Squeak-dev October 2023

squeak-dev@lists.squeakfoundation.org

24 participants
148 discussions

The Trunk: CollectionsTests-ul.401.mcz
by commits＠source.squeak.org 07 Oct '23

07 Oct '23

Levente Uzonyi uploaded a new version of CollectionsTests to project The Trunk: http://source.squeak.org/trunk/CollectionsTests-ul.401.mcz ==================== Summary ==================== Name: CollectionsTests-ul.401 Author: ul Time: 7 October 2023, 10:20:53.836488 am UUID: c18f6621-cbdb-497a-a0d2-2e0e351f7cc4 Ancestors: CollectionsTests-mt.400 Use #allButFirst instead of #copyWithoutFirst. =============== Diff against CollectionsTests-mt.400 =============== Item was changed: ----- Method: LinkedListTest>>firstIndexesOf:in: (in category 'tests - copying with replacement') ----- firstIndexesOf: aSubCollection in: collection " return an OrderedCollection with the first indexes of the occurrences of subCollection in collection " | tmp result currentIndex | tmp:= collection. result:= OrderedCollection new. currentIndex := 1. [tmp isEmpty ]whileFalse: [ (tmp beginsWith: aSubCollection) ifTrue: [ result add: currentIndex. 1 to: aSubCollection size do: [:i | + tmp := tmp allButFirst. - tmp := tmp copyWithoutFirst. currentIndex := currentIndex + 1] ] ifFalse: [ + tmp := tmp allButFirst. - tmp := tmp copyWithoutFirst. currentIndex := currentIndex +1. ] ]. ^ result. ! Item was changed: ----- Method: LinkedListTest>>howMany:in: (in category 'tests - fixture') ----- howMany: aSubCollection in: collection " return an integer representing how many time 'subCollection' appears in 'collection' " | tmp nTime | tmp := collection. nTime:= 0. [tmp isEmpty ]whileFalse: [ (tmp beginsWith: aSubCollection) ifTrue: [ nTime := nTime + 1. + 1 to: aSubCollection size do: [:i | tmp := tmp allButFirst.] - 1 to: aSubCollection size do: [:i | tmp := tmp copyWithoutFirst.] ] + ifFalse: [tmp := tmp allButFirst.] - ifFalse: [tmp := tmp copyWithoutFirst.] ]. ^ nTime. ! Item was changed: ----- Method: LinkedListTest>>testAllButFirst (in category 'tests - subcollections access') ----- testAllButFirst "self debug: #testAllButFirst" + | abf col | col := self moreThan3Elements. abf := col allButFirst. self deny: abf first = col first. + self assert: abf size + 1 = col size. + 1 to: abf size do: [ :i | + self assert: (abf at: i) = (col at: i + 1) ]! - self assert: abf size + 1 = col size! Item was removed: - ----- Method: LinkedListTest>>testCopyWithoutFirst (in category 'tests - copying with or without') ----- - testCopyWithoutFirst - - | result | - result := self nonEmpty copyWithoutFirst. - - self assert: result size = (self nonEmpty size - 1). - - 1 to: result size do: - [:i | - self assert: (result at: i)= (self nonEmpty at: (i + 1))].! Item was changed: ----- Method: LinkedListTest>>testsEndsWith (in category 'tests - begins ends with') ----- testsEndsWith + self assert: (self nonEmpty endsWith: self nonEmpty allButFirst). - self assert: (self nonEmpty endsWith: self nonEmpty copyWithoutFirst). self assert: (self nonEmpty endsWith: self nonEmpty). self deny: (self nonEmpty endsWith: (self nonEmpty copyWith: self nonEmpty first)).!

1 0

The Trunk: Files-ul.206.mcz
by commits＠source.squeak.org 07 Oct '23

07 Oct '23

Levente Uzonyi uploaded a new version of Files to project The Trunk: http://source.squeak.org/trunk/Files-ul.206.mcz ==================== Summary ==================== Name: Files-ul.206 Author: ul Time: 7 October 2023, 10:14:37.49644 am UUID: 3ca897b4-8216-419c-b2be-6ff628652a9e Ancestors: Files-mt.205 Use #allButFirst instead of #copyWithoutFirst. =============== Diff against Files-mt.205 =============== Item was changed: ----- Method: FileDirectory class>>makeRelative: (in category 'platform specific') ----- makeRelative: path "Ensure that path looks like an relative path" ^path first = self pathNameDelimiter + ifTrue: [ path allButFirst ] - ifTrue: [ path copyWithoutFirst ] ifFalse: [ path ]! Item was changed: ----- Method: MacFileDirectory class>>makeAbsolute: (in category 'platform specific') ----- makeAbsolute: path "Ensure that path looks like an absolute path" | absolutePath | (self isAbsolute: path) ifTrue: [ ^path ]. "If a path begins with a colon, it is relative." absolutePath := (path first = $:) + ifTrue: [ path allButFirst ] - ifTrue: [ path copyWithoutFirst ] ifFalse: [ path ]. (self isAbsolute: absolutePath) ifTrue: [ ^absolutePath ]. "Otherwise, if it contains a colon anywhere, it is absolute and the first component is the volume name." ^absolutePath, ':'!

1 0

The Trunk: Installer-Core-ul.454.mcz
by commits＠source.squeak.org 07 Oct '23

07 Oct '23

Levente Uzonyi uploaded a new version of Installer-Core to project The Trunk: http://source.squeak.org/trunk/Installer-Core-ul.454.mcz ==================== Summary ==================== Name: Installer-Core-ul.454 Author: ul Time: 7 October 2023, 10:14:05.797397 am UUID: a09d594d-266c-4575-b81f-ce88d0eb2779 Ancestors: Installer-Core-ct.453 Use #allButFirst instead of #copyWithoutFirst. =============== Diff against Installer-Core-ct.453 =============== Item was changed: ----- Method: InstallerMantis>>maFiles (in category 'mantis') ----- maFiles | file files bugPage id | files := Dictionary new. bugPage := self maPage. [ id := bugPage upToAll: 'href="file_download.php?file_id='; upTo: $&. file := bugPage upToAll: 'amp;type=bug"' ; upTo: $<. + ((file size > 1) and: [file first = $>]) ifTrue: [ files at: file allButFirst put: id ]. - ((file size > 1) and: [file first = $>]) ifTrue: [ files at: file copyWithoutFirst put: id ]. id notEmpty ] whileTrue. ^files !

1 0

The Trunk: PreferenceBrowser-ul.158.mcz
by commits＠source.squeak.org 07 Oct '23

07 Oct '23

Levente Uzonyi uploaded a new version of PreferenceBrowser to project The Trunk: http://source.squeak.org/trunk/PreferenceBrowser-ul.158.mcz ==================== Summary ==================== Name: PreferenceBrowser-ul.158 Author: ul Time: 7 October 2023, 10:13:39.956188 am UUID: 0b7046af-cbb0-43db-9fd9-f20f58898033 Ancestors: PreferenceBrowser-mt.157 Use #allButFirst instead of #copyWithoutFirst. =============== Diff against PreferenceBrowser-mt.157 =============== Item was changed: ----- Method: PBPreferenceButtonMorph>>removeExtraControls (in category 'extra controls') ----- removeExtraControls + self submorphs allButFirst do: [:ea | ea delete]! - self submorphs copyWithoutFirst do: [:ea | ea delete]!

1 0

The Trunk: Traits-ul.314.mcz
by commits＠source.squeak.org 07 Oct '23

07 Oct '23

Levente Uzonyi uploaded a new version of Traits to project The Trunk: http://source.squeak.org/trunk/Traits-ul.314.mcz ==================== Summary ==================== Name: Traits-ul.314 Author: ul Time: 7 October 2023, 10:10:32.815801 am UUID: 96584240-d311-4b11-b124-d11cb42936e7 Ancestors: Traits-eem.313 In SequenceableCollection>>asTraitComposition, use #reduce: instead of a combination of #copyWithoutFirst, #first and #inject:into:. =============== Diff against Traits-eem.313 =============== Item was changed: ----- Method: SequenceableCollection>>asTraitComposition (in category '*Traits') ----- asTraitComposition "For convenience the composition {T1. T2 ...} is the same as T1 + T2 + ..." ^self isEmpty ifFalse: [ self size = 1 ifTrue: [self first asTraitComposition] + ifFalse: [self reduce: [:left :right | left + right]]] - ifFalse: [ - self copyWithoutFirst - inject: self first - into: [:left :right | left + right]]] ifTrue: [ClassDescription newTraitComposition]!

1 0

The Trunk: WebClient-Core-ul.133.mcz
by commits＠source.squeak.org 07 Oct '23

07 Oct '23

Levente Uzonyi uploaded a new version of WebClient-Core to project The Trunk: http://source.squeak.org/trunk/WebClient-Core-ul.133.mcz ==================== Summary ==================== Name: WebClient-Core-ul.133 Author: ul Time: 7 October 2023, 10:02:11.302663 am UUID: 6d91d8f9-0dc5-434b-95b4-486573f7dde9 Ancestors: WebClient-Core-ul.132 In WebSocket07>>readFrameFrom:, use #reduce: instead of a combination of #allButFirst, #first and #inject:into:. Perhaps a streaming solution would be better if there's a chance of frameData being large to avoid n^2 runtime. =============== Diff against WebClient-Core-ul.132 =============== Item was changed: ----- Method: WebSocket07>>readFrameFrom: (in category 'read/write') ----- readFrameFrom: aStream "Reads one frame worth of data" | byte length final mask opcode payload | byte := (aStream next ifNil:[^nil]) asInteger. final := byte anyMask: 128. opcode := byte bitAnd: 15. byte := (aStream next ifNil:[^nil]) asInteger. mask := byte anyMask: 128. length := byte bitAnd: 127. length = 126 ifTrue:[ length := self nextNumber: 2 from: aStream. ] ifFalse:[ length = 127 ifTrue:[length := self nextNumber: 8 from: aStream]. ]. mask ifTrue:[mask := aStream next: 4] ifFalse:[mask := nil]. payload := ByteArray new: length. aStream next: length into: payload. "Unmask payload" mask ifNotNil:[self xor: payload with: mask]. "First see if this is a control frame which can be inserted into fragmented messages." opcode >= 8 ifTrue:[^self handleControl: payload type: opcode]. "Now see if this is a fragmented message" frameType ifNil:[ frameType := opcode. frameData := Array with: payload. ] ifNotNil:[ frameData := frameData copyWith: payload. ]. final ifTrue:[ "This was the final fragment" frameData size = 1 ifTrue:[ payload := frameData first ] ifFalse:[ + payload := frameData reduce: [ :a :b | a, b ]. - payload := frameData allButFirst - inject: frameData first - into:[:each :next| each, next]. ]. frameType = 1 ifTrue:[payload := payload asString utf8ToSqueak]. self handleMessage: payload type: frameType. frameType := frameData := nil. ]. !

1 0

interpreter VM on RISC-V with negative bytecodes per second
by Jecel Assumpcao Jr 05 Oct '23

05 Oct '23

Yesterday I managed to compile the intepreter VM on a StarFive VisionFive 2 board (after replacing the config.guess with a current one having gone through several dead ends before figuring that out). The images in the download page have no indication about being V3 or Spur, so I first tried a 5.0 image and when that didn't work I got a 4.0 one that did. I was a bit surprised at how snappy the interpreted Squeak felt on a 1.5GHz reasonably simple RISC-V. Most other applications on that machine (web browser, Open Office, a digital simulator in Java) run much slower than on my work computer (Intel Core i5-8259U CPU @ 2.30GHz) so I didn't expect that. Running "0 tinyBenchmarks" got me a result of about -660 thousand bytecodes per second and 47 thousand sends per second. I will try to find out how this negative result came about and it is very likely that the other result is completely wrong as well. Here are the results on my work machine (with a Cog VM): 0 tinyBenchmarks '1150561797 bytecodes/sec; 32470183 sends/sec' Having the email client able to evaluate Smalltalk expressions can be very convenient at times ;-) -- Jecel

5 10

The Trunk: Tools-mt.1235.mcz
by commits＠source.squeak.org 05 Oct '23

05 Oct '23

Marcel Taeumel uploaded a new version of Tools to project The Trunk: http://source.squeak.org/trunk/Tools-mt.1235.mcz ==================== Summary ==================== Name: Tools-mt.1235 Author: mt Time: 5 October 2023, 10:28:11.042807 am UUID: 81a8a9b8-417d-9245-827e-0dbb0449c568 Ancestors: Tools-mt.1234 In TreeBrowser, fixes performance issue with "-- all --" class/system category. =============== Diff against Tools-mt.1234 =============== Item was changed: ----- Method: TreeBrowser>>allClassesInCategory (in category 'class tree - support') ----- allClassesInCategory "Answer a list of all class objects in the selected system category." self hasSystemCategorySelected ifFalse: [^ #()]. + self isShowingAllClasses ifTrue: [^ #() "See #classChildren:. No support for #updateTreesIfNeeded."]. self isHierarchy ifTrue: [^ self allClassesForHierarchy]. ^ self class showAllClassesInPackage ifFalse: [self defaultClassList collect: [:className | self environment classNamed: className] thenSelect: [:classOrNil | classOrNil notNil "guard environment"]] ifTrue: [PackageOrganizer default packageNamed: selectedSystemCategory ifPresent: [:pkg | pkg classes] ifAbsent: [self defaultClassList collect: [:className | self environment classNamed: className] thenSelect: [:classOrNil | classOrNil notNil "guard environment"]]]! Item was changed: ----- Method: TreeBrowser>>allClassesInCategoryAndOther (in category 'class tree - support') ----- allClassesInCategoryAndOther "Like #allClassesInCategory, but adds missing inner classes for a complete-looking inheritance tree." | allClasses roots more | + extraClasses := #(). + self hasSystemCategorySelected ifFalse: [^ #()]. + self isShowingAllClasses ifTrue: [^ #() "See #classChildren:. No support for #updateTreesIfNeeded."]. self isHierarchy ifTrue: [^ self allClassesForHierarchy]. + - allClasses := self allClassesInCategory. roots := self classRootsFrom: allClasses. "See #classRoots for the widget callback" more := Set new. "Which inner classes to add?" roots do: [:inner | roots do: [:outer | (inner inheritsFrom: outer) ifTrue: [ | addMore | addMore := true. inner allSuperclassesDo: [:missing | (addMore := addMore and: [missing ~~ outer]) ifTrue: [more add: missing] ]] ]]. self flag: #ugly. "mt: This is an unexpected side effect and clashes with how #updateTreesIfNeeded works... yet, it does not send a superfluous changed event to the view..." extraClasses := more asArray. ^ more ifEmpty: [allClasses] ifNotEmpty: [allClasses, more asArray]!

1 0

renaming a category to a different package does not mark the packages dirty
by Eliot Miranda 04 Oct '23

04 Oct '23

Hi All, I just changed the category of Array>>allLiteralsDo: and Array>>#literalEqual: by renaming Array's "literals" category to "*Kernel-literals". But this failed to mark Collections and Kernel packages dirty in the Monticello browser. This is a bug. SOmeone who understands how SystemChangeNotifier is accessed should fix this. I looked at it but couldn't work out what the right thing to do is quickly enough. Cheers _,,,^..^,,,_ best, Eliot

2 1

The Trunk: Monticello-mt.800.mcz
by commits＠source.squeak.org 04 Oct '23

04 Oct '23

Marcel Taeumel uploaded a new version of Monticello to project The Trunk: http://source.squeak.org/trunk/Monticello-mt.800.mcz ==================== Summary ==================== Name: Monticello-mt.800 Author: mt Time: 4 October 2023, 5:10:04.725674 pm UUID: 03143750-1ae2-e846-95e0-37c48100f5cb Ancestors: Monticello-mt.799 Okay. Class-side instance variables are tricky... :-) =============== Diff against Monticello-mt.799 =============== Item was changed: (PackageInfo named: 'Monticello') postscript: '"Adds support for protocol renaming" + MCWorkingCopy registerForNotifications.'! - MCPackageManager registerForNotifications.'!

1 0

Jump to page: