Squeak-dev August 2000

squeak-dev@lists.squeakfoundation.org

130 participants
418 discussions

Re: [ANN][LINUX] New Debian Packages
by Bruce ONeel 29 Aug '00

29 Aug '00

Hi, ftp://st.cs.uiuc.edu/Smalltalk/Squeak/2.8 is updated with this. cheers bruce Marcus Denker <marcus(a)ira.uka.de> wrote: > Hi! > > The Debian Squeak vm-package has been updated to 2.8pre2 > > ftp://ftp.ira.uka.de/pub/squeak/debian/debian-2.2/2.8/ > > squeak-vm_2.8pre2-1_i386.deb 421 Kb > > > Marcus > > -- > Marcus Denker marcus(a)ira.uka.de phone@home:(0721)6649375 @work:(0721)6082749

1 0

RE: Updates and other SqC projects
by Torsten.Bergmann＠phaidros.com 29 Aug '00

29 Aug '00

Chris Norton wrote: >I recall that Bolot Kerimbaev [bolot(a)cc.gatech.edu] posted a goodie that >provided a mechanism for selecting multiple morphs a year or so ago... >Can't seem to find it on my hard drive, though. You might be inclined to >send him a note on the subject. I think you mean this one. Works fine with 2.9 alpha. Bye Torsten

1 0

[Win32] Slowdowns (was: RE: 2.9a VM oddities...)
by Raab, Andreas 29 Aug '00

29 Aug '00

> > Under Win98, the 2.9a VM seems extremely slow! I finally found the reason for this - there is an ugly side effect of the new event stuff and the attempt of the VM to reduce CPU usage. It's relatively easy to fix but if you experience the slowdown please make sure you have turned off the 'reduce CPU usage' preference (from the F2 menu). Thanks to Kevin and Nathanael for pointing me to the problem! - Andreas

3 2

Newbie question - Constructors
by Mark Greenaway 29 Aug '00

29 Aug '00

I'm using Squeak to try to learn about Smalltalk. At the moment I'm trying to create a Frog class with a collection Frogs stored at class level. I want to somehow override Frog new so that when a Frog instance is created, the new method defined in the Frog class creates the object instance and then adds this instance to the Frogs collection. I've tried new ^super new initialize initialize self class Frogs add: self but the new defined in the Frog class is never called. Can someone please tell me what I'm missing? Is this even the right approach in Smalltalk? Thanks, Mark

2 1

Newbie Morphic Menu Question
by Kevin Fisher 29 Aug '00

29 Aug '00

Sorry for the newbie question...I've done some poking around in various documentation caches on the main Swiki and I've had no luck answering this question... I've been looking at how popup menus are done in Morphic, specifically using Celeste as an example. I was finishing up my enhancements to the CelesteComposition class, where I had added a simple menu to the main PluggableTextMorph (I added cut/copy/paste). For laughs, I decided to add a menu entry for adding attachments (yes, I know there's a button for it at the top!). The problem is, when I activated this menu item, I get 'message not found'. Upon inspection with the debugger I noticed that the #addAttachment message gets sent to the PluggableTextMorph, not the Model 'CelesteComposition'. My question is...how is this different from other popup menus? From what I can tell, popups in the main Celeste window (ie in the Category pane, for example) all call selectors in the main Model (Celeste), not in the list morph the menu is attached to. I got the 'hint' from the comment in Model>>perform:orSendTo: that text editing morphs are treated differently...is this true? If the menu is opened on a text morph, do the messages get sent to the text morph instead of the Model?

1 0

Re: [ICFP Compition] RayTracer with a Function Description Language
by Bert Freudenberg 28 Aug '00

28 Aug '00

On Sun, 27 Aug 2000, Rob Withers wrote: > Good morning, > > I thought I would post a quick overview of the progress made over the > past day. The swiki site is http://callisto.swiki.net. Ned posted a > PPMReaderWriter for output. Andreas has the renderer well in order as > you may imagine. He is looking into the advanced functions like > difference and intersection for solid objects. Bijan created a grammar > for GML which can parse their most complicated example! ;-) Future > directions look like an evaluator, a class heirarchy for the scene > objects and other elements in the environment, including operations. > Rendering and Writing to a file are there. Finally, the compitition > requires stdin as the data feed where they send the gml statementList. > This may mean it would be a good idea to have a headless image to do > this. Anybody else up for a little hacking to get this together? Yes I want! But where's the stuff? I downloaded everything from callisto.swiki.net, but there's no renderer yet, just a parser (how do I call it?) ... Also, http://www.cs.cornell.edu/icfp/task.htm is gone ... Anyone have a copy? -- Bert

2 1

RE: Updates and other SqC projects
by Norton, Chris 28 Aug '00

28 Aug '00

Hi Dan & co. Regarding your comment: "My focus for the next month or so is to get a few morphic hacks (group select, better gridding, etc)..." I recall that Bolot Kerimbaev [bolot(a)cc.gatech.edu] posted a goodie that provided a mechanism for selecting multiple morphs a year or so ago... Can't seem to find it on my hard drive, though. You might be inclined to send him a note on the subject. Cheers, ---==> Chris

1 0

Re: FW: [Qsoul] FW: Bug in class message behaviour
by Mats Nygren 28 Aug '00

28 Aug '00

Johan Fabry <johan.fabry(a)vub.ac.be> wrote: > ObjectOut defines the message #xxxClass, with implementation > "<primitive: 111>", which is identical to #class as defined in Object. > So when ObjectOut needs to get its behaviour, it can use #xxxClass. I had "bad luck" with the details here. > [snip] > > As for debugging transparant proxies, two cases can be made: Do you want > to debug the proxy or the thing the proxy points to? It is clear that > both cases are valid ... So how's the debugger going to know? In my case I think it got the class from one and used instVarAt: in the other. It looked extremely confusing. I didn't investigate it then, knowing more about these things now perhaps I'll get back to it. > I'd go for a standard behaviour of debugging the thing the proxy points > to. After all, we're talking about *transparant* proxies, right? To > debug the proxies you'll have to return to basic stuff, such as > "Transcript show: foo asString". Ugly, I know ... I've been thinking about letting the debugger check if #doesNotUnderstand: is redefined and in that case have some extra options. But more importantly whenever something that is not inheriting from Object is debugged the debugger most be very sly. Perhaps the following is good: ProtoObject "minimal amount of messages" Debuggable(Proto)Object "messages used by debugger (instVarAt: ..) but nothing else" Object > [snip] > > That's the nice thing about meta-level programming: you can shoot > yourself in the foot in *interesting* ways :-) Have tried a few. /Mats > > -- > "You are more than the sum Johan Fabry - Johan.Fabry(a)vub.ac.be > of what you consume. Vrije Universiteit Brussel > Desire is not an occupation." Programming Technology Lab, Room 10F709 > -- KMFDM Pleinlaan 2, 1050 Brussels, Belgium

2 1

Re: IPC and SRP...
by Mats Nygren 28 Aug '00

28 Aug '00

Johan Fabry <johan.fabry(a)vub.ac.be> wrote: > [snip] > Do you know if there is any way we can pass a message to the squeak VM > at startup time, which forces it to start the network code? Check out the startUp:/shutDown machinery, in SystemDictionary. I tried that with PWS, it worked. In your case it will be more difficult I guess, coordinating several images, but more interesting too. This is an interesting thread. /Mats > > -- > "You are more than the sum Johan Fabry - Johan.Fabry(a)vub.ac.be > of what you consume. Vrije Universiteit Brussel > Desire is not an occupation." Programming Technology Lab, Room 10F709 > -- KMFDM Pleinlaan 2, 1050 Brussels, Belgium

2 1

Re: [BUG][FIX] FileOutSlipsFix
by Jesse Welton 28 Aug '00

28 Aug '00

Doug Way wrote: > > Jesse Welton wrote: > > > > Well, that apparently didn't work properly. I guess I'll submit the > > thing the old-fashioned way. > > I noticed when submitting my SubjectPrefixForCS changeset yesterday > (via "mail to list") that the attachment is broken if you use Squeak > 2.8, but works fine with the current 2.9alpha. Hmm, I did send it from 2.9alpha. I thought I was fully updated, but it turns out I'm not. Oops. -Jesse

1 0

← Newer
1
2
3
4
5
6
7
8
9
...
42
Older →

Jump to page: