Squeak-dev April 2005

squeak-dev@lists.squeakfoundation.org

158 participants
287 discussions

Re: Database options was (Re: My first Magma experience ...)
by Chris Muller 04 Apr '05

04 Apr '05

> ODMG (and Magma) seem to block application evolution by storing the binary > objects so that class definitions can't change. It will be difficult to use > them in critical applications if this is correct. Just to clarify, Magma now supports live-class evolution and the ability for instances of many different generations of domain classes to exist in the repository simultaneously, and there is no need for big-bang conversions. Of course, this puts more burden on domain code. There is only one definition of the class active in the image, it must be prepared to handle objects of any generation. - Chris

3 2

Re: SystemDictionary>>changes not understood when installing MiniTraits into a clean image
by Mike 04 Apr '05

04 Apr '05

Hi Stephane, I get the following exception when trying to browse TNumerable: Trait(Object)>>doesNotUnderstand: #superclass Receiver: TNumberable Arguments and temporary variables: aMessage: superclass Receiver's instance variables: methodDict: a MethodDictionary(#adaptToCollection:andSend:->a CompiledMethod (7...etc... traitComposition: {} localSelectors: nil users: an IdentitySet(Complex Number) organization: ('converting' adaptToCollection:andSend:) ('mathematical function...etc... name: #TNumberable environment: a SystemDictionary(lots of globals) classTrait: TNumberable classTrait WhiskerWindow>>selectClass: Receiver: a WhiskerWindow(1378) Arguments and temporary variables: aClassOrMetaclass: TNumberable aClass: TNumberable parent: TNumberable parentClassCollection: an OrderedCollection('TNumberable') parentCategoryCollection: nil categorySections: nil str: nil Receiver's instance variables: bounds: 455@25 corner: 1255@825 owner: a PasteUpMorph(1622) [world] submorphs: #(an AlignmentMorph(1694) a TitledPane(1068)('Class Hierarchy' with ...etc... fullBounds: 455@25 corner: 1255@825 color: Color white extension: a MorphExtension (3900) [eventHandler = an EventHandler; mouseEnterS...etc... borderWidth: 2 borderColor: #raised model: a WhiskerBrowser slotName: nil open: false labelString: 'Whisker' stripes: #(a RectangleMorph(2197) a RectangleMorph(3578)) label: a StringMorph(1629)'Whisker' closeBox: an IconicButton(883) collapseBox: an IconicButton(1514) activeOnlyOnTop: true paneMorphs: an OrderedCollection(a TitledPane(418)('Class Categories' with a La...etc... paneRects: nil collapsedFrame: nil fullFrame: 455@25 corner: 1255@825 isCollapsed: false menuBox: an IconicButton(1905) mustNotClose: false labelWidgetAllowance: 91 updatablePanes: #() allowReframeHandles: true labelArea: an AlignmentMorph(1694) expandBox: an IconicButton(2299) background: a RectangleMorph(2905) lastSelectedMethodPane: a TitledPane(216)('' with nil) classCategoryPane: a TitledPane(418)('Class Categories' with a LayoutFrame(left...etc... classHierarchyPane: a TitledPane(1068)('Class Hierarchy' with a LayoutFrame(lef...etc... classPanes: an OrderedCollection() methodPanes: an OrderedCollection() WhiskerBrowser class>>openExistingBrowserOnClass:selector: Receiver: WhiskerBrowser Arguments and temporary variables: aClass: TNumberable aSelector: nil windows: an OrderedCollection(a WhiskerWindow(1378) a SystemWindow<Browser>(323...etc... collapsedWindows: an OrderedCollection(a SystemWindow(1318)) whiskerWindow: a WhiskerWindow(1378) paneToFlash: nil w: a WhiskerWindow(1378) Receiver's instance variables: superclass: CodeHolder methodDict: a MethodDictionary(#canDiscardEdits->a CompiledMethod (1614) #confi...etc... format: 144 traitComposition: nil localSelectors: nil instanceVariables: #('selectedClasses' 'selectedMethodsDict' 'showVersionDiffs'...etc... organization: ('accessing' showVersionDiffs showVersionDiffs:) ('changing state...etc... subclasses: nil name: #WhiskerBrowser classPool: nil sharedPools: nil environment: a SystemDictionary(lots of globals) category: nil WhiskerBrowser class>>openExistingBrowserOnClass:selector: Receiver: WhiskerBrowser Arguments and temporary variables: aClass: TNumberable aSelector: nil windows: an OrderedCollection(a SystemWindow<Browser>(3236) a SMLoader(3867) a ...etc... collapsedWindows: an OrderedCollection(a SystemWindow(1318)) whiskerWindow: nil paneToFlash: nil w: a SystemWindow(1318) Receiver's instance variables: superclass: CodeHolder methodDict: a MethodDictionary(#canDiscardEdits->a CompiledMethod (1614) #confi...etc... format: 144 traitComposition: nil localSelectors: nil instanceVariables: #('selectedClasses' 'selectedMethodsDict' 'showVersionDiffs'...etc... organization: ('accessing' showVersionDiffs showVersionDiffs:) ('changing state...etc... subclasses: nil name: #WhiskerBrowser classPool: nil sharedPools: nil environment: a SystemDictionary(lots of globals) category: nil --- The full stack --- Trait(Object)>>doesNotUnderstand: #superclass ...etc... Take care, Mike. Receive and share. On Mon Apr 04 05:38:47 PDT 2005, =?ISO-8859-1?Q?st=E9phane_ducasse?= <ducasse(a)iam.unibe.ch> wrote: > have you tried because I do not think why traits would interfere. > > Stef > > On 4 avr. 05, at 14:09, Mike wrote: > >> I understand that Traits is much more complete than MiniTraits, >> but I am in love with the whisker browser, and I suspected that >> using MiniTraits will interfere with it much less than Traits (I >> will check if this fear is justified) > > > >

2 1

Problem Installing WriteBarrier
by Daniel Salama 04 Apr '05

04 Apr '05

Hi, I was installing WriteBarrier from SqueakMap Package Loader (v 24) and I got the following error message: "Error occurred during install: can't find EOCD position" Any ideas? Thanks, Daniel

3 2

[ENH] Speed up horizontal scrollbars
by Steven Swerling 04 Apr '05

04 Apr '05

The attached changeset speeds up initialization of scrollpanes containing horizontal scrollbars by caching scroll range values. Compared to turning off horz scrollbars altogether, "Browser openBrowser" is still about 70ms slower, and "Browser fullOnClass: Browser selector: openBrowser" is about 120ms slower. A signifigant speedup nonetheless with a penalty analagous to turning rounded window corners on.

5 5

RE: Smalltalk/JVM
by Mike 04 Apr '05

04 Apr '05

Oh my God yes, this is fantastic. I've been doing java for years now, its my bread and butter. I have been learning smalltalk over the last 4 or 5 months so far I love it and it would be my language of choice for any new projects I am starting. But there are excellent libraries in Java that I miss - Lucene springs to mind - so if you are telling me that I can have the best of both worlds, I would pull your arm off in my haste to get it. And I suspect there are other developers in the same boat. I also suspect that its not a case of "compile and go" i.e. would I be able to compile Lucene java code and call it just like that from smalltalk code? Take care, Mike. Receive and share. On Fri Apr 01 04:16:08 PST 2005, "Welch, Ronald P (US SSA)" <ronald.p.welch(a)baesystems.com> wrote: > I think there is definitely interest in this. Dynamic languages > on the JVM are a hot topic of discussion right now, and to > my way of thinking anyway, Smalltalk is the poster child of > dynamic languages. > > I am curious though. What do you mean by the statement that > development is done in a "surrogate" Smalltalk environment? > > Ron > ----------=-=-=-=-=-=-=-=-========oOo========-=-=-=-=-=-=-=-=---------- > mailto:Ronald.P.Welch@baesystems.com > Phone:(607)770-3701 > BAE SYSTEMS Controls 600 Main St Johnson City, NY > 13790-1888 > ----------=-=-=-=-=-=-=-=-===================-=-=-=-=-=-=-=-=---------- > -----Original Message----- > From: Jason Jones [mailto:jsj@ksc.com] Sent: Thursday, March 31, > 2005 3:23 PM > To: squeak-dev(a)lists.squeakfoundation.org > Subject: Smalltalk/JVM > > > We were looking for feedback on turning Smalltalk/JVM into an > open > source > project. > > Smalltalk/JVM is a Smalltalk Compiler that compiles to Java > bytecodes > and > class files. > > Is this something any of you out there in the programming > community are > interested in? We would love some feedback. > > Mission Software, Inc. has created a Smalltalk compiler for the > JVM. > This > compiler allows Smalltalk to run on any JVM. The compiler > currently > produces > 100% Java class files fully compatible with the Sun JVM > specification. > This > allows Smalltalk and Java code to interact seamlessly and allows > Smalltalk > programs to run anywhere Java runs. > > > a.. Compiler compiles Smalltalk classes into Java .class files > a.. Java code callable from Smalltalk code > a.. Smalltalk code callable from Java code > a.. Java classes subclassing Smalltalk classes > a.. Smalltalk classes subclassing Java classes > a.. Development is done in a "surrogate" Smalltalk environment > a.. Execution in any JVM environment > a.. Support for Smalltalk blocks > a.. Support for JIT Compilers > a.. Ability to write Servlets in Smalltalk > a.. Ability to write Applets in Smalltalk (No plug-in required) > a.. No special VM required > > Please email any questions to me at jsj(a)ksc.com > > -- Jason Jones > Knowledge Systems Corporation > Smalltalk Industry Council > 919.789.8549 x.21 > www.ksc.com > www.missionsoft.com > > > >

4 3

[BUG] unimplemented method called from TransferMorph>>justDroppedInto: event:
by Alain Plantec 04 Apr '05

04 Apr '05

Hy all, Trying to understand Squeak framework for drag and drop : From Squeak 3.7 (#5989) and Squeak 3.8 - #dropAcceptedMorph: from: is not implemented anywhere. - #dropNotifyRecipient: is never send - #dropNotifyRecipient is only send from TransferMorph>>justDroppedInto: event: and TransferMorph>>delete - dropNotifyRecipient TransfertMorph inst var is only writen from TransfertMorph>>dropNotifyRecipient: and read from TransfertMorph>>dropNotifyRecipient justDroppedInto: targetMorph event: anEvent "If only world wants this TransferMorph, treat it as unaccepted (see also >>delete)." super justDroppedInto: targetMorph event: anEvent. self halt. accepted := targetMorph ~= self world. self animationForMoveSuccess: accepted. accepted ifTrue: [self dropNotifyRecipient ifNotNil: [self dropNotifyRecipient dropAcceptedMorph: self from: targetMorph]]. self delete alain

2 1

Re: Database options was (Re: My first Magma experience ...)
by marcelo Cortez 04 Apr '05

04 Apr '05

Hi all Onmibase also supports class evolution. best regards mdc --- goran.krampe(a)bluefish.se wrote: > Trygve Reenskaug <trygver(a)ifi.uio.no> wrote: > > ODMG (and Magma) seem to block application > evolution by storing the binary > > objects so that class definitions can't change. It > will be difficult to use > > them in critical applications if this is correct. > > > > Cheers, > > --Trygve > > This is not true AFAIK. GemStone/S for example > handles multiple class > versions (at the same time) allowing lazy migration > of instances. This > is AFAIK a much more advanced and flexible approach > than RDBs offer. > > And Magma seems to be following GemStone in this > respect: > > http://minnow.cc.gatech.edu/squeak/5602 > > I think most OODBs have some form of schema/class > evolution facilities. > > regards, Göran > > __________________________________________________ Correo Yahoo! Espacio para todos tus mensajes, antivirus y antispam ¡gratis! ¡Abrí tu cuenta ya! - http://correo.yahoo.com.ar

1 0

[UPDATES] 3.9 alpha 5 more
by stéphane ducasse 04 Apr '05

04 Apr '05

hi all Some updates for 3.9alpha to warn up again. :) 6601HexFixes-laza -- Boris Gaertner -- 16 March 2005 with changes set #6217 the method Integer>>hex was deprecated. This change set is an attempt to replace the remaining senders of hex with either storeStringHex or printStringHex" 6602CleaningUtilitiesAgain-sd -- stephane ducasse -- 30 March 2005 slowly we will get there..." 6603CleaningUtilities-sd -- stephane ducasse -- 30 March 2005 Move some utilities close to the class they use, i.e. Scanner" 6604-prefers-sd Disable the preference for horizontal scrollbars that made everything so sluggish in addition to the rest :) 6605-FixeScrollbars-sd -- stéphane ducasse -- 1 April 2005 fixing old instance of scrollbars now there is no popped up debugger anymore. ' Now we can start to work with 3.9. Stef

6 8

Re: SystemDictionary>>changes not understood when installing MiniTraits into a clean image
by Mike 04 Apr '05

04 Apr '05

Thanks Goran. How can I get the minitraits source code so that I can attempt to fix it? I understand that Traits is much more complete than MiniTraits, but I am in love with the whisker browser, and I suspected that using MiniTraits will interfere with it much less than Traits (I will check if this fear is justified) Thanks for the help. Take care, Mike. Receive and share. On Mon Apr 04 03:42:58 PDT 2005, goran.krampe(a)bluefish.se wrote: > Hi Mike! > > Now I have looked at this and read below. > > Mike <me(a)mikehogan.net> wrote: >> Hi all, >> >> I posted a similar question towards the end of last week, but I >> think I made the post too confusing. So here is the question >> again, boiled down to its essence :-) >> >> I installed a clean image of build 5989, booted it up and tried >> to install MiniTraits using the SqueakMap Package Loader. I got >> the message that SystemDictionary>>changes is not understood. >> >> Can any of you please offer some advice on how to proceed? > > The "snotty" reply here :) is that MiniTraits is categorized for > Squeak > 3.4 (it is an old package and Stephen hasn't been Squeaking for a > while > I think) so trying it in Squeak 3.7 is begging for trouble, SM > even > warns you about that, and this time the warning is warranted > since > MiniTraits is "system level changing stuff" :). > > So the problem is simply that MiniTraits itself (you can see this > if you > open up a changesorter and look at the changes it has introduced) > defines/redefines a couple of low level stuff - in this case it > adds > ClassDescription>>basicRemoveSelector: which is called by > forgetDoIts. > And the new code in basicRemoveSelector: sends #changes to > Smalltalk, > and that behaviour has moved. > > Nowadays you instead write "ChangeSet current" (instead of > "Smalltalk > changes"). > > In other words, MiniTraits itself breaks the system during the > installation of itself. ;) > >> Take care, >> Mike. > > regards, Göran > > PS. The "Traits" package is in a much more working state than > MiniTraits. > > >

3 2

SystemDictionary>>changes not understood when installing MiniTraits into a clean image
by Mike 04 Apr '05

04 Apr '05

Hi all, I posted a similar question towards the end of last week, but I think I made the post too confusing. So here is the question again, boiled down to its essence :-) I installed a clean image of build 5989, booted it up and tried to install MiniTraits using the SqueakMap Package Loader. I got the message that SystemDictionary>>changes is not understood. Can any of you please offer some advice on how to proceed? Take care, Mike. Receive and share.

2 1

Jump to page: