Squeak-dev July 1999

squeak-dev@lists.squeakfoundation.org

139 participants
367 discussions

Re: [ENH] OS-level events
by David Harris 03 Jul '99

03 Jul '99

> Peter William Lount wrote: > > > Hi Peter, > > > > One difference with keyboard and mouse events it is that in the general > > case the "ACTIVE" window consumes them while the other windows ignore them > > unless the event is a one (like a button down event in an inactive window) > > ... > > abstraction and reuse a lot of the code, or is this just wishful thinking > > on > > my part? > > > > Peter > > > > From: Reinier van Loon <R.L.J.M.W.van.Loon(a)inter.nl.net> > > To: squeak(a)cs.uiuc.edu <squeak(a)cs.uiuc.edu> > > Date: Thursday, July 01, 1999 6:06 AM > > Subject: Re: [ENH] OS-level events > > > > >I keep getting the impression that the event systems designed so far are > > >only concerned with keyboard or mouse events. These events are then put in > > >an event queue with some information regarding the event. > > > > > >Why not define an event as <time, object> and event queue as a collection > > >sorted by time? > > >The global event (interrupt) handler detects an event on the queue (tim's > > >work) and gets the rest of the event of the object mentioned in the event. > > > > > >In this way, each object (keyboard, mouse, network, etc.) can have it's > > own > > >buffer. > > >And composite events are detected easily by comparing the time stamps. > > > > > >You could even start new processes to deal with events or have different > > >event handlers wait on the queues of the objects. No 'centralized' event > > >queue anymore. > > > > > >Just an idea. > > > > > >Reinier. > > > > > > A small two-bits: NeWS (Network extensible Windows System -(Java in Display Postscript?) from Sun had an extensive events distribution system. David

1 0

Re: YEOW! ( was: Re: Squeaklets )
by Dwight Hughes 03 Jul '99

03 Jul '99

You are probably seeing the Mac specific VM code in the InterpreterSupportCode class "source files" message category. These methods contain large pieces of C code as strings -- thus you can see them directly in the image. This C code is of course maintained externally in all the other platforms. "Smalltalk discardVMConstruction. Symbol rehash" should get it all out of your image. -- Dwight Les Tyrrell wrote: > > Stephen Pair: > > > To get the demo applet host image, download > > http://www.advantive.com/squeak/applet.image (it's just under 5 meg...no > > need for sources or changes files). > > This looked interesting to me ( I'm a module fan ) so I thought I'd have > a look. As it turns out, my browser doesn't have an obvious option > for downloading a binary directly. So, it expects the above to be > displayable as a page. Well, I figured I'd give it a try, since > I was a bit curious about what the image file would look like. > > Imagine my surprise to find it filled with C code, apparently for the vm! > ( could've been worse... it might have been filled with Java code... ) > > Thinking this was maybe some sort of mistake, although there were some > fairly squirrelly characters in the image as you would expect, I checked > one of my own images and found the same thing. Judging from what I saw, > one gets the impression that a fairly huge chunk of the space in the image > file is taken up by this C code. > > Does anyone have any way of checking this out, and is there a way to > remove this C code from the image to reduce its size should it turn out > to be fairly substantial? ( Perhaps a good use for image segmenting? ) > > les

1 0

Re: Programming with Tuples (was On Reuse) - still reuse
by Les Tyrrell 03 Jul '99

03 Jul '99

Regarding method factoring- It's interesting to note the impression one has of a system that they are trying to repair, vs. the impression they have of the same system when they are merely browsing through it casually. There have been several times when I've called someone over to my screen to show them some munged up collection of gargantuam methods, only to be stymied by the fact that most of the methods are actually quite small. The difference is that the methods I needed to understand in order to do the repair were NOT the small ones, but the very infrequent BIG ones. In my experience, the bigger a method is, the more likely it is to accumulate complaints. I think Terry Raymond's guideline is good- generally, factor things out a fair bit, but don't worry about getting too extreme with it. I actually have adopted a fairly extreme factoring approach for a few years, but could easily back off a notch or two with no ill effects. There's nothing wrong with getting too extreme ( other than the fact that current browsers don't support you well in the task of understanding long chains of messages ) but it doesn't hurt to ask whether the new "sub" methods you create would be reasonably likely to be used by any reasonable clients. Sometimes, even for really huge methods, the answer is no- to this day, I have one in my old grid generation code that bothers me every time I look at it because it is about four pages long with deeply nested blocks. But no reasonable client would ever use any of the submethods I could pull out of it, and more importantly it works. So, let it be... But if it hadn't been working correctly, it would have been refactored in a heartbeat. However, keep in mind that one of the potential clients for a new submethod could be some poor sap trying to wade through your code in order to either understand it or ( more likely ) to repair it. If you have nicely decomposed your methods, this person's effort can be greatly reduced, or perhaps even avoided altogether. Having been that sap many times, I strongly encourage factoring for the purposes of clarity and future malleability. Happy Squeaking! les

1 0

Re: Programming with Tuples (was On Reuse) - still reuse
by Terry Raymond 03 Jul '99

03 Jul '99

On Thu, 1 Jul 1999 15:20:57 -0400, Mark Guzdial wrote: >I've been thinking about reuse, and integration, for a few months now from >the context of maintaining the PWS. It's almost a holiday weekend (for us >in the U.S.), so perhaps I might waste some bandwidth describing some of >these ideas. I doubt that the solutions I've been considering are at all >new, but perhaps someone could kindly point me toward similar thoughts. > >A bunch of people on the PWS list are generating some wonderful additions >to the system. However, these additions often conflict with one another. >The most common conflict is that two additions modify the same method, but >one modifies a couple lines at the top (for example), while the other >modifies a couple of lines at the bottom. > >One answer is, "Better factoring." If that one method were several smaller >methods, then each contributor would only be modifying unit methods. That >may be, but it seems to me that there is no one factoring that would mean >that *all* possible enhancements/additions/modifications to the system >would only touch unitary methods. There's probably a theorem about O-O >systems that makes this case somewhere. [stuff deleted] >There are lots of holes here. For example, how small does that expression >unit have to be to be useful? Do we end up writing assembly code, for >example? But again, it's nearly a holiday weekend, so perhaps you'll >excuse some half-baked discussion. I don't think it is half-baked. I keep going back and forth on the issue myself. >I haven't read much of Bertrand Myer's work, but my guess is that he's >someone who's probably worked out things like this. Any suggestions? Kent beck has written that a method should do only one thing and it should be written at a consistent level of abstraction. I would add to that, that a class should implement only one functional area. On the system that I am working some people have highly factored code, which is easier to reuse. But it is more difficult to understand, you have to look at a lot of trees to figure out what the forest looks like. My current approach is to try to keep method from becoming too factored, 5 - 10 lines seems to be a nice trade-off. Later, if I find I am doing something else similar in another method then I will refactor it. Terry Raymond Crafted Smalltalk (401) 846-6573 http://www.craftedsmalltalk.com

2 2

Squeaklets
by Stephen Pair 03 Jul '99

03 Jul '99

Over the last couple of days, I've played around with enabling "Squeaklets." What I've done is create a Squeaklet host image that will interpret an HTTP url passed to Squeak on the command line as the address of a Squeaklet file. The Squeaklet file is a modified image segment file that constains "resolvers" for out pointers. Currently, this is nothing more that an array of compilable strings that will resolve out pointers to something contained withing the Squeaklet host image (and terminated only with a Character cr) The innefficiency of the compilable strings and the carriage return denoting the boundary between out pointer resolvers and the actual image segment will have to be corrected...but this is just experimentation. I've doctored up ImageSegment a little bit to handle a Squeaklet file. Also, I've created an Applet class that you can send #saveApplet to, and it will take you through a series of steps that allows you to either expand the scope of your applet (by adding to the rootArray), or store a resolver string to resolve an out pointer. I thought I'd go ahead and put a demonstration of this out. Mainly, I'd like to find out if anyone else is exploring this area. The big areas left to be addressed are Squeaklet dependencies (allowing one Squeaklet to depend on another), Squeaklet host image versioning, and cutting the contents of the Squeaklet host image down to something small, but acceptable to everyone. I'd propose that the Squeaklet host image contain the minimum requirements for Morphic, no MVC, networking, multimedia capabilities, the basic Morphic widgets, and of course the base Smalltalk-80 classes...everything else (including things like the class browser, debugger, inspectors and other dev tools) could be provided in their own image segments. This host image should be made to live within any web browsers on any platform. I imagine other people have done some similar things, but it's something I'm very interested in and would like to hear opinions. To get the demo applet host image, download http://www.advantive.com/squeak/applet.image (it's just under 5 meg...no need for sources or changes files). And then start it from a shell (command line) with: "squeak applet.image http://www.advantive.com/SameGameApplet.seg"...this image has only been tested on a 2.4 VM. That should start the applet host image, and immediately try to download the applet segment file from the abover URL. The applet just opens a SameGame morph (admittedly, it's not very exciting), but the morph is actually contained within the applet. The applet is about 33K in size. Keep in mind, the code in that image is very rough. - Stephen

1 0

Newbie squeak question
by Jeff Carr 03 Jul '99

03 Jul '99

Ok I give up --- I am missing something very basic. When in a squeak project, I can FileIn a *.st file. Then what? I can view the class structure, etc --- but how do I build something that executes? The *.st file I have should result in a new browser type, with special functionality.

3 5

[CFP] OOPSLA'99 VM Workshop: Simplicity, Performance andPortability in Virtual Machine Design
by John.Maloney＠disney.com 02 Jul '99

02 Jul '99

Some of you may be interested in the following OOPSLA workshop... ---------------------------------------------------------------- CALL FOR PARTICIPATION OOPSLA'99 WORKSHOP Simplicity, Performance and Portability in Virtual Machine Design Denver, Colorado November 2, 1999 ORGANIZERS John Maloney, Walt Disney Imagineering Antero Taivalsaari, Sun Microsystems Labs Dan Ingalls, Walt Disney Imagineering THEME OF THE WORKSHOP For decades, computer scientists and practitioners have been fascinated by the idea of portable, hardware-independent programs that are capable of running unchanged on a wide variety of machines. In programming languages such as Lisp, Forth, Smalltalk and Java, this independence has been achieved by utilizing a virtual machine (VM): a portable program that implements a virtual processor for that language. Virtual machine design typically involves many tradeoffs. One tradeoff of particular interest to us is the tradeoff between simplicity, which often confers benefits such as portability and reliability, and performance. Historically, the object-oriented language community has needed to prove that object-oriented languages could rival the efficiency of procedural languages, and thus researchers have not always fully explored the benefits of simpler, more portable virtual machine designs. An example of such a performance vs. simplicity tradeoff arises in the choice of a dispatching mechanism for the virtual instruction set. A simple virtual machine might dispatch bytecodes via a case statement, a more complex one might employ some form of threaded code, and a very high-performance one might translate bytecodes into native code dynamically. Each of these alternatives has indirect complexity, portability, and performance costs. While investigating such tradeoffs is one theme of this workshop, we encourage papers about every aspect of virtual machine implementation including: - virtual machine design and implementation * architectural and design issues in building virtual machines for object-oriented languages * the impact of processor/memory architecture on performance * implementation and optimization techniques - experiences in building virtual machines * using object-oriented technology for building virtual machines, * past vs. present: what could we learn from the 1970s and 1980s - current and future directions in virtual machines * embedded virtual machines * server virtual machines * universal (multi-language) virtual machines Since there have been a number of excellent OOSPLA workshops on garbage collection, we would like to gently encourage papers focusing on other aspects of virtual machine design. FORMAT OF THE WORKSHOP Potential workshop participants, either teams or individuals, are invited to submit short papers (2-5 pages) describing their work on virtual machines. Electronic submissions in text form are strongly encouraged. The papers will be reviewed by the organizers, and a subset of them will be selected for brief (10 minute) presentation at the workshop, followed by group discussion. If necessary, attendance to the workshop will be limited to enable fruitful discussion. DEADLINES Submission deadline: August 16, 1999 CONTACT INFORMATION Please send all communication and submissions (preferably in electronic form) to: John Maloney Walt Disney Imagineering P.O. Box 25020 Glendale, CA 91221-5020 E-mail: John.Maloney(a)disney.com Fax: +1 650 428-0843 ----------------------------------------------------------------

2 1

unused message
by Montgomery F. Tidwell 02 Jul '99

02 Jul '99

Howdy, is it possible to turn off the "blah appears to be unused in this method. OK to remove it?" prompt? in a particular method i have defined a temporary and it is used, but every time i save changes, i get this message. \\//_

3 3

RE: On Reuse
by Les Tyrrell 02 Jul '99

02 Jul '99

Correcting my earlier post, Ed Berard's book is titled: "Essays on Object-Oriented Software Engineering", Vol. 1 ( there are no other volumes, btw ) with ISBN 0-13-288895-5 Although it may be out of print, many people can still get it through Interlibrary Loan... Amazon, Barnes and Noble, and Borders are not the only suppliers of books in the world! See your local library for details, especially if it happens to be a university library. Ed's company, The Object Agency, has a web site at www.toa.com. However, these days he's in the website business and appears to not be doing much in the OO world anymore. However, perhaps he still has some copies of the book to sell. les

4 3

Re: unused message
by JArchibald＠aol.com 02 Jul '99

02 Jul '99

On 7/2/99 at 4:36:35 PM EDT, mtidwell(a)mail.com writes: << is it possible to turn off the "blah appears to be unused in this method. OK to remove it?" prompt? in a particular method i have defined a temporary and it is used, but every time i save changes, i get this message. >> What is the text of the message which does this? Jerry Archibald [This may be a duplicate--mailer had a problem with previous version. Sorry.]

1 0

← Newer
1
...
30
31
32
33
34
35
36
37
Older →

Jump to page: