Squeak-dev June 1999

squeak-dev@lists.squeakfoundation.org

139 participants
408 discussions

RE: Hacking Squeak With Apple MPW
by agree＠carltonfields.com 07 Jun '99

07 Jun '99

> One thing that I found with MrC (version 4.1.0a2c1) is that > at the highest > optimization levels, it produces incorrect code for BitBlt, > which is why I > don't use those opt levels. I was likewise unable to get a working executable under the pre-release version 4 compilers. On the other hand, the version 3 compilers work just fine and seem to generate superior performing code. > Andrew, have you found out how to apply the VM patch for > MPW-generated VMs? > MrC generates the same unnecessary bounds check for the > bytecode switch, only > using LR instead of CTR for the branch. Throwing that out > might give you some > of the bytecode level performance you're missing. I've > included my version of > the patch method in case you need it. I found that the existing patch code seems to do the trick with the version 3 optimized code, simply replacing "6" with "5". A quick glance at Hans' code seems to indicate he did the same thing. As I'm no PowerPC wizard, I'd be obliged if Hans would check my work and see if I'm missing something.

1 0

Re: Squeak sighting in Dr Dobbs
by Reinier van Loon 07 Jun '99

07 Jun '99

>Unfortunately, I can't find X509.st, I looked at the ftp site >(ftp.ddj.com) but could only find X509.txt). Anybody (Paul?) >know where it is? > It is in X509.zip. Reinier.

2 1

Problems with sound
by prb＠thinkage.on.ca 07 Jun '99

07 Jun '99

I was experimenting with playing sampled sounds and GSM compression and I've run across a problem. I have a GSM compressed file about 45 minutes in length. It had been recorded and compressed completely outside of squeak. I wanted to see if I could play it within squeak. I did the following in a workspace with squeak 2.4c. --------------------------------------------------------------------------- gsmFile := (FileStream readOnlyFileNamed: '/home/prb/scifri2.gsm') binary. gsmChannels := Array new: 1. gsmChannels at: 1 put: gsmFile contentsOfEntireFile. gsmFile close. gsm := CompressedSoundData new codecName: #GSMCodec; soundClassName: #SampledSound. gsm channels: gsmChannels; samplingRate: 8000; firstSample: 1; loopEnd: (gsmChannels at: 1) size; loopLength: 0.0; perceivedPitch: 100.0; gain: 1.0. Time millisecondsToRun: [uncompressed := gsm asSound]. SoundPlayer startPlayerProcessBufferSize: 1000 rate: 8000 stereo: false; stopReverb; playSound: uncompressed. --------------------------------------------------------------------------- Everything up to the playback worked exactly as I expected. The actual playback failed quite early with a segmentation fault as follows --------------------------------------------------------------------------- ttyp5[9]% squeak -memory 60m /dev/dsp: Operation not supported by device /dev/dsp: Operation not supported by device Segmentation fault 1131731908 SmallInteger>doesNotUnderstand: 1131731688 AbstractSound>playSampleCount:into:startingAt: 1131732532 [] in SoundPlayer class>playLoop 1131730908 OrderedCollection>do: 1131731376 [] in SoundPlayer class>playLoop 1131730376 Semaphore>critical: 1131543036 SoundPlayer class>playLoop 1131547244 [] in SoundPlayer class>startPlayerProcessBufferSize:rate:stereo: 1131547400 [] in BlockContext>newProcess --------------------------------------------------------------------------- The '/dev/dsp' problems are an unrelated problem I've not yet investigated. I'm running on a linux machine that's using the ALSA sound drivers rather than the OSS sound drivers. ALSA is 'supposed' to be 100% compatible with OSS, but it would seem that it is not. These errors always occur when the squeak image attempts to make it's first sound. I then duplicated the problem on a Win95 OSR2 system. The crash.dmp file I got back has an identical traceback. It also says "Current byte code: 240" and "Primitive index: 0". The above suggests one or more of the following: 1) I've pushed things further than I should have tried. A 45MByte 16 bit sample is just too big to handle in such a silly way. 2) I've not properly done what I was wanting missing some subtle necessity. 3) A bug in squeak. If it's not my bug I'll happily try to debug it. I'll probably need a hint I'm not sure how to approach an exploding VM problem. GDB I assume... - paul

2 1

Re: Some questions about Morph & Squeak
by Lex Spoon 07 Jun '99

07 Jun '99

"Giovanni Giorgi" <giovanni.giorgi(a)mlab.disco.unimi.it> wrote: > Excuse me, > I'd like to know if there is a tutorial on the Morph GUI. > I have some difficult to use the SystemWindow class. > I try to use it, but I have a little problem. > I open a SystemWindow, fill it with my buttons, and lik it to my > class-model. > But when I try to minimize it, the Window disapper!! > I suppose i am doing something wron. > Exactly, what is the right way to use SystemWindow(s)? Look at FileList or Scamper for fairly simply examples of SystemWindow usage. In Morphic, views and controllers tend to get combined, so its more like model/view than model/view/controller. > > The second question is very simple: Squeak code blocks have temp. > variables? No. Although there has been at least one patch posted to allow this. You can use temporary variables of the method to stand in for them, though. Lex

1 0

Squeak Online Book with Java Tree
by Torsten.Bergmann＠phaidros.com 07 Jun '99

07 Jun '99

Hi, the Squeak Online Book at http://www.phaidros.com/DIGITALIS/index.htm now have a Java tree for navigation. It also includes the Squeak Alice description with some images. (see chapter Squeak and 3D) Bye Torsten (alias Cronos)

1 0

Re: [fixed ENH] 3D Look for Morphic Windows
by Peter Smet 07 Jun '99

07 Jun '99

This is what I love about the internet. I posted my comment just before I left work tonight. I get home, and the things been fixed already. Great. I should also have added that despite the walkback everything seemed to work perfectly. Keep up the good work, Peter >On Mon, 7 Jun 1999, Peter Smet wrote: > >> Bert, this is excellent! >> I noticed that in 2.4b that your change set filed in with >> no problems when I was in a morphic world, but >> got the following walkback in the default MVC window: >> >> UndefinedObject(Object)>>doesNotUnderstand: >> SystemWindow class>>windowsIn:satisfying: >> Preferences class>>noteThatFlag:justChangedTo: >> Preferences class>>setPreference:toValue: >> Preferences class>>disable: >> UndefinedObject>>DoIt >> Compiler>>evaluate:in:to:notifying:ifFail: >> Compiler class>>evaluate:for:notifying:logged: >> Compiler class>>evaluate:for:logged: >> Compiler class>>evaluate:logged: > >Oops! Sorry. I didn't check preferences toggling in MVC. In Preferences >class>>noteThatFlag:justChangedTo: World must be replaced with >"self currentWorld". I'll just repost the changeset. > >Thanks for beta testing :-) >/bert > >-- > Bert Freudenberg Department of > Simulation and > Computer Graphics > http://isgwww.cs.uni-magdeburg.de/isg/bert.html Univ. of Magdeburg >

1 0

Re: [ENH(?)] 3D Look for Morphic Windows - error
by Peter Smet 07 Jun '99

07 Jun '99

Bert, this is excellent! I noticed that in 2.4b that your change set filed in with no problems when I was in a morphic world, but got the following walkback in the default MVC window: UndefinedObject(Object)>>doesNotUnderstand: SystemWindow class>>windowsIn:satisfying: Preferences class>>noteThatFlag:justChangedTo: Preferences class>>setPreference:toValue: Preferences class>>disable: UndefinedObject>>DoIt Compiler>>evaluate:in:to:notifying:ifFail: Compiler class>>evaluate:for:notifying:logged: Compiler class>>evaluate:for:logged: Compiler class>>evaluate:logged: It appears to be due to #submorphs being sent to an undefined object (aWorld) windowsIn: aWorld satisfying: windowBlock ^ aWorld submorphs select: [:m | (m isKindOf: SystemWindow) and: [windowBlock value: m]] Peter >Hi all Squeak color haters ;-) > >I personally got very used to Squeaks coloring scheme, and I don't >experience hurting eyes as some of you reported ... But still, the colors >seem to be quite an obstacle for beginners. I heared that complaints for >years (okay, not *that* many years). Anyway, I wondered how hard it would >be to get a "conventional" gray 3d look - it wasn't (well, at least for >Morphic). Now here it comes: > >"Change Set: 3dlook-bf >Date: 21 May 1999 >Author: Bert Freudenberg > >Adds a 3D look to Morphic windows. Enable uniformWindowColors in >preferences to try." > >It basically works by setting all panes borderColor to #raised. The rest >is preferences handling, some cosmetic stuff (one pixel wide borders >didn't look good with the default hilite/shadow colors), and some extra >code for handling a browser's switches (these seem to me like a rather >crude hack that doesn't really fit into the SystemWindow/ScrollPane >scheme). > >Enjoy, >/bert > >-- > Bert Freudenberg Department of > Simulation and > Computer Graphics > http://isgwww.cs.uni-magdeburg.de/isg/bert.html Univ. of Magdeburg > >

2 1

[BUG] Float atan:
by Alexander Lazarevic 07 Jun '99

07 Jun '99

Just a little bug I found: The atan: method in Float should return values between -Pi and Pi. It returns false values for the third quadrant, e.g. "-1 atan: -1" returns 3.92699081698724 (225°), which should be -2.356194490192345 (-135°). Here is a patch: Content-Type: application/octet-stream Content-Disposition: attachment; filename="Float-arcTan.st" Attachment converted: Anon:Float-arcTan.st (????/----) (0000AAC2) Alex. -- ------ Alexander Lazarevic ------------------------------------------------ -- Otto-von-Guericke-Universitaet mailto: lazarevi(a)cs.uni-magdeburg.de -- -- Computational Visualistics http://www.cs.uni-magdeburg.de/~lazarevi --

1 0

[ENH(?)] 3D Look for Morphic Windows
by Bert Freudenberg 07 Jun '99

07 Jun '99

Hi all Squeak color haters ;-) I personally got very used to Squeaks coloring scheme, and I don't experience hurting eyes as some of you reported ... But still, the colors seem to be quite an obstacle for beginners. I heared that complaints for years (okay, not *that* many years). Anyway, I wondered how hard it would be to get a "conventional" gray 3d look - it wasn't (well, at least for Morphic). Now here it comes: "Change Set: 3dlook-bf Date: 21 May 1999 Author: Bert Freudenberg Adds a 3D look to Morphic windows. Enable uniformWindowColors in preferences to try." It basically works by setting all panes borderColor to #raised. The rest is preferences handling, some cosmetic stuff (one pixel wide borders didn't look good with the default hilite/shadow colors), and some extra code for handling a browser's switches (these seem to me like a rather crude hack that doesn't really fit into the SystemWindow/ScrollPane scheme). Enjoy, /bert -- Bert Freudenberg Department of Simulation and Computer Graphics http://isgwww.cs.uni-magdeburg.de/isg/bert.html Univ. of Magdeburg Content-Type: TEXT/PLAIN; charset=US-ASCII; name="3dlook-bf.cs" Content-ID: <Pine.LNX.3.96.990607085307.14323C(a)balloon.cs.uni-magdeburg.de> Content-Description: 'From Squeak 2.3 of January 14, 1999 on 7 June 1999 at 7:34:13 am'! "Change Set: 3dlook-bf Date: 21 May 1999 Author: Bert Freudenberg Adds a 3D look to Morphic windows. Enable uniformWindowColors in preferences to try."! !Object methodsFor: 'user interface' stamp: 'bf 6/7/1999 07:30'! effectiveBackgroundColor "Take Preference #uniformWindowColors into account when answering my default background color" ^Color colorFrom: (Preferences uniformWindowColors ifTrue: [Preferences uniformWindowBackground] ifFalse: [self defaultBackgroundColor])! ! !BorderedMorph methodsFor: 'drawing' stamp: 'bf 5/21/1999 14:17'! drawOn: aCanvas "Draw a rectangle with a solid, inset, or raised border. Note: the raised border color is generated from the receiver's own color, while the inset border color is generated from the color of its owner. This behavior is visually more consistent. Thanks to Hans-Martin Mosner." | insetColor | borderWidth = 0 ifTrue: [ "no border" aCanvas fillRectangle: bounds color: color. ^ self]. borderColor == #raised ifTrue: [ ^ aCanvas frameAndFillRectangle: bounds fillColor: color borderWidth: borderWidth topLeftColor: (color muchLighterIf: borderWidth = 1) bottomRightColor: (color muchDarkerIf: borderWidth = 1)]. borderColor == #inset ifTrue: [ insetColor _ owner colorForInsets. ^ aCanvas frameAndFillRectangle: bounds fillColor: color borderWidth: borderWidth topLeftColor: (insetColor muchDarkerIf: borderWidth = 1) bottomRightColor: (insetColor muchLighterIf: borderWidth = 1)]. "solid color border" aCanvas frameAndFillRectangle: bounds fillColor: color borderWidth: borderWidth borderColor: borderColor.! ! !Browser methodsFor: 'initialize-release' stamp: 'bf 5/21/1999 14:23'! buildMorphicSwitches | instanceSwitch commentSwitch classSwitch row aColor | instanceSwitch _ PluggableButtonMorph on: self getState: #instanceMessagesIndicated action: #indicateInstanceMessages. instanceSwitch label: 'instance'; askBeforeChanging: true. commentSwitch _ PluggableButtonMorph on: self getState: #classCommentIndicated action: #plusButtonHit. commentSwitch label: '?' asText allBold; askBeforeChanging: true; setBalloonText: 'class comment'. classSwitch _ PluggableButtonMorph on: self getState: #classMessagesIndicated action: #indicateClassMessages. classSwitch label: 'class'; askBeforeChanging: true. row _ AlignmentMorph newRow hResizing: #spaceFill; vResizing: #spaceFill; inset: 0; borderColor: Color transparent; addMorphBack: instanceSwitch; addMorphBack: commentSwitch; addMorphBack: classSwitch. aColor _ self effectiveBackgroundColor. row submorphs do: [:m | m color: aColor; onColor: aColor darker offColor: aColor; borderColor: Preferences borderColorForPanes]. ^ row ! ! !Color methodsFor: 'transformations' stamp: 'bf 5/21/1999 14:05'! muchDarker ^ self * 0.233 ! ! !Color methodsFor: 'transformations' stamp: 'bf 5/21/1999 14:16'! muchDarkerIf: aBoolean ^ aBoolean ifTrue: [self muchDarker] ifFalse: [self darker] ! ! !Color methodsFor: 'transformations' stamp: 'bf 5/21/1999 14:16'! muchLighterIf: aBoolean ^ aBoolean ifTrue: [self muchLighter] ifFalse: [self lighter] ! ! !Preferences class methodsFor: 'parameters' stamp: 'bf 5/21/1999 14:22'! borderColorForPanes ^ Preferences uniformWindowColors ifTrue: [#raised] ifFalse: [Color black]! ! !Preferences class methodsFor: 'parameters' stamp: 'bf 5/21/1999 16:14'! uniformWindowBackground ^ Color veryVeryLightGray! ! !Preferences class methodsFor: 'preferences dictionary' stamp: 'bf 5/21/1999 16:02'! noteThatFlag: prefSymbol justChangedTo: aBoolean "Provides a hook so that a user's toggling of a preference might precipitate some immediate action" | keep | (prefSymbol == #useGlobalFlaps) ifTrue: [aBoolean ifFalse: "Turning off use of flaps" [keep _ self confirm: 'Do you want to preserve the existing global flaps for future use?'. Utilities globalFlapTabsIfAny do: [:aFlapTab | Utilities removeFlapTab: aFlapTab keepInList: keep. aFlapTab isInWorld ifTrue: [self halt]]. keep ifFalse: [Utilities clobberFlapTabList]] ifTrue: "Turning on use of flaps" [Smalltalk isMorphic ifTrue: [self currentWorld addGlobalFlaps]]]. (prefSymbol endsWith: 'WindowColors') ifTrue: [(SystemWindow windowsIn: World satisfying: [:w | true]) do: [:w | w updatePaneColors]]! ! !ScriptEditorMorph methodsFor: 'textually-coded scripts' stamp: 'bf 5/21/1999 11:43'! addTextualScript: aBrowser "Put a message browser right into the header alignment morph" | window aMorph codePane | submorphs last class == PluggableTextMorph ifTrue: [^ self]. window _ self. aMorph _ PluggableListMorph on: aBrowser list: #messageListSingleton selected: #indexIsOne changeSelected: #indexIsOne: menu: #messageListMenu:shifted:. aMorph bounds: (window topLeft extent: 200@12). "will get moved" window addMorphBack: aMorph. aMorph borderWidth: 1; color: aBrowser effectiveBackgroundColor. codePane _ PluggableTextMorph on: aBrowser text: #contents accept: #contents:notifying: readSelection: #contentsSelection menu: #codePaneMenu:shifted:. "editString ifNotNil: [codePane editString: editString. codePane hasUnacceptedEdits: true]." codePane bounds: (window topLeft extent: 200@120). "will get moved" window addMorphBack: codePane. codePane borderWidth: 1; retractableOrNot; "make it stay" color: (Color perform: aBrowser defaultBackgroundColor). " self addMorph: aMorph. aMorph borderWidth: 1; color: (Color perform: aBrowser defaultBackgroundColor); bounds: "! ! !ScrollPane methodsFor: 'initialization' stamp: 'bf 5/21/1999 14:49'! initialize super initialize. borderWidth _ 2. borderColor _ Preferences borderColorForPanes. retractableScrollBar _ scrollBarOnLeft _ (Preferences valueOfFlag: #inboardScrollbars) not. scrollBar := ScrollBar new model: self slotName: 'scrollBar'. scrollBar borderWidth: 2; borderColor: #inset. retractableScrollBar ifFalse: [self addMorph: scrollBar]. scroller := TransformMorph new color: Color transparent. scroller offset: -3@0. self addMorph: scroller. self on: #mouseEnter send: #mouseEnter: to: self. self on: #mouseLeave send: #mouseLeave: to: self. self extent: 150@120! ! !SystemWindow methodsFor: 'initialization' stamp: 'bf 5/21/1999 14:53'! initialize | aFont | super initialize. isCollapsed _ false. activeOnlyOnTop _ true. paneMorphs _ Array new. paneRects _ Array new. borderColor _ Color black. borderWidth _ 1. color _ Color black. aFont _ Preferences fontForScriptorButtons. stripes _ Array with: (RectangleMorph newBounds: bounds) "see extent:" with: (RectangleMorph newBounds: bounds). self addMorph: (stripes first borderWidth: 1). self addMorph: (stripes second borderWidth: 2). self addMorph: (label _ StringMorph new contents: labelString; font: (TextStyle default fontAt: 2) emphasis: 1). self addMorph: (closeBox _ SimpleButtonMorph new borderWidth: 0; label: 'X' font: aFont; color: Color transparent; actionSelector: #delete; target: self; extent: 16@16). self addMorph: (collapseBox _ SimpleButtonMorph new borderWidth: 0; label: 'O' font: aFont; color: Color transparent; actionSelector: #collapseOrExpand; target: self; extent: 16@16). Preferences noviceMode ifTrue: [closeBox setBalloonText: 'close window'. collapseBox setBalloonText: 'collapse/expand window']. self on: #mouseEnter send: #spawnReframeHandle: to: self. self on: #mouseLeave send: #spawnReframeHandle: to: self. label on: #mouseDown send: #relabelEvent: to: self. self extent: 300@200! ! !SystemWindow methodsFor: 'panes' stamp: 'bf 5/21/1999 11:44'! paneColor Display depth > 2 ifTrue: [^model effectiveBackgroundColor]. ^ Color white! ! !SystemWindow methodsFor: 'panes' stamp: 'bf 5/27/1999 08:11'! updatePaneColors "Useful when changing from monochrome to color display" | col borderCol m | col _ self paneColor. borderCol _ Preferences borderColorForPanes. self setStripeColorsFrom: col. paneMorphs do: [:p | p color: col; borderColor: borderCol]. "This is for the browsers switches - copied from Browser>>buildMorphicSwitches" self flag: #XXX. "Is there a better way to do this?" (model isKindOf: Browser) ifTrue: [ m _ self findA: AlignmentMorph. m ifNotNil: [ m submorphs do: [:b | b color: col; onColor: col darker offColor: col; borderColor: borderCol]]] ! ! "Postscript: Make preference #uniformWindowColors explicit." Preferences disable: #uniformWindowColors!

1 0

Re: Documentation
by Vandendriessche, Frederik 07 Jun '99

07 Jun '99

I would like to contribute in documenting. What I would also suggest is ; not only written text documentation, but also (UML) models on the classes... Am I the only one who finds that models are missing here ? I believe not. Frederik Vandendriessche * ..32/0 2 711 37 83 L ..32/0 2 711 39 47 ---------------------------------------- First they ignore you. Then they laugh at you. Than they fight you. Then you win. -- Gandhi ----------------------------------------

1 0

Jump to page: