Squeak-dev July 2014

squeak-dev@lists.squeakfoundation.org

31 participants
223 discussions

The Inbox: Collections-cmm.576.mcz
by commits＠source.squeak.org 24 Jul '14

24 Jul '14

Chris Muller uploaded a new version of Collections to project The Inbox: http://source.squeak.org/inbox/Collections-cmm.576.mcz ==================== Summary ==================== Name: Collections-cmm.576 Author: cmm Time: 24 July 2014, 10:39:06.452 am UUID: 17b60ac6-55b3-4706-b1c8-d826e6ad5657 Ancestors: Collections-eem.575 When a SequenceableCollection contains duplicate keys, #findBinaryIndex:do:ifNone: should answer the index of the _earliest_. =============== Diff against Collections-eem.575 =============== Item was changed: ----- Method: SequenceableCollection>>findBinaryIndex:do:ifNone: (in category 'enumerating') ----- + findBinaryIndex: aBlock do: actionBlock ifNone: exceptionBlock - findBinaryIndex: aBlock do: actionBlock ifNone: exceptionBlock "Search for an element in the receiver using binary search. The argument aBlock is a one-element block returning 0 - if the element is the one searched for <0 - if the search should continue in the first half >0 - if the search should continue in the second half If found, evaluate actionBlock with the index as argument If no matching element is found, evaluate exceptionBlock, with the indexes of the 'bounding' elements as optional arguments. Warning: Might give invalid indexes, see examples below. Examples: + #(1 3 5 7 11 11 15 23) - #(1 3 5 7 11 15 23) findBinaryIndex: [ :arg | 11 - arg ] do: [ :found | found ] ifNone: [ :a :b | ('between: ', {a. b} printString)] #(1 3 5 7 11 15 23) findBinaryIndex: [ :arg | 12 - arg ] do: [ :found | found ] ifNone: [ :a :b | ('between: ', {a. b} printString) ] #(1 3 5 7 11 15 23) d findBinaryIndex: [ :arg | 0.5 - arg ] do: [ :found | found ] ifNone: [ :a :b | ('between: ', {a. b} printString) ] #(1 3 5 7 11 15 23) findBinaryIndex: [ :arg | 25 - arg ] do: [ :found | found ] ifNone: [ :a :b | ('between: ',{a. b} printString) ] " | index low high | low := 1. high := self size. + [ index := high + low // 2. + low > high ] whileFalse: + [ | test | + test := aBlock value: (self at: index). + test = 0 + ifTrue: + [ "In case there are duplicate keys, walk backward to the first." + [ index > 1 and: [ (aBlock value: (self at: index - 1)) = 0 ] ] whileTrue: [ index := index - 1 ]. + ^ actionBlock value: index ] + ifFalse: + [ test > 0 - [ - index := high + low // 2. - low > high ] whileFalse: [ - | test | - test := aBlock value: (self at: index). - test = 0 - ifTrue: [ ^actionBlock value: index ] - ifFalse: [ test > 0 ifTrue: [ low := index + 1 ] ifFalse: [ high := index - 1 ] ] ]. + ^ exceptionBlock + cull: high + cull: low! - ^exceptionBlock cull: high cull: low!

1 0

The Trunk: Kernel-nice.860.mcz
by commits＠source.squeak.org 24 Jul '14

24 Jul '14

Nicolas Cellier uploaded a new version of Kernel to project The Trunk: http://source.squeak.org/trunk/Kernel-nice.860.mcz ==================== Summary ==================== Name: Kernel-nice.860 Author: nice Time: 24 July 2014, 1:00:22.683 pm UUID: 368a9aba-7c65-ad4b-9499-75a652bed57f Ancestors: Kernel-eem.859 Avoid engaging LargeInteger arithmetic (asTrueFraction) just for testing the sign of a Float =============== Diff against Kernel-eem.859 =============== Item was changed: ----- Method: Float>>sign (in category 'mathematical functions') ----- sign "Answer 1 if the receiver is greater than 0, -1 if less than 0, else 0. Handle IEEE-754 negative-zero by reporting a sign of -1" + self > 0.0 ifTrue: [^ 1]. + (self < 0.0 or: [((self at: 1) bitShift: -31) = 1]) ifTrue: [^ -1]. - self > 0 ifTrue: [^ 1]. - (self < 0 or: [((self at: 1) bitShift: -31) = 1]) ifTrue: [^ -1]. ^ 0!

1 0

Daily Commit Log
by commits＠source.squeak.org 24 Jul '14

24 Jul '14

Changes to Trunk (http://source.squeak.org/trunk.html) in the last 24 hours: http://lists.squeakfoundation.org/pipermail/packages/2014-July/007191.html Name: KernelTests-dtl.276 Ancestors: KernelTests-eem.275 Add tests for digit access in large integers. Given a 20 byte large positive or negative integer, verify that digit length, digit at, and digit replacement produce the expected results. =============================================

1 0

Re: [squeak-dev] The Trunk: Graphics-nice.296.mcz
by Nicolas Cellier 24 Jul '14

24 Jul '14

2014-07-22 1:29 GMT+02:00 <commits(a)source.squeak.org>: > Nicolas Cellier uploaded a new version of Graphics to project The Trunk: > http://source.squeak.org/trunk/Graphics-nice.296.mcz > > ==================== Summary ==================== > > Name: Graphics-nice.296 > Author: nice > Time: 22 July 2014, 1:28:48.356 am > UUID: 3f6d869e-9929-4e08-b137-aae22dd3eee6 > Ancestors: Graphics-nice.295 > > Don't try to accelerate GIF reading by copying the stream in memory. > This should be the purpose of a decent file buffering policy, and if ever > that was not the case, then we should fix the file buffering policy rather > than trying to patch each and every client site. > > Especially when the patch is testing that the stream is in memory with a > class identity to ReadWriteStream!!! > Frankly, how do we know that it's not a RWBinaryOrTextStream ;) > > So why reading an image with a ReadWriteStream? > Oh, sure, this is in ImageReadWriter, so we may need a ReadWriteStream? > No, no, and no! > This thing will either read or write but NEVER read and write. > One simple role, one simple class, no UberPotentSwissKnifeStream with > UberComplexStateToManage (readPosition + writePosition + readLimit + > writeLimit + readBuffer + writeBuffer + readWriteBufferInteraction). > > What's coming next? > Easy to guess, see what (ImageReadWriter>>on:) does frivously to our > stream: > reset??? Hey, ImageReadWriter, why do you take the responsibility to reset > the stream? Did you ask the permission? Did you think of possible side > effects? > the answer is summarized there: readNativeResourceFrom: byteStream | img aStream | (byteStream isKindOf: FileStream) ifTrue:[ "Ugly, but ImageReadWriter will send #reset which is implemented as #reopen and we may not be able to do so." aStream := RWBinaryOrTextStream with: byteStream contents. ] ifFalse:[ aStream := byteStream. ]. Ugly says it all. > I'm quite sure those reset has something to do with following usage: > (ReadWriteStream with: 'foo') next -> nil > but: > (ReadWriteStream with: 'foo') reset; next -> $f > Pure Read/Write madness. > There is more room for cleanup here, but only after I get rid of more > R/WThing (TM), otherwise I could break some insane expectations. > > =============== Diff against Graphics-nice.295 =============== > > Item was changed: > ----- Method: AnimatedGIFReadWriter>>allImages (in category 'accessing') > ----- > allImages > | body colorTable | > - stream class == ReadWriteStream ifFalse: [ > - stream binary. > - self on: (ReadWriteStream with: (stream > contentsOfEntireFile))]. > localColorTable := nil. > forms := OrderedCollection new. > delays := OrderedCollection new. > comments := OrderedCollection new. > self readHeader. > [(body := self readBody) == nil] > whileFalse: [colorTable := localColorTable > ifNil: [colorPalette]. > transparentIndex > ifNotNil: [transparentIndex + 1 > > colorTable size > ifTrue: [colorTable := > colorTable forceTo: transparentIndex + 1 paddingWith: Color white]. > colorTable at: transparentIndex + > 1 put: Color transparent]. > body colors: colorTable. > forms add: body. > delays add: delay]. > ^ forms! > > Item was changed: > ----- Method: GIFReadWriter>>nextImage (in category 'accessing') ----- > nextImage > + "Read in the next GIF image from the stream." > - "Read in the next GIF image from the stream. Read it all into > - memory first for speed." > > | f thisImageColorTable | > - stream class == ReadWriteStream ifFalse: [ > - stream binary. > - self on: (ReadWriteStream with: (stream > contentsOfEntireFile))]. > > localColorTable := nil. > self readHeader. > f := self readBody. > self close. > f == nil ifTrue: [^ self error: 'corrupt GIF file']. > > thisImageColorTable := localColorTable ifNil: [colorPalette]. > transparentIndex ifNotNil: [ > transparentIndex + 1 > thisImageColorTable size ifTrue: [ > thisImageColorTable := thisImageColorTable > forceTo: transparentIndex + 1 > paddingWith: Color white > ]. > thisImageColorTable at: transparentIndex + 1 put: Color > transparent > ]. > f colors: thisImageColorTable. > ^ f > ! > > >

3 3

The Trunk: KernelTests-dtl.276.mcz
by commits＠source.squeak.org 23 Jul '14

23 Jul '14

David T. Lewis uploaded a new version of KernelTests to project The Trunk: http://source.squeak.org/trunk/KernelTests-dtl.276.mcz ==================== Summary ==================== Name: KernelTests-dtl.276 Author: dtl Time: 22 July 2014, 10:38:16.634 pm UUID: f6128ad6-87e1-449f-b34d-23c4ae265462 Ancestors: KernelTests-eem.275 Add tests for digit access in large integers. Given a 20 byte large positive or negative integer, verify that digit length, digit at, and digit replacement produce the expected results. =============== Diff against KernelTests-eem.275 =============== Item was added: + ----- Method: LargeNegativeIntegerTest>>testDigitAt (in category 'tests') ----- + testDigitAt + + | lni | + lni := -114605103402541699037609980192546360895434064385. + 1 to: 20 do: [:i | | digit | + digit := lni digitAt: i. + self assert: i equals: digit] + ! Item was added: + ----- Method: LargeNegativeIntegerTest>>testDigitAtPut (in category 'tests') ----- + testDigitAtPut + + | lni | + lni := LargeNegativeInteger new: 20. + 1 to: 20 do: [:i | lni digitAt: i put: i]. + self assert: -114605103402541699037609980192546360895434064385equals: lni + ! Item was added: + ----- Method: LargeNegativeIntegerTest>>testDigitLength (in category 'tests') ----- + testDigitLength + + | lni | + lni := -114605103402541699037609980192546360895434064385. + self assert: 20 equals: lni digitLength + ! Item was added: + ----- Method: LargeNegativeIntegerTest>>testReplaceFromToWithStartingAt (in category 'tests') ----- + testReplaceFromToWithStartingAt + + | lni20 lni7 | + lni20 := LargeNegativeInteger new: 20. + 1 to: 20 do: [:i | lni20 digitAt: i put: i]. + lni7 := LargeNegativeInteger new: 7. + 1 to: 7 do: [:i | lni7 digitAt: i put: 11 - i]. + lni20 replaceFrom: 6 to: 10 with: lni7 startingAt: 2. + "unmodified digits" + (1 to: 5) , (11 to: 20) do: [:e | | digit | + digit := lni20 digitAt: e. + self assert: e equals: digit]. + "replaced digits" + 6 to: 10 do: [:e | | digit replacementDigit | + digit := lni20 digitAt: e. + replacementDigit := lni7 digitAt: e - 4. + self assert: replacementDigit equals: digit] + ! Item was added: + ----- Method: LargePositiveIntegerTest>>testDigitAt (in category 'tests') ----- + testDigitAt + + | lpi | + lpi := 114605103402541699037609980192546360895434064385. + 1 to: 20 do: [:i | | digit | + digit := lpi digitAt: i. + self assert: i equals: digit] + ! Item was added: + ----- Method: LargePositiveIntegerTest>>testDigitAtPut (in category 'tests') ----- + testDigitAtPut + + | lpi | + lpi := LargePositiveInteger new: 20. + 1 to: 20 do: [:i | lpi digitAt: i put: i]. + self assert: 114605103402541699037609980192546360895434064385equals: lpi + ! Item was added: + ----- Method: LargePositiveIntegerTest>>testDigitLength (in category 'tests') ----- + testDigitLength + + | lpi | + lpi := 114605103402541699037609980192546360895434064385. + self assert: 20 equals: lpi digitLength + ! Item was added: + ----- Method: LargePositiveIntegerTest>>testReplaceFromToWithStartingAt (in category 'tests') ----- + testReplaceFromToWithStartingAt + + | lpi20 lpi7 | + lpi20 := LargePositiveInteger new: 20. + 1 to: 20 do: [:i | lpi20 digitAt: i put: i]. + lpi7 := LargePositiveInteger new: 7. + 1 to: 7 do: [:i | lpi7 digitAt: i put: 11 - i]. + lpi20 replaceFrom: 6 to: 10 with: lpi7 startingAt: 2. + "unmodified digits" + (1 to: 5) , (11 to: 20) do: [:e | | digit | + digit := lpi20 digitAt: e. + self assert: e equals: digit]. + "replaced digits" + 6 to: 10 do: [:e | | digit replacementDigit | + digit := lpi20 digitAt: e. + replacementDigit := lpi7 digitAt: e - 4. + self assert: replacementDigit equals: digit] + !

1 0

Daily Commit Log
by commits＠source.squeak.org 23 Jul '14

23 Jul '14

Changes to Trunk (http://source.squeak.org/trunk.html) in the last 24 hours: http://lists.squeakfoundation.org/pipermail/packages/2014-July/007187.html Name: Graphics-nice.297 Ancestors: Graphics-nice.296 RWStream shoot them up: score of the evening nice 1/RWStream 0. If writing with a WriteStream just works, why summon the devil? ============================================= http://lists.squeakfoundation.org/pipermail/packages/2014-July/007188.html Name: MorphicExtras-nice.153 Ancestors: MorphicExtras-nice.152 RWStream shoot them up part 2: How to write an EPS preview using a WriteStream. ============================================= http://lists.squeakfoundation.org/pipermail/packages/2014-July/007189.html Name: Nebraska-nice.37 Ancestors: Nebraska-nice.36 RWStream shoot them up part 3. How to decodeImage: without a RWBinaryOrTextStream. (Let's not summon the devil when we don't need his superpowers). ============================================= http://lists.squeakfoundation.org/pipermail/packages/2014-July/007190.html Name: System-nice.680 Ancestors: System-kfr.679 RWStream Shoot them up part 4: How to drop an image without a RWBinaryOrTextStream. I feel like applying a balm on chicken pox one by one. There are a lot of them, my image is crippled, and it's itching considerably! =============================================

1 0

[Bug] some morph asEPS will fail
by Nicolas Cellier 23 Jul '14

23 Jul '14

Because anEPSCanvas doesNotUnderstand: aCanvas frameAndFillRoundRect: bounds radius: 8 fillStyle: gradient borderWidth: 1 borderColor: borderColor When adding new protocol to a specific canvas, one should think of the brothers and sisters in Canvas hierarchy...

1 0

The Trunk: System-nice.680.mcz
by commits＠source.squeak.org 23 Jul '14

23 Jul '14

Nicolas Cellier uploaded a new version of System to project The Trunk: http://source.squeak.org/trunk/System-nice.680.mcz ==================== Summary ==================== Name: System-nice.680 Author: nice Time: 23 July 2014, 12:15:58.869 am UUID: 46ed1f7b-d575-4d6d-9985-5e0e9a40143d Ancestors: System-kfr.679 RWStream Shoot them up part 4: How to drop an image without a RWBinaryOrTextStream. I feel like applying a balm on chicken pox one by one. There are a lot of them, my image is crippled, and it's itching considerably! =============== Diff against System-kfr.679 =============== Item was changed: ----- Method: ExternalDropHandler class>>defaultImageHandler (in category 'private') ----- defaultImageHandler ^ExternalDropHandler type: 'image/' extension: nil action: [:stream :pasteUp :event | | image sketch | stream binary. + image := Form fromBinaryStream: stream contents asByteArray readStream. - image := Form fromBinaryStream: ((RWBinaryOrTextStream with: stream contents) reset). Project current resourceManager addResource: image url: (FileDirectory urlForFileNamed: stream name) asString. sketch := World drawingClass withForm: image. pasteUp addMorph: sketch centeredNear: event position. image := sketch := nil]!

1 0

The Trunk: Nebraska-nice.37.mcz
by commits＠source.squeak.org 23 Jul '14

23 Jul '14

Nicolas Cellier uploaded a new version of Nebraska to project The Trunk: http://source.squeak.org/trunk/Nebraska-nice.37.mcz ==================== Summary ==================== Name: Nebraska-nice.37 Author: nice Time: 23 July 2014, 12:03:09.798 am UUID: cd1e75c9-63de-4f9c-80b2-d4928cf48bff Ancestors: Nebraska-nice.36 RWStream shoot them up part 3. How to decodeImage: without a RWBinaryOrTextStream. (Let's not summon the devil when we don't need his superpowers). =============== Diff against Nebraska-nice.36 =============== Item was changed: ----- Method: CanvasDecoder class>>decodeImage: (in category 'decoding') ----- decodeImage: string | bitsStart depth width height bits rs numColors colorArray | bitsStart := string indexOf: $|. bitsStart = 0 ifTrue: [^nil]. rs := ReadStream on: string. rs peek == $C ifTrue: [ rs next. numColors := Integer readFromString: (rs upTo: $,). colorArray := Array new: numColors. 1 to: numColors do: [ :i | colorArray at: i put: (self decodeColor: (rs next: 12)) ]. ]. depth := Integer readFromString: (rs upTo: $,). width := Integer readFromString: (rs upTo: $,). height := Integer readFromString: (rs upTo: $|). + bits := Bitmap newFromStream: rs upToEnd asByteArray readStream. - bits := Bitmap newFromStream: (RWBinaryOrTextStream with: rs upToEnd) binary reset. colorArray ifNil: [ ^Form extent: width@height depth: depth bits: bits ]. ^(ColorForm extent: width@height depth: depth bits: bits) colors: colorArray !

1 0

The Trunk: MorphicExtras-nice.153.mcz
by commits＠source.squeak.org 22 Jul '14

22 Jul '14

Nicolas Cellier uploaded a new version of MorphicExtras to project The Trunk: http://source.squeak.org/trunk/MorphicExtras-nice.153.mcz ==================== Summary ==================== Name: MorphicExtras-nice.153 Author: nice Time: 22 July 2014, 11:47:16.784 pm UUID: 411a90c9-8122-4287-b19a-3498e2d264cd Ancestors: MorphicExtras-nice.152 RWStream shoot them up part 2: How to write an EPS preview using a WriteStream. =============== Diff against MorphicExtras-nice.152 =============== Item was changed: ----- Method: EPSCanvas>>writeEPSPreviewImageFor: (in category 'private') ----- writeEPSPreviewImageFor: aMorph + | form stream string newExtent | - | form stream string lines newExtent | newExtent := (aMorph width roundUpTo: 8) @ aMorph height. form := aMorph imageForm: 1 forRectangle: (aMorph bounds origin extent: newExtent). + stream := WriteStream on: (String new: (form bits byteSize * 2.04) asInteger). - stream := RWBinaryOrTextStream on: (String new: (form bits byteSize * 2.04) asInteger). form storePostscriptHexOn: stream. string := stream contents. - lines := string occurrencesOf: Character cr. "%%BeginPreview: 80 24 1 24" "width height depth " + target print: '%%BeginPreview: '; write: newExtent; space; write: form depth; space; write: string lineCount; cr. - target print: '%%BeginPreview: '; write: newExtent; space; write: form depth; space; write: lines; cr. + string linesDo: [:e | target nextPut: $%; nextPutAll: e; cr.]. - stream position: 0. - [ stream atEnd ] whileFalse: [ - target nextPut: $%; nextPutAll: (stream upTo: Character cr); cr. - lines := lines - 1. - ]. target print: '%%EndPreview'; cr. !

1 0

← Newer
1
...
6
7
8
9
10
11
12
...
23
Older →

Jump to page: