Squeak-dev September 2004

squeak-dev@lists.squeakfoundation.org

150 participants
437 discussions

[UPDATES] 2 for Squeak3.7gamma
by dway＠mailcan.com 03 Sep '04

03 Sep '04

I thought fixing condenseSources was a relatively critical fix, so I'm adding it as a final update for 3.7gamma. (Plus a comment-only change to TwoWayScrollPane indicating that it's obsolete.) Well, this fixes *a* problem with condenseSources, anyway... testing it just now with the gamma1 image, I'm seeing a different problem (nil isSeparator) which I don't think I was seeing with earlier images... sigh. Well, that will have to wait for 3.8, which will be a short release anyway. No other critical-sounding bugs have come up for 3.7gamma, so this should be it. I'll release a gamma2 candidate image shortly, which includes these updates plus any appropriate Welcome Window changes from Marcus/Stephane's Full image. - Doug ------------------------------ 5986CondenseSources-bg-dew -- Boris Gaertner and Doug Way -- 24 August 2004 Smalltalk condenseSources runs into an error that is caused by the fact, that the SourceFileVersionString is stored at two places - in SystemDictionary and in SmalltalkImage. This proposal removes SourceFileVersionString from SystemDictionary so that it's only stored in SmalltalkImage. Cleaned up for 3.7gamma. -dew" 5987TwoWayScrollComment-dew -- Doug Way -- 24 August 2004 Questions/problems have come up a few times recently about TwoWayScrollPane, which is now obsolete. This is simply a class comment change which makes it clear that it is obsolete." <This automated message brought to you by the UpdateIncorporationTool.> <See the UpdateIncorporationTool package on SqueakMap for details.>

6 18

[Q][BUG] Why rotated morph change class ?
by Lic. Edgar J. De Cleene 03 Sep '04

03 Sep '04

I doing Squeak teaching to student's of my class. They are about 20 years old and see something OOP and Smalltalk before using Smalltalk Express as part of his curricula. When I show Squeak to them in lab , I took a EllipseMorph from flaps, duplicate several times , changing color via halo and to one of ellipses I do rotate via halo too. I open a Workspace and type: World submorphs do: [ :m| (m isKindOf: EllipseMorph ) ifTrue: [ m delete ]. Guess what: The rotate one don't delete. I know what should be originate a TransformationMorph , so I pick again via halo and raise a inspector. Surprise!! The inspectors says what still is a EllipseMorph. Then I subclass Trash can and intercept dropping whit a self halt. Now I see what morph dropped was TransformationMorph. They says to me what this is "lack of consistency" and " a real object don't should change class by the fact you rotate it" And how I don't have clever answers ,here I asking to list what to say. Edgar

6 7

[BUG][FIX] ProjectViewMorphInWindowFix
by karl.ramberg＠chello.se 03 Sep '04

03 Sep '04

"Change Set: ProjectViewMorphInWindowFix Date: 3 September 2004 Author: Karl Ramberg ImageMorphs gets their borderWidth added to their extent, unlike other morphs. This is a small fix to check whether a ImageMorph is added to a SystemWindow and make sure it does not get a border added."

1 0

[BUG](FIX] double initialization in ParagraphEditor
by Boris Gaertner 03 Sep '04

03 Sep '04

This bug report is for Squeak 3.7 gamma latest update: #5987 (both for the basic image and for the full image) In the class protocol of ParagraphEditor we read * in method new: new "Answer a new instance of me with a null Paragraph to be edited." | aParagraphEditor | aParagraphEditor _ super new. " --> " aParagraphEditor initialize. " < -- " aParagraphEditor changeParagraph: '' asParagraph. ^aParagraphEditor * in method newParagraph: newParagraph: aParagraph "Answer an instance of me with aParagraph as the text to be edited." | aParagraphEditor | aParagraphEditor _ super new. " --> " aParagraphEditor initialize. " <-- " aParagraphEditor changeParagraph: aParagraph. ^aParagraphEditor The statements aParagraphEditor initialize (labelled with comments by me) are not needed any more, as initialization is called when super new is evaluated. The attached change set is a proposal to remove the unneeded second initialization. This is not a serious bug, but I think it it a bug nevertheless. Greetings, Boris

1 0

Images
by Damien Cassou 03 Sep '04

03 Sep '04

Hi, I've tried this code : (HTTPSocket httpShowGif: 'http://www.google.fr/intl/fr_fr/images/logo.gif') display It displays the logo but a bug too : VM: unix - a SmalltalkImage Image: Squeak3.7gamma [latest update: #5985] SecurityManager state: Restricted: false FileAccess: true SocketAccess: true Working Dir /home/quetzalcoatl/squeak-3.7 Trusted Dir /home/quetzalcoatl/squeak-3.7/secure Untrusted Dir /home/quetzalcoatl/squeak-3.7/untrusted HTTPSocket class(Object)>>doesNotUnderstand: #display Receiver: HTTPSocket Arguments and temporary variables: aMessage: display Receiver's instance variables: superclass: OldSimpleClientSocket methodDict: a MethodDictionary(#contentType->a CompiledMethod (492) #contentTyp...etc... format: 150 instanceVariables: #('headerTokens' 'headers' 'responseCode') organization: ('as yet unclassified' contentType contentType: contentsLength: g...etc... subclasses: nil name: #HTTPSocket classPool: a Dictionary(#HTTPBlabEmail->'' #HTTPPort->80 #HTTPProxyCredentials-...etc... sharedPools: nil environment: a SystemDictionary(lots of globals) category: nil UndefinedObject>>DoIt Receiver: nil Arguments and temporary variables: Receiver's instance variables: nil Compiler>>evaluate:in:to:notifying:ifFail:logged: Receiver: a Compiler Arguments and temporary variables: textOrStream: a ReadWriteStream 'Transcript show: ''Début du test'' ; cr. Trans...etc... aContext: nil receiver: nil aRequestor: a TextMorphEditor failBlock: [] in TextMorphEditor(ParagraphEditor)>>evaluateSelection {[FakeClas...etc... logFlag: true methodNode: DoIt ^ (HTTPSocket httpShowGif: 'http://www.google.fr/intl/fr_fr/i...etc... method: a CompiledMethod (3854) value: nil selector: #DoIt Receiver's instance variables: sourceStream: a ReadWriteStream 'Transcript show: ''Début du test'' ; cr. Trans...etc... requestor: a TextMorphEditor class: UndefinedObject context: nil parserClass: Parser cacheDoItNode: true [] in TextMorphEditor(ParagraphEditor)>>evaluateSelection {[rcvr class evaluatorClass new evaluate: self selectionAsStream in: ctxt...]} Arguments and temporary variables: result: nil rcvr: nil ctxt: nil ex: nil --- The full stack --- HTTPSocket class(Object)>>doesNotUnderstand: #display UndefinedObject>>DoIt Compiler>>evaluate:in:to:notifying:ifFail:logged: [] in TextMorphEditor(ParagraphEditor)>>evaluateSelection {[rcvr class evaluatorClass new evaluate: self selectionAsStream in: ctxt...]} - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - BlockContext>>on:do: TextMorphEditor(ParagraphEditor)>>evaluateSelection TextMorphEditor(ParagraphEditor)>>doIt [] in TextMorphEditor(ParagraphEditor)>>doIt: {[self doIt]} TextMorphEditor(Controller)>>terminateAndInitializeAround: TextMorphEditor(ParagraphEditor)>>doIt: TextMorphEditor(ParagraphEditor)>>dispatchOnCharacter:with: TextMorphEditor>>dispatchOnCharacter:with: TextMorphEditor(ParagraphEditor)>>readKeyboard TextMorphEditor>>readKeyboard [] in TextMorphForEditView(TextMorph)>>keyStroke: {[editor readKeyboard]} TextMorphForEditView(TextMorph)>>handleInteraction:fromEvent: TextMorphForEditView>>handleInteraction:fromEvent: TextMorphForEditView(TextMorph)>>keyStroke: TextMorphForEditView>>keyStroke: TextMorphForEditView(TextMorph)>>handleKeystroke: KeyboardEvent>>sentTo: TextMorphForEditView(Morph)>>handleEvent: TextMorphForEditView(Morph)>>handleFocusEvent: [] in HandMorph>>sendFocusEvent:to:clear: {[ActiveHand := self. ActiveEvent := anEvent. result := focusHolder han...]} [] in PasteUpMorph>>becomeActiveDuring: {[aBlock value]} BlockContext>>on:do: PasteUpMorph>>becomeActiveDuring: HandMorph>>sendFocusEvent:to:clear: HandMorph>>sendEvent:focus:clear: HandMorph>>sendKeyboardEvent: HandMorph>>handleEvent: HandMorph>>processEvents [] in WorldState>>doOneCycleNowFor: {[:h | ActiveHand := h. h processEvents. capturingGesture := capturingGest...]} Array(SequenceableCollection)>>do: WorldState>>handsDo: WorldState>>doOneCycleNowFor: WorldState>>doOneCycleFor: PasteUpMorph>>doOneCycle [] in Project class>>spawnNewProcess {[[World doOneCycle. Processor yield. false] whileFalse. nil]} [] in BlockContext>>newProcess {[self value. Processor terminateActive]}

3 3

bar graph problem
by Matthias Wiese 03 Sep '04

03 Sep '04

Hi, I'm trying to make a bar graph. I used the AlignmentMorph class, but I had Problems to show a scaling and to plot titles. Can someone tell me how I can do this? Thanx, Matthias ________________________________________________________________ Verschicken Sie romantische, coole und witzige Bilder per SMS! Jetzt neu bei WEB.DE FreeMail: http://freemail.web.de/?mc=021193

1 0

Bar graph
by Matthias Wiese 03 Sep '04

03 Sep '04

____________________________________________________ Aufnehmen, abschicken, nah sein - So einfach ist WEB.DE Video-Mail: http://freemail.web.de/?mc=021200

1 0

m17n/Squeak Multilingualization on SqueakMap
by Karl Ramberg 03 Sep '04

03 Sep '04

Could the developers of m17n update it's package info on SqueakMap. It's hard to find the latest and greatest wading though the mail archive. Karl

2 1

Condense Changes for 3.7 Full?
by Doug Way 03 Sep '04

03 Sep '04

Marcus & I were just discussing whether we should do a "Smalltalk condenseChanges" for the 3.7 Full release. Normally, we don't do it for point releases (3.4, 3.5, 3.6). You lose the intermediate method version information, except for the latest version, plus the first version in the .sources file. But it makes the .changes file a lot smaller (11MB instead of 20MB). For the reasons I state below, I think it might be reasonable to condenseChanges on the Full image to make it a more reasonable download size. But leave the Basic image alone. Thoughts? - Doug Begin forwarded message: > From: Marcus Denker <denker(a)iam.unibe.ch> > Date: Fri Aug 27, 2004 5:37:46 PM America/Detroit > To: Doug Way <dway(a)mailcan.com> > Subject: Re: Full Image 3.7g for testing > > > Am 27.08.2004 um 23:25 schrieb Doug Way: > >> On Friday, August 27, 2004, at 04:30 AM, Marcus Denker wrote: > >>> ( I did a Smalltalk condenseChanges). >> >> ... I'm on the fence on whether we should do condenseChanges. I >> don't think it's been done in the past anytime during Squeak 2.x or >> 3.x. (except of course when we started over with new .changes at 2.0 >> and 3.0) >> >> When you condense changes you lose all of the intermediate method >> versions, except for the most recent and the one in the .sources >> file. (I believe) This would be inconvenient for Basic image >> development, and it might mess up the ConflictChecker. On the other >> hand, maybe it's okay for the Full image to have condensed changes, >> since active development on the image itself really just happens in >> Basic, not Full. And of course the .changes file becomes much >> smaller. So I guess doing it for Full would be okay with me.

5 4

Maui utility (was: MVC vs. Morphic)
by Chris Muller 03 Sep '04

03 Sep '04

I have been exploring Morphic trying to create a user-interface utility that is not windowy. It presents rectangles on the screen that each represent a single domain object to which you can drag messages of its class right out of the system browsers and drop them on the rectangle. You can then click on the message to run the method. For binary and keyword messages, a little black-hole is present where you drop other morphs as parameters (the object they represent ends up being the parameter, not the Morph itself). I'm calling it "Maui" and I can hardly wait to finish and share it. Learning and working with Morphic, for me, has been "hard fun" (to steal from one of Alans videos). But I'm glad, frankly, that Morphic is not conventional. I think its exceptional, and its basic-building-blocks approach has been much less constraining than what I remember from Motif. Morphic has made it reasonably possible to implement ideas I only imagined with my eyes closed, unconstrained by the bounds and parameters of a consistent, window-based UI framework. Until something better lets me do that, here's to Morphic! - Chris

2 1

Jump to page: