Squeak-dev November 2006

squeak-dev@lists.squeakfoundation.org

157 participants
223 discussions

suggested server (headless) swiki package for mac os x?
by Daniel Green 27 Nov '06

27 Nov '06

I'd like to setup a server (headless) wiki running on a mac os x machine to allow some students to post their e-toys and so forth. I haven't had good luck trying comswiki (http://minnow.cc.gatech.edu/ swiki/16) as a headless option for the mac squeak VM. The VM maintainer suggests I ask the lists what other swiki options. Any suggestions? Thanks. Best regards, - DanG

1 0

Error while updating Squeak...
by kironsky＠grisoft.cz 26 Nov '06

26 Nov '06

Hi! I tried to update both the Squeak3.9g-7056 and Squeak3.9-RC1-7061 image. In both cases I get an error: "Error: No package found with name beginning with SqueakMap2 loader". Sorry for the eventual double posting, recently I had no time to watch the mailing list. Elod

2 1

Socket connection timeout...
by kironsky＠grisoft.cz 26 Nov '06

26 Nov '06

Hi! I am trying to connect to a server from Squeak 3.8.1#6747. It is the Robocup Soccer Server that by default opens the UDP port 6000 for connections. So I run the server and execute the following script in Workspace: s:=Socket newUDP. s disconnect. s connectToHostNamed: 'localhost' port: 6000. I get an exception telling me: "ConnectionTimedOut: Cannot connect to 127.0.0.1:6000". The strange thing is, that if I run some sample clients for this server, they can all connect, so the port is surely opened (I checked it also with a port scanner). If I do not disconnect the port before connectiong, I get "InvalidSocketStatusException: Socket status must Unconnected before opening a new connection" exception. I checked my firewall settings, but neither adding UDP 6000 port to exceptions, nor disabling my firewall helped. I am running Fedora Core 5 x64 on an AMD Athlon 64 3200+ processor with 1GB RAM. Do you have any idea what can be the problem? I tried SocketStream class also but I get the same exception. Cheers, Elod

2 2

[Q] recursion
by David Faught 26 Nov '06

26 Nov '06

Hi Klaus and Stéphane, Thank you both for your help. It looks like #clone is indeed faster and will work okay for copying a block like this, as there is no reason to worry about instance variables. My recursive turtle thanks you too! Dave Stéphane Rollandin wrote: >there is also #clone, which seems faster >on my computer: > >fact _ [:n | n isZero ifTrue: [1] > ifFalse: [n * (fact copy value: (n - 1))]]. > >[fact copy value: 6] bench > ==> '157928.2143571286 per second.' > >fact _ [:n | n isZero ifTrue: [1] > ifFalse: [n * (fact clone value: (n - 1))]]. > >[fact clone value: 6] bench > ==> '208975.4049190162 per second.' Klaus D. Witzel wrote: >Hi David, try this: > > | spiral | > self vi:1. > self r: 0.0 g: 0.8 b: 0.0. self lw: 1. > spiral := [ :size :angle | > (size < 100) ifTrue: [ > self fo: size. > self tr: angle. > spiral copy value: size + 2 value: angle. > ]. > ]. > spiral copy value: 0 value: 91. > >Indeed, if you don't make a copy of a recursing block, it is (as the error >message says) already being evaluated. But not so the copy. Note that >every #value:value: needs its own copy.

2 2

Re: SoundRecorder
by Dominique Dutoit 26 Nov '06

26 Nov '06

I think I have found the culprit. If you listen carefully to the recorded sound, you will notice that frames are dropped at regular intervals. Since this pattern is quite obvious and the VM doesn't seem to screw the sound data, it should be something happening at the image level. Looking for a clue in a 3.9 alpha version, I found a change made by "dgd" on April 4, 2006: "workaround for OSS emulation on top on ALSA (on Linux environments)" (Delay forMilliseconds: 20) wait. If the last line is commented out, the sound recorder just works as expected. The same code is present in the OLPC and Squeakland images and both work pretty well without this delay. CoreAudio records the audio input in real time but SoundRecorder picks the data once in a while. The end result is that the sound data is stored in a different time frame that the sampling rate and the sound plays faster. Now I am not sure what can be done to have a cross-platform solution, but as far I have seen the problem is not on the Mac VM side.

3 2

Tests are yellow!
by Ralph Johnson 26 Nov '06

26 Nov '06

"Tests are yellow" doesn't sound nearly as nice as "Tests are green", does it? That is exactly what I felt when I got rid of the last failing test for Squeak 3.9 and the bar on the Test Runner was *still* yellow! It says that there are 2 expected failures. However, it doesn't list any methods. I don't know whether this means that it was a failure, but it was supposed to be a failure, or whether it was supposed to be a failure, but wasn't. If everything is OK, why is the Test Runner yellow? If everything is not OK, why doesn't it tell me which tests are indicating a problem? Can you help me either make the Test Runner green or figure out which test is the problem so I can fix it? I have attached two change files for your amusement. They are both for Squeak3.9-finel-7067. The first will, despite its name, make the tests yellow. There will be no tests left to fix, though, so it illustrates what I am complaining about. Hans-Martin Mosner did most of the work to make this happen. Andreas Raab said we could delete the FloatMathPluginTests and Colin Putney said we could delte the ToolBuilderTest. All I did was ask the original question, delete tests people said could be deleted, and move the other failing tests to subclasses in another category, which I deleted. I have also attached this other cateogry, which is called FailingTests. Don't file it in unless you want the Test Runner to be red! -Ralph Johnson

3 2

OLPC
by Pavel Krivanek 26 Nov '06

26 Nov '06

Hi all, I played with the OLPC system image (http://tuttlesvc.teacherhosting.com/wordpress/?p=251) now and my feelings are quite puzzled - first of all, it seems like almost all GUI can be done using Squeak and it's big loss that we aren't able to offer Squeak in state when it can do this job :-( You may see it here http://www.youtube.com/watch?v=DwzCsOFxT-U&mode=related or here http://www.ivr-usability.com/olpc/olpc.html. -- Pavel

4 4

[Q] recursion
by David Faught 26 Nov '06

26 Nov '06

This is really a Smalltalk/Squeak language question, although it may not sound like it at first. There is a limited Squeak programming environment in the MockTurtle turtle graphics package for Croquet, essentially like a WorkSpace where the code is dynamically compiled and executed like a DoIt, and not retained as a method in the turtle class. Since many turtle graphics examples use recursion to do interesting things, I'm wondering if there is a way to do recursion in this kind of environment, maybe using named blocks? Thanks for any input you may have. Simple examples would be much appreciated!

4 4

Re: [News] Google AdWords campaign targeting http://www.squeak.org
by Klaus D. Witzel 26 Nov '06

26 Nov '06

Google's hiring Smalltalk programmers and they run ads on my keywords 8-) Try it yourself: - http://www.google.com/search?ie=UTF-8&oe=UTF-8&hl=en&q=smalltalk&gl=US After half a dozen or so of manual refreshs, you will see their Smalltalk ad: Smalltalk at Google? Dig SmallTalk & object oriented design? Get a Java XP job at Google www.google.com/jobs/ I run my squeak.org campaign now for 36 hours, during which their CPC bid was raised from pennies to bucks, between a factor of 50 and 100. My pockets are not deep enough to compete against Google's own ads, sorry ;-) Nevertheless, running this campaign was quite an experience :) /Klaus On Wed, 22 Nov 2006 20:09:14 +0100, Klaus D. Witzel wrote: > Hi folks, > > I've just started a $$ Google AdWords campaign and would like to receive > feedback when you see www.squeak.org appearing in a Goole ad (wording, ad > position, whatever your impression, please let me know). > > I would appreciate if you DO*NOT*CLICK*THE*AD, because it costs me $$ per > click, but if you cannot resist feel free to do so ;-) > > /Klaus > > _______________________________________________ > News mailing list > News(a)lists.squeakfoundation.org > http://lists.squeakfoundation.org/cgi-bin/mailman/listinfo/news

4 5

MorphicToolBuilderTests
by Ralph Johnson 25 Nov '06

25 Nov '06

I'm trying to make all the tests green in 3.9. Hans-Martin Mosner got a lot cleaned up, and I'm looking at the rest. The one that is giving me problems at the moment is MorphicToolBuilderTests>>testAddAction. This test is actually inherited from ToolBuilderTests. It dies because it sends the #receiver message to a symbol. It pulls an "action" out of an "itemSpec" and thinks that it will be something with a receiver, but it is just a symbol. To me, it looks like a bug in the test, but I don't know much about ToolBuilder. Can somebody who knows a little about ToolBuilder take a look and tell me how to fix this? Should we just delete #testAddAction from ToolBuilderTests? Can we fix the test? -Ralph Johnson

3 2

← Newer
1
2
3
4
5
6
7
...
23
Older →

Jump to page: