Squeak-dev September 2002

squeak-dev@lists.squeakfoundation.org

165 participants
288 discussions

FFI on WinCE VM?
by Aaron J Reichow 07 Sep '02

07 Sep '02

Yoshiki and other gurus, How hard would it be to get FFI working on the WinCE VM? I'm particularily interested in the Handheld PC 2000 VM ARM (platform I'm using now) and the PocketPC ARM VMs. Should FFI be easy to get working in WinCE? Regards, Aaron Aaron Reichow :: UMD ACM Pres :: http://www.d.umn.edu/~reic0024/ "corporation, n: an ingenious device for obtaining individual profit without individual responsibility." :: ambrose bierce

3 4

Re: Which mac VM (was RE: String>>hash performance)
by Richard A. O'Keefe 07 Sep '02

07 Sep '02

John M McIntosh <johnmci(a)smalltalkconsulting.com> *very* helpfully wrote: In general the application used must have the largest version number, an internally assigned number that I typically increment for each VM distributed. In case of version number duplication then the application with the latest creation date is used, if duplication then it's usually the one you've last moved within the finder. Ah, os-x might have some different opinions here. Not relevant to me because I'm running MacOS 8.6. Where's the version information? Is this a MacOS thing? Ah, ResEdit tells me that 3.2 has version 3.2.7 Beta 7 00-USA and 3.0Alpha8MT has version 3.0.8 Alpha 8 00-USA Now in the 3.0.x series of VM there was some VM distributed that have version numbers out of sequence, so if you've got lots of them about then it's possible an older VM might launch because of a mistake on my part, although I think if you have a 3.2.x VM this shouldn't be a problem. For example double-clicking on an image on my powerbook launches a beta test VM I compiled yesterday, on another office machine of mine used for testing which contains 37 some VM of various forms (many not distributed to the public) causes 3.2.8b5 to run because that's the latest. To ensure that you get the VM you expect to run you could use Sherlock and find all applications of type FAST IE do a custom search for file type is APPL and creator is 'FAST' I have Date 'vers' Name 2000.01.04 <<missing>> Squeak VM 2.7 PPC 2000.08.28 2.8.0 Final 0 Squeak VM 2.8 2001.02.28 3.0.8 Alpha 8 Squeak 3.0Alpha8MT 2001.06.31 3.1.8 Alpha 0 Squeak 3.0 (the browser plugin) 2002.05.31 3.2.7 Beta 7 Squeak 3.2.7b6 Classic 2002.05.31 3.2.7 Beta 7 Squeak 3.2 VM (a copy in a shared folder) The 'vers' resource doesn't seem to be helping MacOS 8.6 find the most recent version. Then delete the ones you don't want to use. It's not clear to me that I don't want to use any of them. I had imagined that the browser plugin VM was different for a reason; not that I have any idea what the reason might be. If you want to see whether something works in an older Squeak, you _don't_ want to do your testing in a VM that might have some bugs fixed. And of course, older VMs asked for less memory; on a 64MB machine it is surprising just how little you can have running at the same time as a Squeak 3.2 VM. As noted people who collect VMs for a living seem to do the drag and drop invocations just to be sure what they run is what they expect. Yes, but I don't collect VMs for a living, and I do want to fire Squeak 3.0 and Squeak 3.2 up from DragThing. (For later MacOS users, think of the Dock.) I'm rebuilding my desktop file. Gosh, it takes a long time. Nope, still picks 3.0. Changed Creator of 3.0 and plugin, nope, still picks them even though Creator doesn't match any more. Hmmm. Don't burn bridges entirely, drop 3.0 VM(s) onto MacGzip. Well, stap me vitals! Now it picks 2.8. Drop 2.8 and 2.7 onto MacGzip. I don't know how MacOS picks 'em, but it certainly isn't newness... Hokay, I've compressed all but the latest VM, so that I _can_ test with older VMs by uncompress followed by drag-and-drop. Should something about this be in an installation guide somewhere? Should a new release image (or one derived from it) pop up a warning if run on an older VM?

2 1

Re: [Survey] About the cognitive surcharge of super
by Stephane Ducasse 07 Sep '02

07 Sep '02

Hi ian this is fun because squeakers are much better than the oo expert (j..) I know ;). Just a point. The fact that class is not sent is not really good. I would say that ideally it should be bytecoded only when class is not redefined in the class or one of its superclasses. Redefining of class let me implement shadow classes (instance of behavior) acting as metobject (as in CodA) transparently in VW. Without the possibility to redefine class I as well as codA could not have made it in a so nice way. Stef On vendredi, septembre 6, 2002, at 07:15 PM, Ian Piumarta wrote: > Steph, > > On Fri, 6 Sep 2002, Stephane Ducasse wrote: >> >> MyClass>>test1 >> ^ super == self > > The only difference betwen super and self is related to where lookup > starts -- the receiver is identical in both cases. That leaves only > two > possibilities: > > 1) if you defined MyClass to have nil as superclass (or to be a > subclass > of some class that does not have Object in the superclass chain > someplace) then you get message not understood (#== is not found > in lookup). > > 2) if Object is somewhere above MyClass then the answer is true. > >> MyClass>>test2 >> ^ self class == super class > > Self and super are the same object. The result _has_ to be true. (The > same observation about superclasses and #doesNotUnderstand: applies > here > too, but only in the send to super. [The send of #class to self > doesn't > cause a message send at all -- the VM fetches the class directly when > it > sees the special send bytecode. In the send to super, the message > really > is sent and a primitive in Object or ProtoObject or wherever fetches > the > class of the receiver, which is self.]) > > Regards, > > Ian > > Dr. Stéphane DUCASSE (ducasse(a)iam.unibe.ch) http://www.iam.unibe.ch/~ducasse/ "if you knew today was your last day on earth, what would you do different? ... especially if, by doing something different, today might not be your last day on earth" Calvin&Hobbes

2 2

Re: Squeak book !
by Bruce ONeel 07 Sep '02

07 Sep '02

I'll put one vote for a reasonable collection of base modules that lets beginners download one thing and get started. We can argue endlessly over what that list should be :-) cheers bruce goran.hultgren(a)bluefish.se wrote: > Hi all! > > Avi Bryant <avi(a)beta4.com> wrote: > > > Just to add some history, when the chapter was commissioned > > > (er.."requested") Comanche was far less done. It got a lot more done over > > > the period of the writing, but it's still not in the image. > > > > Why isn't it? It's clearly used far more than PWS is, and it includes > > several classes (SocketStream, ConnectionHandler, TimeStamp) that I > > frequently use even in non-web contexts. Including the base Comanche > > classes in the image seems like a big win to me. But maybe that's just > > me. > > I agree. But on the other hand we need to stop thinking about the > "image" and start thinking in the terms of Modules! > > I have always thought that we are splitting up the image into Modules > and the border of the image will then be less clear. Should we have some > form of "blessing" of Modules meaning that they are considered to be > Modules we all together try to maintain and that we, as in "the Squeak > community" consider them to be of some fundamental importance/use? Such > Modules would then typically be included in the "base distribution" of > Squeak. Or we simply stop including Modules in the base and instead make > the base "minimal" and then let the user/developer use SqueakMap to find > what he/she needs. Just like the apt system in Debian. They also have > collections of Modules suitable for different "tasks" - that might also > be an interesting model. > > Anyway, just thinking aloud. > > > Avi > > regards, Göran

4 3

chat apps in squeak.
by Niko Schwarz 06 Sep '02

06 Sep '02

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Hello, I was writing a tiny messanger similar to kxicq, mostly to fool around a bit with encryption. I dont have much time lately, so i was trying to keep the efforts small. My idea was to write a small java server and then access it from squeak to have the client squeaking. That sure would be a reasonable approach, but i got kinda pissed when xml-rpc for java didnt wanna work together with mycgiserver, and now id prefer to find a tiny solution to swap simple information with a public server for zero costs elsewhere in the net. is anyone aware of such a possibility? (this isnt urgent; if there is none im gonna write my own xml-rpc or modify xml-rpc so that the packages work -- its just that im so tired of this approach..) regards, nick - -- Those who express random thoughts to legislative committees are often surprised and appalled to find themselves the instigators of law. -- Mark B. Cohen -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.0.6 (GNU/Linux) Comment: For info see http://www.gnupg.org iD8DBQE9eNsxc9uyCiO8RNQRAmgJAJ9ujJijHRICrWkvfcyFEoTEQeK8MgCgkLRE 6Ton3eSK2XBORjVh2Y3Ydtk= =fudw -----END PGP SIGNATURE-----

4 5

XML-RPC
by marbusse＠t-online.de 06 Sep '02

06 Sep '02

Hi! http://minnow.cc.gatech.edu/squeak/1399 says that Bolot Kerimbaev had done some work implementing a XML-RPC server for squeak. Looking into the latest comanche-changeset, I can't find classes for that... Does anyone have an implementation on his harddisk? Regards, Markus

2 3

Re: Squeak book !
by Richard A. O'Keefe 06 Sep '02

06 Sep '02

Bijan Parsia <bparsia(a)email.unc.edu> wrote: (Just to be clear, I don't at all dispute Richard's claims about them, nor do I take them as harshing on our efforts in making it.) Just to make this clear, I wouldn't bother to make sure the nuBlue was a quick arm's reach away if I thought it was a _bad_ book. It's a window into an exciting project, full of ideas ready to explode in lots of directions. Richard, was PWS simply not working? or was there a problem getting the configuration files? No, it was a documentation problem. Basically, I got totally confused about what files I was supposed to get and what I was supposed to do with them. Here's how it goes. Open a browser, find class PWS, look at the class comment. It starts: PWS is the PluggableWebServer. Documentation for it can be found at http://www.cc.gatech.edu/fac/mark.guzdial/squeak/pws/. Copy the URL (Cmd-V), open Scamper, paste the url in, and we find Pluggable WebServer NOTE Most of the information here is superseded and is better updated at http://pbl.cc.gatech.edu/myswiki The word "NOTE" is extra big and scary. Follow the link, and you find yourself at Swiki Swiki Welcome to the Swiki Swiki Swiki Front Page This is a place to talk about Swiki, also known as CoWeb .... Doesn't sound like PWS documentation. The first link you find for PWS takes you right back to the page that says it is obsolete. Persist a bit more and you arrive at FAQ on PWS and Swiki There is documentation on PWS and Swiki at http://www.cc.gatech.edu/fac/mark.guzdial/squeak/pws/, but it's pretty old now and not well-oriented toward the needs of new Swiki-users. We've >>started<< at Documenting PWS. Until new documentation can be drafted, let's start with an FAQ here. Basically, I got lost in a maze of web pages all telling me they were no good. I downloaded the stuff that unpacks into an initial sort of Swiki folder, but decided that I could wait to find out how to use it. Now, when I am trying to read documentation, I *hate* going clicky-clicky. I want (more, as a bear of very little brain who doesn't get lost in cities but gets lost very very easily in hypertext, I *need*) something I can put on paper, take home, and read sitting in a comfortable armchair without any computer anywhere near me. If there are examples to run on a computer, fine, I can run them _after_ I've understodd what I need to understand. I was interested because I thought it might be a good idea to set up some local Swikis. For example, in some of the papers I teach, students have to work in groups. It sounded as though a Swiki (with only the members of a group authorised to change it, and the lecturer(s) also authorised to read it for marking purposes) might be a good thing to offer.

3 2

Looking for geometry code
by goran.hultgren＠bluefish.se 06 Sep '02

06 Sep '02

I am looking for geometry Smalltalk code for: 1. Calculating the convex hull of 2 simple polygons (non-convex). 2. Combining simple polygons (some form of "shrinkwrapping" creating less waste than a convex hull). 3. Calculating or approximating "no-fit polygons" for non-convex polygons using Minkowski sums etc. If any of the above ring a bell, I am interested! Otherwise we will just build these things from scratch, it doesn't seem too hard and we have found descriptions of some neat algorithms on the net. I just wanted to check if there is code out there already for these kind of things. regards, Göran

1 0

Re: [FIX] IntegerFromBytesFix-sr
by Richard A. O'Keefe 06 Sep '02

06 Sep '02

Stephan Rudlof <sr(a)evolgo.de> pointed out that unpacking SmallInteger maxVal and trying to put the bytes back together with Integer>>byte1:byte2:byte3:byte4: gives you a LargePositiveInteger instead of a SmallInteger, and provided a correction. On looking at #byte1:byte2:byte3:byte4:, I see a documentation problem. The comment doesn't explain whether it is supposed to construct a *signed* 4-byte integer or an *unsigned* 4-byte integer. Could I ask for a change to the comment as well, so that it reads "Answer an integer in the range 0 to 2**32-1, given its base-256 digits (byte4 byte3 byte2 byte1). The result is a SmallInteger if possible, otherwise a LargeInteger. Only the most significant byte needs to be checked to decide which."

2 1

[FIX] IntegerFromBytesFix-sr v2
by Stephan Rudlof 06 Sep '02

06 Sep '02

"Change Set: IntegerFromBytesFix-sr v2 Date: 2 September 2002 Author: Stephan Rudlof Fixes a bug creating LargeIntegers, where they should be SmallIntegers. Probably not detected before, since LargeIntegers are normalized frequently. Try: | val | val _ SmallInteger maxVal. {val. val + 1} collect: [:arg | (Integer byte1: (arg digitAt: 1) byte2: (arg digitAt: 2) byte3: (arg digitAt: 3) byte4: (arg digitAt: 4)) class]. before and after filing in. History: v2 : improved method comment suggested by Richard A. O'Keefe v1 : bug fix" -- Stephan Rudlof (sr(a)evolgo.de) "Genius doesn't work on an assembly line basis. You can't simply say, 'Today I will be brilliant.'" -- Kirk, "The Ultimate Computer", stardate 4731.3

1 0

← Newer
1
...
20
21
22
23
24
25
26
...
29
Older →

Jump to page: