Squeak-dev January 2007

squeak-dev@lists.squeakfoundation.org

160 participants
184 discussions

[ANN] New version of SmallWorld
by Russell Allen 07 Jan '07

07 Jan '07

Happy new year everyone! SmallWorld is a version of Tim Budd's Little Smalltalk which he rewrote a few years back to run on the JVM. I was playing around, and have made some small changes to it; mainly to put the image into the jar file - which means that SmallWorld is now a single executable jar file weighing in at only 72kb including vm and image! For those people unfamiliar with Little Smalltalk, it is not a full Smalltalk 80 system but has been simplified in a number of ways. The current image contains about 4000 objects and is missing many of the normal features of a Smalltalk system (like a debugger, for example :) Nevertheless, it's a fun little system to play with and both the image and the vm should be easily understandable. Play with it online at http://russell-allen.com/smallworld/ Best wishes, Russell Russell Allen mail(a)russell-allen.com

2 1

[Q] BerkelyDB Plugin?
by Sungjin Chun 06 Jan '07

06 Jan '07

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Hi, The link in squeak map does not reachable; so I've lost where I can find it. Can anyone has this or let me know where can I find it? Thanks in advance. -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.2.2 (GNU/Linux) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iD8DBQFFn0CZQqspS1+XJHgRAjlfAJ94B1gz6ROO4v70Idho/cumsD4ulwCcCEqp 9nOUr4I8fNyNwEv5gnQ4PT4= =fDF0 -----END PGP SIGNATURE-----

3 2

Webteam report jan 6 2007
by karl 06 Jan '07

06 Jan '07

The webteam list has been quite relaxed lately. Squeak.org got news feeds (news.squeak.org)on the front page. (by Brad Fuller) I went trough squeak.org and changed all links to new swiki address. Next to do is add links to the new tutorial stuff by the doc team. Happy new Squeak year. Karl

1 0

ODBC in squeak
by Frank Urbach 05 Jan '07

05 Jan '07

Hi all, Is there somebody who use the ODBC-package from Diego? Is there any chance handle errors in the application using this. My problem is that all errors are running into the debugger. For me it isn't a problem. For an application used by "non-sqeakers" this is to much. Subclassing and inherit the error-handling messages wasn't successfully. Any pointers are very appreciated. Thanks in advance, Frank

1 0

Monticello Extension
by Noury Bouraqadi 05 Jan '07

05 Jan '07

Hi, I extended classes definitions and wanted to have these extensions taken into account into monticello. So, I extended MCClassDefinition in my package. Things work fine until I switched to a fresh image and wanted to install my package. I found out that there is a bootstrapping problem, I need my MC extension installed before actually loading my package. Putting the extension into a required package doesn't help. My solution is to use an "installer", i.e. a class that loads my package. However, I have some questions: -I guess other people face this problem. I'm sure that Traits have this problem. You can't install the traits package in an image that doesn't already have MCTraitDefinition... Any other examples? -What are the other available solutions ? I guess, there should be some tool similar to what I called "installer". -Is this issue is or will be addressed in Monticello 2? -Is this problem is really specific to Monticello and Squeak? Anybody knows about similar issue in other Smalltalk dialects or even other languages? Noury ------------------------------------------------------------------ Dr. Noury Bouraqadi - Enseignant/Chercheur ARMINES - Ecole des Mines de Douai - Dept. I.A. http://csl.ensm-douai.fr/noury European Smalltalk Users Group Board http://www.esug.org Squeak: a Free Smalltalk http://www.squeak.org ------------------------------------------------------------------

2 1

debugging similar algorithms
by Ralph Boland 05 Jan '07

05 Jan '07

I have implemented a tool for helping me debug some algorithms that are very similar but I need some help in improving how it works. I have a set of implementations of algorithms that all compute the same result. I normally run two implementations (lets say C and E) at a time and compare the runs for speed, space used, and other characteristics. If the implementations are correct both runs return the same result but if there is a bug the results may be different. If the results are different I track down the problem by running a method called traceBug. (Tracebug is started by pressing a button). The high level part of each implementation is the same but they each invoke a method which is different for each implementation. Let's say C invokes Mc and E invokes Me where Mc and Me are different. Note that Mc (if it has no bugs) always computes the same result for the same input parameters. Similarly for Me. Conceptually, TraceBug works as follows. Repeat If C has not been started start it. Continue running C until Mc is invoked. Store the result of Mc, in Ac. If E has not been started start it. Continue running E until Me is invoked. Store the result of Me in Ae. Until Ac and Ae are different. // since there is a bug this must happen Ac and Ae are arrays. Since they are different, for some index i, Ac[i] != Ae[i]. TraceBug now invokes M'c where is M'c takes the same parameters as Mc except for one additional parameter i. M'c does the same computations as Mc except that when the value Ac[i] is about to be computed a halt is invoked thus bringing up the debugger. I can now use the debugger to investigate how Ac[i] is computed! I also invoke traceBug so that the halt occurs when Ae[i] is about to be computed. Thus by running traceBug twice I am able to bring up two debuggers somewhere near where my bug is! Having two debuggers up at the same time is very useful in finding where my bug is! I would like to be able to improve traceBug in two ways. First, I would like to only press the traceBug button once. This means that once the debugger is brought up traceBug needs to be restarted so that it can cause the second debugger to come up. Obviously this can be done but I don't know how to do it. Second, instead of restarting traceBug from the beginning, I would like to have it be restarted at the point immediately after the line where M'c is invoked. It could then immediately invoke M'e. This would avoid running C and E twice and thus be faster. This is not that important to me since C and E are usually fast (a few seconds), but I would like to know how to do this and I may deal with slower algorithms in the future. Thanks in advance for any advice. Ralph

3 2

New Smalltalk Job offer
by Noury Bouraqadi 05 Jan '07

05 Jan '07

Hi everybody, I just added to my web page a new Smalltalk job offer in France http://csl.ensm-douai.fr/noury/SmalltalkJobAndInternshipOffers Thanks for those who told me that they did found a job using my list. It's always good to learn that some effort is useful. Best wishes for the new year, Noury ------------------------------------------------------------------ Dr. Noury Bouraqadi - Enseignant/Chercheur ARMINES - Ecole des Mines de Douai - Dept. I.A. http://csl.ensm-douai.fr/noury European Smalltalk Users Group Board http://www.esug.org Squeak: a Free Smalltalk http://www.squeak.org ------------------------------------------------------------------

1 0

Re: relational for what?
by Stephan Eggermont 05 Jan '07

05 Jan '07

>>What, really? There are many possible reasons that Google don't use an >>RDBMS to index the web: stupidity, arrogance, excessive cost of an RDBMS, >>sound engineering decisions, or a combination of these. >I'm not saying Google are idiots. Clearly not. I was basically just >questioning using them as some sort of counter point against RDBM systems. >I think you could do the same thing they did with a RDBMS, but not on a >bunch of low end computers. You would have to spend some cash. If you take a look at the access patterns of google you should be able to see that if you wanted to eliminate a lot of cpu's, you'd need an exchange rate of 1 cpu for 100 disks or so. That's not cost-effective at all. Stephan Eggermont

1 0

What is the extent of a truncated rectangle? (reply to milan from heading = forwardDir... thread)
by Jerome Peace 05 Jan '07

05 Jan '07

What is the extent of a truncated rectangle? (from heading = forwardDir... thread) was heading = forwardDirection + rotationDegrees (was Re: How can we deal with Message rot?) Hi Milan, Thanks for your interest in this issue. Look at: http://bugs.impara.de/view.php?id=2453 0002453: [Bugs] What is the extent of a truncated rectangle? and for a specific example http://bugs.impara.de/view.php?id=2709 0002709: Sample instances have their leftmost row of pixels clipped off. The latter has a fix for the polygon curve morph sample instance. Which included extending ceiling and floor to the Point and Rectangle classes. I confess I was lazy and snuck these methods into the morphic package instead of negotiating with Andreas and getting the truncation and roundoff methods adopted into the graphic package. That still needs to be done. The current formalities of modularization are easier to work around that work with. I've been coming across varients of this bug since Oct 2004. (Mantis #0000360) There are numerous reports on mantis that touch on this. also search for reports with "render" or for reports with "drama" (as in dramatic demonstration) for reports with "transf" (as in transform) or reports from "wiz" many of my reports deal with these kind of things. HTH. Yours in service, --Jerome Peace Previous message: > >Milan Zimmermann milan.zimmermann at sympatico.ca >Tue Jan 2 14:33:04 UTC 2007 wrote: > >On 2006 December 29 01:27, Jerome Peace wrote: ><<snip>> >> Many of the deep bugs I run into have the >> TransformationMorph decorator as their root. >> Lots of the gribble leaving problems come from a >> mismatch of assumptions between TfMorph ( which >> believes you can translate the origin into the 3rd >> quadrant) while every body else believes that truncate >> is the same as floor because all points are in the >> 1st quadrant. Clipping boxes get calculated wrong and >> are often of by one just when you need them not to be. > >Jerome, > >Could you point out a few concrete examples (class, method) of this behaviour? >I started work on a (exteremely limited) set of tests for etoys and morphic, >and perhaps it would make sense to document what you found in a test; >although documenting something that is (if I understand) considered broken >may not be as valuable, it should be easy to convert the test code if/once >the code under test is refactored. > >Thanks Milan > >> >> >> Results: screen gribbles abound. >> >> >- inEtoys they are not used directly, but through the >> >> Player, and this is >> >> >why your equation does not hold for each individual >> >> morph. I think >> >> >you would have to eliminate TransformMorph >> >> completely, which might >> >> >complicate other Morphs - until now, that "ugly" >> >> transformation stuff >> >> >was factored out. >> >> As I look at the code I see constant attempts to worry >> about which morph is actually being talked to renderer >> or renderee. The code practically begs to be >> refactored. Its just a big job at the moment. >> I await enlightenment and inspiration. Meanwhile I am >> trying refactorings that lay the ground work for the >> simplification of the Tf mess. >> >> Yours in service, --Jerome Peace > __________________________________________________ Do You Yahoo!? Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.com

2 1

The Weekly Juan #8: "Some books I’ve been reading"
by Juan Vuletich 05 Jan '07

05 Jan '07

Hi Folks, The Weekly Juan #8 is out at http://www.jvuletich.org/issues/Issue0008.htm . This week I talk a bit about several books I've been reading lately. Hope you like it. Any comment is welcome. Cheers, Juan Vuletich

1 0

Jump to page: