Squeak-dev November 2010

squeak-dev@lists.squeakfoundation.org

77 participants
254 discussions

New Trunk Image: 10548
by Casey Ransberger 03 Nov '10

03 Nov '10

Grab it here: http://ftp.squeak.org/trunk/Squeak4.2-10548-alpha.zip 2805 run, 2767 passes, 9 expected failures, 28 failures, 1 errors, 0 unexpected passes Mac OS X 10.6.4 (Squeak4.2.5beta1U) The error is in WeakFinalizersTest>>#testNewFinalizationSupported, which it would seem is not too big of a deal (just means the VM doesn't support Igor's new finalization stuff yet.) Brave and noble souls will download it on various platforms, run the tests, and generally make sure the image is good. -- Casey Ransberger

5 9

Daily Commit Log
by commits＠source.squeak.org 03 Nov '10

03 Nov '10

Changes to Trunk (http://source.squeak.org/trunk.html) in the last 24 hours: http://lists.squeakfoundation.org/pipermail/packages/2010-November/003871.h… Name: Tools-ul.275 Ancestors: Tools-ul.274 - fix: copy class instance variables and class comments when creating a copy of a class ============================================= http://lists.squeakfoundation.org/pipermail/packages/2010-November/003872.h… Name: Collections-ul.400 Ancestors: Collections-ul.399 - use paragma declarations for #inline: and #var:declareC: ============================================= http://lists.squeakfoundation.org/pipermail/packages/2010-November/003873.h… Name: Graphics-ul.152 Ancestors: Graphics-dtl.151 - use paragma declarations for #inline: and #var:declareC: ============================================= http://lists.squeakfoundation.org/pipermail/packages/2010-November/003874.h… Name: Sound-ul.26 Ancestors: Sound-dtl.25 - use paragma declarations for #inline: and #var:declareC: ============================================= http://lists.squeakfoundation.org/pipermail/packages/2010-November/003875.h… Name: Files-ul.90 Ancestors: Files-ul.89 - Integrated Eliot's stdio changes with a minor fix/optimization ============================================= http://lists.squeakfoundation.org/pipermail/packages/2010-November/003876.h… Name: Multilingual-ul.130 Ancestors: Multilingual-cmm.129 - make sure MultiByteFileStream can also be used as an stdio stream ============================================= http://lists.squeakfoundation.org/pipermail/packages/2010-November/003877.h… Name: Files-ul.91 Ancestors: Files-ul.90 - initialize StdioHandles without restarting the image ============================================= http://lists.squeakfoundation.org/pipermail/packages/2010-November/003878.h… Name: Collections-ul.401 Ancestors: Collections-ul.400 - added Symbol >> #cull:, so unary symbols can be used instead of monadic blocks where #cull: is sent to them (e.g. #ifNotNil:) ============================================= http://lists.squeakfoundation.org/pipermail/packages/2010-November/003879.h… Name: CollectionsTests-ul.174 Ancestors: CollectionsTests-ar.173 - mark WeakFinalizersTest >> #testNewFinalizationSupported as expectedFailure if the VM doesn't support the new finalization scheme ============================================= http://lists.squeakfoundation.org/pipermail/packages/2010-November/003880.h… Name: Collections-ul.402 Ancestors: Collections-ul.401 - enhanced AttributedTextStream so it can be used as a TextStream in some cases. - recategorized a few methods, fixed some comments ============================================= http://lists.squeakfoundation.org/pipermail/packages/2010-November/003881.h… Name: System-ul.389 Ancestors: System-ul.388 - a bit faster ClassDiffBuilder >> #split: - TextDiffBuilder >> #buildDisplayPatch uses AttributedTextStream instead of TextStream, so it takes O(n) time to build a display patch, instead of O(n^2) =============================================

1 0

Collection's #fold: vs #reduce:
by Levente Uzonyi 02 Nov '10

02 Nov '10

Hi, we have these two methods which do exactly the same thing. #reduce: was added by Andreas during the developement of Squeak 4.1. #fold was added by Eliot for Cog VMMaker compatibility. One of them is superfluous. I can image the following solutions: 1) Replace senders of #fold: in VMMaker to use #reduce:, and remove #fold: from the image. 2) Replace #fold:'s implementation to self reduce: aBlock. I benchmarked the two methods and #reduce: is a bit faster on CogVM. There's no difference on SqueakVM. Cheers, Levente

7 20

Re: [squeak-dev] SocketStream not a Stream?
by Ralph Boland 02 Nov '10

02 Nov '10

Message: 16 Date: Tue, 02 Nov 2010 08:06:11 +0100 From: G?ran Krampe <goran(a)krampe.se> Subject: Re: [squeak-dev] SocketStream not a Stream? To: squeak-dev(a)lists.squeakfoundation.org Message-ID: <4CCFB863.1080302(a)krampe.se> Content-Type: text/plain; charset=ISO-8859-1; format=flowed > On 11/02/2010 04:55 AM, Andreas Raab wrote: > > Hi - > > > > I (accidentally) noticed that SocketStream isn't a subclass of Stream. > > Is there any reason why this should remain true? It strikes me as odd as > > Stream has very few requirements (basically #next, #nextPut:, and > > #atEnd) which are trivially fulfilled by SocketStream. Anyone seeing a > > reason not to change this? > > > > Cheers, > > - Andreas > Well, when I wrote the current version I think I decided against it > since I did not want any "inherited bugs" so to speak. If a clean and > useful inheritance is possible, then sure. The other reason it might be > unwise is that it does both input and output - but of course that does > not mean a useful inheritance is possible anyway. > regards, Göran I don't know anything about SocketStream so my comments may be way off base. If SocketStream does both output and input then it sounds like it should have an instance variable (a Stream) for input and an instance variable (another Stream) for output. On the other hand if what you mean by doing input and output is that is reads and writes to the same place then it sounds like SocketStream should not only be a subclass of Stream but a subclass of ReadWriteStream. Like I said, I don't know anything about SocketStream. Regards, Ralph Boland

4 5

The Trunk: System-ul.389.mcz
by commits＠source.squeak.org 02 Nov '10

02 Nov '10

Levente Uzonyi uploaded a new version of System to project The Trunk: http://source.squeak.org/trunk/System-ul.389.mcz ==================== Summary ==================== Name: System-ul.389 Author: ul Time: 2 November 2010, 10:06:25.693 pm UUID: 77647d29-a395-fe4e-a7d0-013850bdfae2 Ancestors: System-ul.388 - a bit faster ClassDiffBuilder >> #split: - TextDiffBuilder >> #buildDisplayPatch uses AttributedTextStream instead of TextStream, so it takes O(n) time to build a display patch, instead of O(n^2) =============== Diff against System-ul.388 =============== Item was changed: ----- Method: ClassDiffBuilder>>split: (in category 'private') ----- split: aString "I return an array with aString splitted by Character >> #separators." ^Array streamContents: [ :stream | + | input separators | + input := aString readStream. + separators := CharacterSet separators. + [ input atEnd ] whileFalse: [ + | word separator | + word := input + upToAnyOf: separators + do: [ :matchingSeparator | + separator := matchingSeparator ]. + stream nextPut: (separator + ifNil: [ word ] + ifNotNil: [ word, separator asString ]) ] ]! - | out | - out := aString copy writeStream. - aString do: [ :c | - out nextPut: c. - c isSeparator ifTrue:[ - stream nextPut: out contents. - out reset ] ]. - out position = 0 ifFalse: [ - stream nextPut: out contents ] ]! Item was changed: ----- Method: TextDiffBuilder>>buildDisplayPatch (in category 'creating patches') ----- buildDisplayPatch + | stream | + stream := AttributedTextStream new. + self + patchSequenceDoIfMatch: [ :string | + self print: string withAttributes: NormalTextAttributes on: stream ] + ifInsert: [ :string | + self print: string withAttributes: InsertTextAttributes on: stream ] + ifRemove: [ :string | + self print: string withAttributes: RemoveTextAttributes on: stream ]. + ^stream contents! - ^Text streamContents: [ :stream | - self - patchSequenceDoIfMatch: [ :string | - self print: string withAttributes: NormalTextAttributes on: stream ] - ifInsert: [ :string | - self print: string withAttributes: InsertTextAttributes on: stream ] - ifRemove: [ :string | - self print: string withAttributes: RemoveTextAttributes on: stream ] ]!

1 0

The Trunk: Collections-ul.402.mcz
by commits＠source.squeak.org 02 Nov '10

02 Nov '10

Levente Uzonyi uploaded a new version of Collections to project The Trunk: http://source.squeak.org/trunk/Collections-ul.402.mcz ==================== Summary ==================== Name: Collections-ul.402 Author: ul Time: 2 November 2010, 10:03:54.289 pm UUID: c07fdead-7f3a-544f-82dd-4bfe4ef28365 Ancestors: Collections-ul.401 - enhanced AttributedTextStream so it can be used as a TextStream in some cases. - recategorized a few methods, fixed some comments =============== Diff against Collections-ul.401 =============== Item was changed: + ----- Method: AttributedTextStream>>contents (in category 'accessing') ----- - ----- Method: AttributedTextStream>>contents (in category 'retrieving the text') ----- contents | ans | currentRun > 0 ifTrue:[ attributeValues nextPut: currentAttributes. attributeRuns nextPut: currentRun. currentRun := 0]. ans := Text string: characters contents runs: (RunArray runs: attributeRuns contents values: attributeValues contents). ^ans! Item was added: + ----- Method: AttributedTextStream>>cr (in category 'character writing') ----- + cr + "Append a carriage return character to the receiver." + + self nextPut: Character cr! Item was added: + ----- Method: AttributedTextStream>>crlf (in category 'character writing') ----- + crlf + "Append a carriage return and a line feed to the receiver." + + self nextPut: Character cr; nextPut: Character lf! Item was changed: + ----- Method: AttributedTextStream>>currentAttributes (in category 'accessing') ----- - ----- Method: AttributedTextStream>>currentAttributes (in category 'access') ----- currentAttributes "return the current attributes" ^currentAttributes! Item was changed: + ----- Method: AttributedTextStream>>currentAttributes: (in category 'accessing') ----- - ----- Method: AttributedTextStream>>currentAttributes: (in category 'access') ----- currentAttributes: newAttributes "set the current attributes" (currentRun > 0 and:[currentAttributes ~= newAttributes]) ifTrue:[ attributeRuns nextPut: currentRun. attributeValues nextPut: currentAttributes. currentRun := 0. ]. currentAttributes := newAttributes. ! Item was changed: + ----- Method: AttributedTextStream>>initialize (in category 'initialize-release') ----- - ----- Method: AttributedTextStream>>initialize (in category 'private-initialization') ----- initialize + + characters := String new writeStream. + currentAttributes := #(). - characters := WriteStream on: String new. - currentAttributes := OrderedCollection new. currentRun := 0. + attributeValues := (Array new: 50) writeStream. + attributeRuns := (Array new: 50) writeStream! - attributeValues := WriteStream on: (Array new: 50). - attributeRuns := WriteStream on: (Array new: 50). ! Item was added: + ----- Method: AttributedTextStream>>lf (in category 'character writing') ----- + lf + "Append a line feed character to the receiver." + + self nextPut: Character lf! Item was changed: + ----- Method: AttributedTextStream>>nextPut: (in category 'accessing') ----- - ----- Method: AttributedTextStream>>nextPut: (in category 'stream protocol') ----- nextPut: aChar currentRun := currentRun + 1. ^characters nextPut: aChar! Item was changed: + ----- Method: AttributedTextStream>>nextPutAll: (in category 'accessing') ----- - ----- Method: AttributedTextStream>>nextPutAll: (in category 'stream protocol') ----- nextPutAll: aString "add an entire string with the same attributes" currentRun := currentRun + aString size. ^characters nextPutAll: aString.! Item was changed: + ----- Method: AttributedTextStream>>size (in category 'accessing') ----- - ----- Method: AttributedTextStream>>size (in category 'access') ----- size "number of characters in the stream so far" ^characters size! Item was added: + ----- Method: AttributedTextStream>>withAttribute:do: (in category 'accessing') ----- + withAttribute: attribute do: aBlock + + ^self withAttributes: { attribute } do: aBlock! Item was added: + ----- Method: AttributedTextStream>>withAttributes:do: (in category 'accessing') ----- + withAttributes: attributes do: aBlock + + | previousAttributes | + previousAttributes := currentAttributes. + [ + self currentAttributes: attributes. + aBlock value ] + ensure: [ self currentAttributes: previousAttributes ]! Item was changed: + ----- Method: Stream>>printOn: (in category 'printing') ----- - ----- Method: Stream>>printOn: (in category 'accessing') ----- printOn: stream super printOn: stream. stream space. self contents printOn: stream. ! Item was changed: ----- Method: WriteStream>>cr (in category 'character writing') ----- cr + "Append a carriage return character to the receiver." - "Append a return character to the receiver." self nextPut: Character cr! Item was changed: ----- Method: WriteStream>>crlf (in category 'character writing') ----- crlf + "Append a carriage return and a line feed to the receiver." - "Append a line feed character to the receiver." self nextPut: Character cr; nextPut: Character lf!

1 0

The Trunk: CollectionsTests-ul.174.mcz
by commits＠source.squeak.org 02 Nov '10

02 Nov '10

Levente Uzonyi uploaded a new version of CollectionsTests to project The Trunk: http://source.squeak.org/trunk/CollectionsTests-ul.174.mcz ==================== Summary ==================== Name: CollectionsTests-ul.174 Author: ul Time: 2 November 2010, 6:31:29.485 pm UUID: 3c5e057a-fe6f-e847-9595-4e45e1135ed8 Ancestors: CollectionsTests-ar.173 - mark WeakFinalizersTest >> #testNewFinalizationSupported as expectedFailure if the VM doesn't support the new finalization scheme =============== Diff against CollectionsTests-ar.173 =============== Item was added: + ----- Method: WeakFinalizersTest>>expectedFailures (in category 'as yet unclassified') ----- + expectedFailures + + ^WeakFinalizationList hasNewFinalization + ifTrue: [ #() ] + ifFalse: [ #(testNewFinalizationSupported) ]!

1 0

The Trunk: Collections-ul.401.mcz
by commits＠source.squeak.org 02 Nov '10

02 Nov '10

Levente Uzonyi uploaded a new version of Collections to project The Trunk: http://source.squeak.org/trunk/Collections-ul.401.mcz ==================== Summary ==================== Name: Collections-ul.401 Author: ul Time: 2 November 2010, 6:27:15.942 pm UUID: 0a1c53c0-73df-d84f-ab48-5d976d110df2 Ancestors: Collections-ul.400 - added Symbol >> #cull:, so unary symbols can be used instead of monadic blocks where #cull: is sent to them (e.g. #ifNotNil:) =============== Diff against Collections-ul.400 =============== Item was added: + ----- Method: Symbol>>cull: (in category 'evaluating') ----- + cull: anObject + + ^anObject perform: self!

1 0

Questions about parsing and orthogonal constructs....
by gi.joe 02 Nov '10

02 Nov '10

Hello everyone. I'm new to Squeak, and I chose to do a research paper for school on it. I have a few questions about the language. My first question is what components in Squeak can be successfully parsed by the compiler without necessarily creating a full program. For example, can a blank method be created? Create variables without using them? etc. My second question is what constructs in the language are NOT orthogonal. I'm not sure if there is a different term for this, but basically it means what constructs cannot be used within a construct of the same type. For example, can you use a function within a function? If any help can be provided, I would greatly appreciate it. Some examples and references to the ANSI Smalltalk Reference Manual would be much appreciated as well. I was trying to find the ansers in the ANSI manual haven't had much luck. -- View this message in context: http://forum.world.st/Questions-about-parsing-and-orthogonal-constructs-tp3… Sent from the Squeak - Dev mailing list archive at Nabble.com.

4 3

SocketStream not a Stream?
by Andreas Raab 02 Nov '10

02 Nov '10

Hi - I (accidentally) noticed that SocketStream isn't a subclass of Stream. Is there any reason why this should remain true? It strikes me as odd as Stream has very few requirements (basically #next, #nextPut:, and #atEnd) which are trivially fulfilled by SocketStream. Anyone seeing a reason not to change this? Cheers, - Andreas

3 2

Jump to page: