins (in category 'accessing') ----- + adjustOrigins + "Answer if any of the origin numbers were class Fraction or undefined. Assure all origin numbers are non-fractional and defined." + + | answer | + (answer := destX isFraction | destY isFraction + | sourceX isFraction | sourceY isFraction + | clipX isFraction | clipY isFraction + | destX isNil | destY isNil + | sourceX isNil | sourceY isNil + | clipX isNil | clipY isNil ) + ifFalse: [ ^answer ] . + + destX :=( destX ifNil:[ 0]) asNonFraction . + destY := (destY ifNil:[ 0]) asNonFraction . + sourceX := (sourceX ifNil:[ 0]) asNonFraction . + sourceY := (sourceY ifNil:[ 0]) asNonFraction . + clipX := (clipX ifNil:[0]) asNonFraction . + clipY := (clipY ifNil:[0]) asNonFraction . + ^ answer! Item was changed: ----- Method: BitBlt>>clipRange (in category 'private') ----- clipRange + "Fill in the lazy state if needed and insure no numbers are class Fraction. + Clip and adjust source origin and extent appropriately" + - "clip and adjust source origin and extent appropriately" - "first in x" | sx sy dx dy bbW bbH | + self adjustExtents | self adjustOrigins . "True when something changed." - "fill in the lazy state if needed" - destX ifNil:[destX := 0]. - destY ifNil:[destY := 0]. - width ifNil:[width := destForm width]. - height ifNil:[height := destForm height]. - sourceX ifNil:[sourceX := 0]. - sourceY ifNil:[sourceY := 0]. - clipX ifNil:[clipX := 0]. - clipY ifNil:[clipY := 0]. - clipWidth ifNil:[clipWidth := destForm width]. - clipHeight ifNil:[clipHeight := destForm height]. + "first in x" destX >= clipX ifTrue: [sx := sourceX. dx := destX. bbW := width] ifFalse: [sx := sourceX + (clipX - destX). bbW := width - (clipX - destX). dx := clipX]. (dx + bbW) > (clipX + clipWidth) ifTrue: [bbW := bbW - ((dx + bbW) - (clipX + clipWidth))]. "then in y" destY >= clipY ifTrue: [sy := sourceY. dy := destY. bbH := height] ifFalse: [sy := sourceY + clipY - destY. bbH := height - (clipY - destY). dy := clipY]. (dy + bbH) > (clipY + clipHeight) ifTrue: [bbH := bbH - ((dy + bbH) - (clipY + clipHeight))]. sourceForm ifNotNil:[ sx < 0 ifTrue: [dx := dx - sx. bbW := bbW + sx. sx := 0]. sx + bbW > sourceForm width ifTrue: [bbW := bbW - (sx + bbW - sourceForm width)]. sy < 0 ifTrue: [dy := dy - sy. bbH := bbH + sy. sy := 0]. sy + bbH > sourceForm height ifTrue: [bbH := bbH - (sy + bbH - sourceForm height)]. ]. (bbW <= 0 or:[bbH <= 0]) ifTrue:[ sourceX := sourceY := destX := destY := clipX := clipY := width := height := 0. ^true]. (sx = sourceX and:[sy = sourceY and:[dx = destX and:[dy = destY and:[bbW = width and:[bbH = height]]]]]) ifTrue:[^false]. sourceX := sx. sourceY := sy. destX := dx. destY := dy. width := bbW. height := bbH. ^true! Item was changed: ----- Method: BitBlt>>copyBits (in category 'copying') ----- copyBits "Primitive. Perform the movement of bits from the source form to the destination form. Fail if any variables are not of the right type (Integer, Float, or Form) or if the combination rule is not implemented. In addition to the original 16 combination rules, this BitBlt supports 16 fail (to simulate paint) 17 fail (to simulate mask) 18 sourceWord + destinationWord 19 sourceWord - destinationWord 20 rgbAdd: sourceWord with: destinationWord 21 rgbSub: sourceWord with: destinationWord 22 rgbDiff: sourceWord with: destinationWord 23 tallyIntoMap: destinationWord 24 alphaBlend: sourceWord with: destinationWord 25 pixPaint: sourceWord with: destinationWord 26 pixMask: sourceWord with: destinationWord 27 rgbMax: sourceWord with: destinationWord 28 rgbMin: sourceWord with: destinationWord 29 rgbMin: sourceWord bitInvert32 with: destinationWord " <primitive: 'primitiveCopyBits' module: 'BitBltPlugin'> "Check for compressed source, destination or halftone forms" (combinationRule >= 30 and: [combinationRule <= 31]) ifTrue: ["No alpha specified -- re-run with alpha = 1.0" ^ self copyBitsTranslucent: 255]. ((sourceForm isForm) and: [sourceForm unhibernate]) ifTrue: [^ self copyBits]. ((destForm isForm) and: [destForm unhibernate]) ifTrue: [^ self copyBits]. ((halftoneForm isForm) and: [halftoneForm unhibernate]) ifTrue: [^ self copyBits]. "Check for unimplmented rules" combinationRule = Form oldPaint ifTrue: [^ self paintBits]. combinationRule = Form oldErase1bitShape ifTrue: [^ self eraseBits]. "Check if BitBlt doesn't support full color maps" (colorMap notNil and:[colorMap isColormap]) ifTrue:[ colorMap := colorMap colors. ^self copyBits]. + "Check for class Fraction numbers and undefined paramenters." + self adjustExtents ifTrue: [ ^self copyBits ]. + self adjustOrigins ifTrue: [ ^self copyBits ]. + "Check if clipping gots us way out of range" + self clipRange ifTrue:[ ^self copyBits]. - self clipRange ifTrue:[self roundVariables. ^self copyBitsAgain]. + self error: 'Bad BitBlt arg; proceed to round args..'. - self error: 'Bad BitBlt arg (Fraction?); proceed to convert.'. "Convert all numeric parameters to integers and try again." self roundVariables. ^ self copyBitsAgain!

1 0

The Inbox: Tests-wiz.129.mcz
by commits＠source.squeak.org 24 May '11

24 May '11

A new version of Tests was added to project The Inbox: http://source.squeak.org/inbox/Tests-wiz.129.mcz ==================== Summary ==================== Name: Tests-wiz.129 Author: wiz Time: 23 May 2011, 8:41:13.662 pm UUID: 6adddaae-d5f8-417f-ada0-da25040053e0 Ancestors: Tests-nice.128 Tests for class fraction parameters being passed to copy bit. Two main tests should fail and pass after test is implemented. The third test was for debugging one of the main tests. It should run green before and after. =============== Diff against Tests-nice.128 =============== Item was added: + ----- Method: BitBltClipBugs>>testFloatingFractionOrigin (in category 'as yet unclassified') ----- + testFloatingFractionOrigin + "Copy bits should be able to handle fractional extents w/o error." + "self debug: #testFloatingFractionOrigin" + | f1 bb f2 | + f1 := Form extent: 100@100 depth: 1. + f2 := Form extent: 100@100 depth: 1. + bb := BitBlt toForm: f1. + bb combinationRule: 3. + bb sourceForm: f2. + bb destOrigin: 0 @ (0.5) . + bb width: 100; height: 200. + self shouldnt:[bb copyBits] raise: Error. + ! Item was added: + ----- Method: BitBltClipBugs>>testFractionalExtent (in category 'as yet unclassified') ----- + testFractionalExtent + "Copy bits should be able to handle fractional extents w/o error." + | f1 bb f2 | + f1 := Form extent: 100@100 depth: 1. + f2 := Form extent: 100@100 depth: 1. + bb := BitBlt toForm: f1. + bb combinationRule: 3. + bb sourceForm: f2. + bb destOrigin: 0 asPoint. + bb width: 100; height: 201/2. + self shouldnt:[bb copyBits] raise: Error. + ! Item was added: + ----- Method: BitBltClipBugs>>testFractionalOrigin (in category 'as yet unclassified') ----- + testFractionalOrigin + "Copy bits should be able to handle fractional extents w/o error." + | f1 bb f2 | + f1 := Form extent: 100@100 depth: 1. + f2 := Form extent: 100@100 depth: 1. + bb := BitBlt toForm: f1. + bb combinationRule: 3. + bb sourceForm: f2. + bb destOrigin: 0 @ (1/2) .. + bb width: 100; height: 200. + self shouldnt:[bb copyBits] raise: Error. + !

1 0

The Inbox: Kernel-wiz.590.mcz
by commits＠source.squeak.org 24 May '11

24 May '11

A new version of Kernel was added to project The Inbox: http://source.squeak.org/inbox/Kernel-wiz.590.mcz ==================== Summary ==================== Name: Kernel-wiz.590 Author: wiz Time: 23 May 2011, 8:36:40.385 pm UUID: 4c655b5c-5794-4e0b-8160-f696e5ce7fd6 Ancestors: Kernel-nice.589 Add asNonFraction to Fraction Number and Object. Added to object so nil will understand. Methods are essential for fixing copyBit bug see M7635. =============== Diff against Kernel-nice.589 =============== Item was added: + ----- Method: Fraction>>asNonFraction (in category 'converting') ----- + asNonFraction + "Answer the receiver asFloat. This is here for the benifit of the copyBits primitive which can handle floats but not fractions. It has to be stated this way so we don't turn integers into floats. See inheritance for backstop" + + ^self asFloat! Item was added: + ----- Method: Number>>asNonFraction (in category 'converting') ----- + asNonFraction + "Answer the receiver. This is here for the benifit of the copyBits primitive which can handle floats but not fractions. It has to be stated this way so we don't turn integers into floats. See Fraction for active method" + + ^self ! Item was added: + ----- Method: Object>>asNonFraction (in category 'converting') ----- + asNonFraction + "Answer the receiver. This is here for the benifit of the copyBits primitive which can handle floats but not fractions. It has to be stated this way so we don't turn integers into floats. See Fraction for active method. Lazy initialization requires this work of nil also." + + ^self !

1 0

Daily Commit Log
by commits＠source.squeak.org 24 May '11

24 May '11

Changes to Trunk (http://source.squeak.org/trunk.html) in the last 24 hours: http://lists.squeakfoundation.org/pipermail/packages/2011-May/004729.html Name: Morphic-bf.543 Ancestors: Morphic-dtl.542 bullet-proof Morphic drawing error handling =============================================

1 0

[BUG][M7635][Fixed w /Tests] Fraction class co-ordinates
by Jerome Peace 23 May '11

23 May '11

I have fixed the problem in copyBits so that it is safe to use Fraction class parameters in grafports and else where. I have left the error notification in place but added repairs above that. This should fix most problems. A changeset is included with the report. The key is a #asNonFraction method that changes Fractions (and nothing else) into Floats. The primitive seems able to handle it from there. If this is valuable to squeak or Pharo or Cuis then someone else must harvest it from the change set. I prefer to spend my time working on more puzzling matters. A tip of the hat and a great thanks to Bert for working on the other half of the problem. If he made the error notification work properly there are going to be a lot of happy people out there including this one. Cheers, Yours in curiosity and service, --Jerome Peace

1 0

The Inbox: Kernel-cmm.586.mcz
by commits＠source.squeak.org 23 May '11

23 May '11

A new version of Kernel was added to project The Inbox: http://source.squeak.org/inbox/Kernel-cmm.586.mcz ==================== Summary ==================== Name: Kernel-cmm.586 Author: cmm Time: 23 May 2011, 2:54:46.188 pm UUID: 08000000-1508-901b-1508-901b14000000 Ancestors: Kernel-fbs.585 Move Object>>'tracing' messages to ProtoObject to allow tracing of proxy objects. =============== Diff against Kernel-fbs.585 =============== Item was removed: - ----- Method: Object>>chasePointers (in category 'tracing') ----- - chasePointers - PointerFinder on: self! Item was removed: - ----- Method: Object>>explorePointers (in category 'tracing') ----- - explorePointers - PointerExplorer new openExplorerFor: self! Item was removed: - ----- Method: Object>>inboundPointers (in category 'tracing') ----- - inboundPointers - "Answers a collection of all objects in the system that point to myself" - - ^ self inboundPointersExcluding: #()! Item was removed: - ----- Method: Object>>inboundPointersExcluding: (in category 'tracing') ----- - inboundPointersExcluding: objectsToExclude - "Answer a list of all objects in the system that point to me, excluding those in the collection of objectsToExclude. I do my best to avoid creating any temporary objects that point to myself, especially method and block contexts. Adapted from PointerFinder class >> #pointersTo:except:" - - | anObj pointers objectsToAlwaysExclude | - Smalltalk garbageCollect. - "big collection shouldn't grow, so it's contents array is always the same" - pointers := OrderedCollection new: 1000. - - "#allObjectsDo: and #pointsTo: are expanded inline to keep spurious - method and block contexts out of the results" - anObj := self someObject. - [0 == anObj] whileFalse: [ "We must use #== here, to avoid leaving the loop when anObj is another number that's equal to 0 (e.g. 0.0)." - anObj isInMemory - ifTrue: [((anObj instVarsInclude: self) - or: [anObj class == self]) - ifTrue: [pointers add: anObj]]. - anObj := anObj nextObject]. - - objectsToAlwaysExclude := { - pointers collector. - thisContext. - thisContext sender. - thisContext sender sender. - objectsToExclude. - }. - - ^ pointers removeAllSuchThat: [:ea | - (objectsToAlwaysExclude identityIncludes: ea) - or: [objectsToExclude identityIncludes: ea]]! Item was removed: - ----- Method: Object>>outboundPointers (in category 'tracing') ----- - outboundPointers - "Answers a list of all objects I am causing not to be garbage-collected" - - | collection | - collection := OrderedCollection new. - self outboundPointersDo: [:ea | collection add: ea]. - ^ collection! Item was removed: - ----- Method: Object>>outboundPointersDo: (in category 'tracing') ----- - outboundPointersDo: aBlock - "do aBlock for every object I point to, exactly how the garbage collector would. Adapted from PointerFinder >> #followObject:" - - aBlock value: self class. - 1 to: self class instSize do: [:i | aBlock value: (self instVarAt: i)]. - 1 to: self basicSize do: [:i | aBlock value: (self basicAt: i)].! Item was added: + ----- Method: ProtoObject>>chasePointers (in category 'tracing') ----- + chasePointers + PointerFinder on: self! Item was added: + ----- Method: ProtoObject>>explorePointers (in category 'tracing') ----- + explorePointers + PointerExplorer new openExplorerFor: self! Item was added: + ----- Method: ProtoObject>>inboundPointers (in category 'tracing') ----- + inboundPointers + "Answers a collection of all objects in the system that point to myself" + + ^ self inboundPointersExcluding: #()! Item was added: + ----- Method: ProtoObject>>inboundPointersExcluding: (in category 'tracing') ----- + inboundPointersExcluding: objectsToExclude + "Answer a list of all objects in the system that point to me, excluding those in the collection of objectsToExclude. I do my best to avoid creating any temporary objects that point to myself, especially method and block contexts. Adapted from PointerFinder class >> #pointersTo:except:" + + | anObj pointers objectsToAlwaysExclude | + Smalltalk garbageCollect. + "big collection shouldn't grow, so it's contents array is always the same" + pointers := OrderedCollection new: 1000. + + "#allObjectsDo: and #pointsTo: are expanded inline to keep spurious + method and block contexts out of the results" + anObj := self someObject. + [0 == anObj] whileFalse: [ "We must use #== here, to avoid leaving the loop when anObj is another number that's equal to 0 (e.g. 0.0)." + anObj isInMemory + ifTrue: [((anObj instVarsInclude: self) + or: [anObj class == self]) + ifTrue: [pointers add: anObj]]. + anObj := anObj nextObject]. + + objectsToAlwaysExclude := { + pointers collector. + thisContext. + thisContext sender. + thisContext sender sender. + objectsToExclude. + }. + + ^ pointers removeAllSuchThat: [:ea | + (objectsToAlwaysExclude identityIncludes: ea) + or: [objectsToExclude identityIncludes: ea]]! Item was added: + ----- Method: ProtoObject>>outboundPointers (in category 'tracing') ----- + outboundPointers + "Answers a list of all objects I am causing not to be garbage-collected" + + | collection | + collection := OrderedCollection new. + self outboundPointersDo: [:ea | collection add: ea]. + ^ collection! Item was added: + ----- Method: ProtoObject>>outboundPointersDo: (in category 'tracing') ----- + outboundPointersDo: aBlock + "do aBlock for every object I point to, exactly how the garbage collector would. Adapted from PointerFinder >> #followObject:" + + aBlock value: self class. + 1 to: self class instSize do: [:i | aBlock value: (self instVarAt: i)]. + 1 to: self basicSize do: [:i | aBlock value: (self basicAt: i)].!

1 0

The Inbox: CollectionsTests-topa.185.mcz
by commits＠source.squeak.org 23 May '11

23 May '11

A new version of CollectionsTests was added to project The Inbox: http://source.squeak.org/inbox/CollectionsTests-topa.185.mcz ==================== Summary ==================== Name: CollectionsTests-topa.185 Author: topa Time: 23 May 2011, 1:50:29.396 pm UUID: 8e536e32-1fb9-4bdd-8d4b-3088b46e41a9 Ancestors: CollectionsTests-nice.184 Mall testcase for RWBinaryOrTextStream on exisiting collections. =============== Diff against CollectionsTests-nice.184 =============== Item was added: + ----- Method: RWBinaryOrTextStreamTest>>testExisiting (in category 'tests') ----- + testExisiting + + | string stream | + + string := 'hello'. + + stream := RWBinaryOrTextStream on: string. + self assert: stream contents = string. + + stream := RWBinaryOrTextStream on: string. + stream setToEnd. + self assert: stream position = string size. + !

1 0

← Newer
1
2
3
4
5
6
7
8
9
...
31
Older →

Jump to page: