Squeak-dev June 1998

squeak-dev@lists.squeakfoundation.org

83 participants
130 discussions

Re: floatarray primitives
by Les Tyrrell 02 Jun '98

02 Jun '98

A good starting point would be to go through the BLAS- these are used quite heavily in numerical work. I would also look a bit at the vector features in FORTRAN 90- shorthands for array sectioning, etc. I also had an interest a few years ago in doing this... no time. But it should be straightforward, and I think this approach is a good way to go. The fundamental thing that was missing was the notion of homogenous arrays of known, primitive types such as float, double, and integer ( to corresponed with comparable abilities in FORTRAN ). The Vector/Matrix primitives would be super to have. I wouldn't expect it to be as fast as FORTRAN, but it ought to be pretty good- I think they still say that C is slower, I've heard about 30% in some cases. So, there is perhaps some leeway to do better than C, or C++ for that matter. So, for what it's worth, I strongly encourage you to do this- I wish I could offer more, but time is very short and I really ought to get those Modules working for everyone. les

1 0

Re: floatarray primitives
by Juan Cires Martinez 02 Jun '98

02 Jun '98

I have in a very low priority background process the project of writing a FloatBlt class, that will do for float arrays what BitBlt does for forms. Maybe we could get some kind of organized effort to do it. I guess the first steps would be to define an interface, then implement it in Squeak, then decide what methods should be primitives, and finally, to write the primitives.

1 0

How can turn off the sound?
by lhooq＠mail.hitel.net 02 Jun '98

02 Jun '98

Hi. Question from a newbie.. In Squeak2.0 (windows) BookMorph, anything I touch results a primitive fail error on SoundPlayer class. I guess my lack of soundcard is making the problem. How can I turn off the sound ? Thanks ! (Changhee Lee) lhooq(a)hitel.net

1 0

Re: floatarray primitives
by Eric King 02 Jun '98

02 Jun '98

:I am looking for floatarray which are implemented in primitives for speed. :I want to use them for intensive vector and matrix type calculations. I'm also interested in some capabilities like this, but I don't think it's very high on the Squeak team's to-do list, and I haven't found any third parties who've attempted this yet. There's also the question as to how much such a library would bloat the Squeak VM. Personally, I think a high-speed & portable math library would be a wonderful addition to Squeak and go a long way towards increasing its performance. I am willing to help out on such a project but how should such a library be structured? Matrix routines seem like a good place to start, but what sorts of matrix routines? You mention a floatarray class, should it be strictly one dimensional for simplicity, or n-dimensional for generality? (I'm a voxel person, so I have an interest in seeing fast support for 3D arrays ;)) -Eric

1 0

get me off your mailing list
by "Fabi·n" Mario Chesta 02 Jun '98

02 Jun '98

somebody else signed me up for your mailing list, i would like to be removed from your mailing list immediatly. remove my address fmsoft(a)rocketmail.com please === Fabian Mario Chesta fmsoft(a)rocketmail.com _________________________________________________________ DO YOU YAHOO!? Get your free @yahoo.com address at http://mail.yahoo.com

1 0

debugger crash
by Aik-Siong Koh 02 Jun '98

02 Jun '98

<x-rich>I am trying to learn how C code is generated so I tried to step through the following code: "self halt. AbstractSound cCodeForSoundPrimitives" with the debugger, but the VM crashed with the following dump: <bigger> Current byte code: 0 Primitive index: 88 Stack dump follows: 19678900 [] in BlockContext>newProcess I am using squeak 2.0 on NT 4.0. </bigger> All the best Aik-Siong Koh </x-rich>

1 0

Latest updates
by Dan Ingalls 02 Jun '98

02 Jun '98

Folks - I have just put out five new updates... 064ServerSortBy-di.cs (1998.05.27 08:36:54 5,516) Supports sorting by date, name or size in remote directories. 065serversInRoot-di.cs (1998.05.27 08:36:21 8,183) Automatically includes all known servers in the root list (what you see when you choose [ ] ). Also fixes 'add server' and adds 'remove server'. Also fixes a bug that prevented one from seeing file names with spaces. Also fixes a bug that prevented looking at directories other than the leaf directory. Lets you freely go back and forth from Net to Disk. 066DebuggerFixes-wod.cs (1998.05.27 09:17:51 3,799) Bill Dargel's debugger fixes from a previous message - many thanks! 067tk-chooseServer-DG.cs (1998.05.27 08:58:34 3,148) Lets you to abort by choosing no server group when you start to load updates. 068tk-saveMorph-BG.cs (1998.05.27 08:58:45 16,079) Allows saving morphs with scripts. The unit of saving is the PasteUpMorph (i.e. Playfield). In the halo of a PasteUpMorph, in its menu, choose "playfield options...", then "save morph in file". It is OK if viewers and scripts are open at the time of saving. But, any script that is not *named* will be inaccessable later. So name you scripts before saving. I (TK) will issue a later update that explicitly asks you to name the scripts. To bring a playfield back, in the World menu, under "file...", choose "read morph(s) from file". It will come in and stick to the cursor. (We need to get a file picker in here, because we all know it is bad interface to ask the user to type an existing name.) Enjoy - Dan

2 2

Re: Newbie Question
by Les Tyrrell 02 Jun '98

02 Jun '98

Oh, it's much worse than that- Link is the superclass of Process. I've done this a few times, porting code from Digitalk applications to VisualWorks. Link is a handy name, but a dangerous one in this case. Oddly enough, I believe it is possible to add an instance variable to Object in VisualWorks- perhaps this is also possible in Squeak. The problem with redefining Link is that the instance variables and class kind ( variable byte, etc ) probably got changed, hosing the variable bindings in the methods for Process. If the instance variables originally needed for Process still were defined, perhaps you would survive that change. But of course the real issue is that you wanted to name a NEW class Link, and that is a Modules & Namespaces issue- we do not have either Modules or Namespaces yet. les

1 0

Re: st browser of Java code
by Bruce Boyer 01 Jun '98

01 Jun '98

At 12:23 PM 6/1/98 -0400, Jason Karney wrote: >I thought it would be nice to have a class browser/development tool for Java >that worked like Squeak's.... and the portability of Squeak would be an >advantage for development. Have you taken a look at the class browser in Parts for Java? Since it was designed by Smalltalkers, maybe it's what you have in mind. You can download a trial version from www.objectshare.com. (Please excuse the shameful advertising.) Bruce ************ Bruce Boyer Lead Technical Writer ObjectShare email: bboyer(a)objectshare.com ************

1 0

Updates 'R' Us
by Dan Ingalls 01 Jun '98

01 Jun '98

Folks - Well, you have heard about automatic code updates in Squeak 2.0. Now is your chance to try them out. If you are connected to the Internet, all you should have to do is: Start with a reasonably clean copy of 2.0, choose 'help...' from the screen menu, and then choose 'update code from server' in the next menu. After a modest delay, you should see a progress bar as each update rolls in, and then a confirmation of how many updates were received (this will depend on when you get around to trying this). You should then save a copy of this image, as it will save you time not to have to repeat this process every time you want to get a clean image. To give you a sense of how this can impact life between changes, I believe that every bug report so far since the release of 2.0 is dealt with in this first set of six changes. 058CSFileIn-di.cs Causes fileIn to recognize and save ChangeSet preambles. 059FileListScrollbarFix-di A proper fix to the erratic FileList scrollbar. 060NewServerOrg-di.cs Moves server globals into class ServerDirectory. 061MorphicPaneMenus-di Enables option menus in morphic browsers. 062DebuggerPcFix-ar.st Fixes pc selection during debugger step. 063aboutSystemFix-tao Makes aboutThisSystem work when you have no updates (which won't be so after the update but, oh, well...;-) The update facility is primarily the work of Ted Kaehler, although it owes much to the underlying socket support built a while back by John Maloney. - Dan

8 10