Squeak-dev August 2000

squeak-dev@lists.squeakfoundation.org

130 participants
418 discussions

Basically private to Ian (was Re: [BUG][FIX]Squeak2.8 VM support files broken
by Tim Rowledge 30 Aug '00

30 Aug '00

Ian, this reply direct to you bounced. Again. Why on earth does email to you bounce pretty much every other time? Is it just that the French net is trying to maintain a minimum level of French content, like the tv system? tim Ian Piumarta wrote: > > > - sqPlatformSpecific.h; your version doesn't have the 'offending' > > lines at all; > > Odd. We're talking about <textutils.h> no? I remember moving them to > before the #endif /* macintosh */ a week ago, and sending the Johns a > bug report. Yeah, that's the stuff. Wonder what happened? The 'bad' lines were in sources generated from the update 2359 image, but not from the tarfile I found on your website; I guess that might point to an explanation. I grabbed something like a 2.1Mb tarfile which seemed to be recent. I'm up in the mountains right now (had to dash up here to cope with forest fire problems) so I can't actually check the filename. Oh, actually I just got to your website ok and it was ~piumarta/squeak/unix/Squeak2.8-src.tar of 17 Aug. > > > - platform.exports; your version is missing the XFNs for Unix and Acorn, > > What XFNs for Unix? Unix doesn't have any. (If you're referring to > the setFooPollFunction() abominations that I found in the SF sources, > that particular bug has been summarily excised and fed to the trash > compactor weeks ago. ;-) What did you do to get rid of the need for the socket & sound poll? Are you using threads or summat? Hope it's something I can use on the Acorn! tim PS So, what about this jitter thingy?

2 1

Re: [Feature] another clientserver framework
by Mats Nygren 30 Aug '00

30 Aug '00

"Göran" Hultgren <gohu(a)rocketmail.com> wrote: >[about different communication frameworks] > I will post a summary when I am done. Thanks. Looking forward to it. And if you will describe a little what you are doing with it, so much the better. /Mats

1 0

Re: [BUG] Network data loss.
by Bob Arning 30 Aug '00

30 Aug '00

On Tue, 29 Aug 2000 15:16:07 +0200 Johan Fabry <johan.fabry(a)vub.ac.be> wrote: >> >The bug is that sometimes the client sends data, but never gets a reply. >> >This is because, apparently, the server never recieves the data. Also, >> >clientside, even though socket DataAvailable returns true, when reading >> >data from the socket no data is read. This can be seen by the 'W' >> >printouts on the transcript. >> >> Two things that might be of interest: >> - on the server, is a connection ever made in the cases that fail? Or does it simply fail to receive all or some of the 79 characters? In other words, how far does the server get? > >As far as I can see, no connection is ever registered on the server. It >never gets to the handlePacket: loop where data is read from the socket, >and it does not sit waiting for data to appear on the socket in >handleConnection:. As far as I can see, the server never detects the >socket. Hi Johan, If I understand this correctly, the client thinks he has a connection (he would not have gotten to the W part otherwise), but the server appears not to be aware of the connection. I have seen behavior like this before and my best guess at what was happening in that case was that, prior to the current server being started, another listen on the same port wasn't properly cleaned up, leaving something in the OS accepting connections without passing them on to the current server. Two experiments might shed some light on this: - When the problem is occurring, stop the current server and then see if a client can connect. I would try the connect a number of times in case it is an intermittent thing. - Does rebooting the server machine help? Cheers, Bob

2 1

Re: Mac 2.8 Browser plugin
by Karl Ramberg 30 Aug '00

30 Aug '00

The mac plugin runs fine. :-) The biggest issue is the installation. On the Bert's download page it tells you to put the image in SystemFolder:Preferences:Squeak etc. SystemFolder and Preferences will have a name localized after the language MacOS is using. I have to admit I got confused because I do have a Preferences folder in my Swedish MacOS and that is created by some programs that do not follow the MacOS guidelines and use named path names or something. The folder I should use is called Programinställningar. Well, there is enough room for confusion. Are there any free mac installers that could free us from these silly problems and let us keep the focus on interesting stuff ? Karl

5 5

Re: [BUG][FIX]Squeak2.8 VM support files broken
by Henrik Gedenryd 30 Aug '00

30 Aug '00

Tim Rowledge wrote: > ... from the latest 2.8 image (as far as I can see, ipdate 2359) the sq.h > file still has a broken macro and the sqPlatformSpecific.h has Mac > specific code outside the Mac section. > Oh yeah, the platform exports file was wrong as well. Tim, could you coordinate these fixes with the latest updates for 2.9a? Otherwise, they are in risk of becoming the Neanderthals of 2.8 (another line of descent eradicated them). John McIntosh has made a lot of updates to the Mac sources in the 24xx, some involving sq.h problems, and quite possibly also these fixes. It is unfortunate that the long latency of releasing 2.8 means that a lot of fixes for 2.8, such as these, are now really part of 2.9a. Let's hope these missing fixes in 2.8 are not substantial enough to give it a bad rep. Henrik

1 0

Squeak Projects from Content image. Was RE: Help with Squeak pres entation to DoD!
by Torsten.Bergmann＠phaidros.com 30 Aug '00

30 Aug '00

The best way is to let Squeak speak for itself. You should create a "content image" with little demos. I can help you a little bit with the following demo projects from my own "content image". To run them you need: - a Squeak 2.9alpha image with the latest changes - 1024x768 screen resolution, set Squeak to full screen mode - open a workspace in a new morphic project Evaluate the following lines to download the projects: Project thumbnailFromUrl: 'http://software.freepage.de/cronos/squeak/Projects/SqueakPlatforms.pr' "I think Squeak runs on more platforms than Java" Project thumbnailFromUrl: 'http://software.freepage.de/cronos/squeak/Projects/SqueakBooks.pr'. "Contains a book with a general overview on Squeak" Project thumbnailFromUrl: 'http://software.freepage.de/cronos/squeak/Projects/SqueakFFI.pr' "Windows only - showing FFI (Foreign Function Interface) capabilities" Project thumbnailFromUrl: 'http://software.freepage.de/cronos/squeak/Projects/FinalFrontier.pr'. "The final frontier segment from Andreas Raab with an additional background" Project thumbnailFromUrl: 'http://software.freepage.de/cronos/squeak/Projects/SqueakGames.pr'. "The games you will find inside of Squeak" Project thumbnailFromUrl: 'http://software.freepage.de/cronos/squeak/Projects/SqueakGraphic.pr'. "Shows what you can do with graphics in squeak" Project thumbnailFromUrl: 'http://software.freepage.de/cronos/squeak/Projects/SqueakPooh.pr'. "The Teddy implementation (Squeak Pooh)" Project thumbnailFromUrl: 'http://software.freepage.de/cronos/squeak/Projects/SqueakShockwave.pr' "Squeak Shockwave integration" Project thumbnailFromUrl: 'http://software.freepage.de/cronos/squeak/Projects/SqueakMusic.pr' "Sound's awfull after loading, change the instruments and it will work" Project thumbnailFromUrl: 'http://software.freepage.de/cronos/squeak/Projects/ActiveTextures.pr'. "Shows reactive 3D textures in Squeak" I will provide some more as soon as I have time. The last one seems to have some problems. Everything is for beta testers only !!! See http://software.freepage.de/cronos/squeak/Projects/overview.gif for an project overview. Hopefully Dan is not angry that I directly send them to the list so early. There is a much more you can show: - more 3D (open a B3DSceneExplorerMorph or the bunny sample in PWM-7) - the webbrowser Scamper - the mailtool Celeste - the other Play with me (PWM) projects - the code update capability - Swiki Servers (Check out class PWS) - communication via EToy system (see EToyListenerMorph class comment and EToyChatMorph, ...) You can send morphs from one squeak image to another in 2.9alpha !!! - the character recognizer (use CMD-R in a workspace) - Sound recording (see RecordingControlsMorph) - Sound processing and voice recognition (see SpectrumAnalyzerMorph and PhonemeRecognizerMorph) - Postscript and True Type font support - Morphic Scripting and Macro Recorder (EventRecorderMorph) - the speech engine (evaluate Speaker manWithEditor say: 'With this editor you can change my voice.' ) - Multilanguage Squeak - Facial animation project (Goodie, see Mailinglist last month) - Nebraska (server image with network terminals, check NebraskaServerMorph) - SUnit Framework (from Extreme Programming) - Fabrik components (see PWM-6) You can also show graphic effects like star wars scrolling: | warp src dest | "Star wars like scrolling" warp _ (WarpBlt toForm: Display) cellSize: 1; sourceForm: Display; cellSize: 2; combinationRule: Form over. src _ {-200@-100. -50@100. 50@100. 200@-100}. dest _ 0@0 corner: 200@100. Display restoreAfter: [ [Sensor anyButtonPressed] whileFalse: [warp copyQuad: src + Sensor cursorPoint toRect: dest]] or the world smallest drawing program: | myPen | myPen := Pen new defaultNib: 2; place: Sensor mousePoint. [Sensor waitButton ; yellowButtonPressed] whileFalse: [Sensor redButtonPressed ifTrue: [myPen goto: Sensor mousePoint]]. To test the Internet capabilities evaluate: MIDIFileReader playURLNamed: 'http://squeak.org/Squeak2.0/midi/wtellovr.mid'. or (FlashMorphReader on: (HTTPSocket httpGet: 'http://www.audi.co.uk/flash/intro1.swf' accept:'application/x-shockwave-flash')) processFile startPlaying openInWorld. or HTTPSocket httpShowGif: 'http://squeak.org/Squeak2.0/midi/Squeakers.GIF'. or Project thumbnailFromUrl: 'http://www.squeak.org/Squeak2.0/2.7segments/SqueakEasy.extSeg'. Hope this will help you. Bye Torsten -----Original Message----- From: jchludzinski(a)worldkey.net [mailto:jchludzinski@worldkey.net] Sent: Dienstag, 29. August 2000 23:31 To: Squeak mailing list! Subject: Help with Squeak presentation to DoD! I'm going to be giving a presentation to some DoD folks about the wonders of Squeak and why it ought to be seriously consider as a candidate for development work (for a wargaming project). I'm currently using it for the prototype, so naturally the question arose: Should we continue to use Squeak (Smalltalk in general vs. C++, Java, etc. [the usual suspects])? I would GREATLY appreciate any input! In addition, I've noticed but haven't used either the GUI (Morphic) features or any support for Web/Internet development. Could someone please address these. To date, I've been using Squeak for the sim-engine and models (tanks, planes, etc.). The GUI is some legacy Java code from a previous sim. ---John

1 0

Re: Help with Squeak presentation to DoD!
by Treis, Ken 30 Aug '00

30 Aug '00

jchludzinski(a)worldkey.net wrote: > I'm going to be giving a presentation to some DoD folks about the > wonders of Squeak and why it ought to be seriously consider as a > candidate for development work (for a wargaming project). I'm > currently using it for the prototype, so naturally the question > arose: Should we continue to use Squeak (Smalltalk in general vs. > C++, Java, etc. [the usual suspects])? To me, one of the most compelling reasons for choosing Smalltalk is that you can change it. As much as possible, the major "features" of the language are themselves defined in the language. The class library, code tools, debugger, and windowing system are all written in Smalltalk. Even the syntax of the language is defined in Smalltalk. If you don't like it, or if it doesn't suit your project, you can change it. Use your tools to change your tools. Squeak takes this capability to another level by defining the VM in Smalltalk. That means that even the lowest levels can be changed to suit your needs, if necessary. The "open-source" nature of the VM has helped it has also helped make Squeak run on so many different platforms so seamlessly. > I would GREATLY appreciate any input! In addition, I've noticed > but haven't used either the GUI (Morphic) features or any support > for Web/Internet development. Could someone please address these. That's too bad, because it might make a really fun Morphic app. Drag this plane onto that tank and combat ensues. DnD for the DoD. :) You might also be able to make use of the "stepping" in Morphic for your event time. Each unit moves a certain distance, adjusts course, etc. when stepped. It's also fairly easy for morphs in a world to figure out if they're near (or in contact with) some other morph -- a demo called "Alan's Fanciful Gears" (or something similar, sorry Alan) makes use of these capabilities -- and have them take action because of that. > To date, I've been using Squeak for the sim-engine and models (tanks, > planes, etc.). The GUI is some legacy Java code from a > previous sim. -- Ken Treis / Software Engineer Fravenic Skreiggser Software Collective ktreis(a)keyww.com --- "There is something fascinating about science. One gets such wholesale returns of conjecture out of such a trifling investment of fact." -- Mark Twain

1 0

[BUG][FIX][UNIX] B3DRasterizerPlugin class>>b3dTypesH
by Marcus Denker 29 Aug '00

29 Aug '00

Hi! The following code was added to b3dTypes.h in the Unix 2.7 VM, but has not been added to B3DRasterizerPlugin class>>b3dTypesH. Without this, the colors of Squeak3D on Linux-x86 are wrong. /* 1999/12/4 12:24:25 ikp -- fix for Unix hosts */ #if defined(HAVE_CONFIG_H) # include "sqUnixConfig.h" # if defined(WORDS_BIGENDIAN) # define MSB_FIRST # undef LSB_FIRST # else # undef MSB_FIRST # define LSB_FIRST # endif #endif Marcus

1 0

Re: FW: [Qsoul] FW: Bug in class message behaviour
by schwa＠cc.gatech.edu 29 Aug '00

29 Aug '00

Jon Hylands <Jon(a)huv.com> wrote: > On Mon, 28 Aug 2000 19:14:48, Mats Nygren <nygren(a)sics.se> wrote: > > > I think there is a distinction to be made: > > - some proxies like ObjectOut becomes their real objects. They inherit ProtoObject directly and will turn into their real object when debugged. (Probably before that). > > - others like a decorator will remain a distinct object and never become what they proxy. These would be very difficult to debug unless they handled the debugger messages themselves so should inherit DebuggableProtoObject. > > My feeling is that proxies should be as transparent as much as they > can be. If you're debugging proxies, and it shouldn't happen very > often, you can afford to do a little nastiness. > I'm not sure that I agree. The issue isn't whether you're debugging the proxy, it's whether you're dealing with a proxy or the actual thing. I don't like the idea of the system lying to me about what I'm seeing. I'd suggest something like a split view that shows both the proxy and all it's internal state, as well as the state of the object wrapped by the proxy. The System-Prototypes changeset posted recently creates Inspectors on PrototypeObjects that work in a way similar to this. I guess there's no reason that it couldn't be configurable, though, so if you want proxies to be invisible, be my guest. I do like the idea of DebuggableProtoObject, as I was forced to add the requisite methods to a subclass of ProtoObject that I created for a purpose where it wasn't just a transparent proxy. Joshua > You don't want to have to go through hoops to debug your application, > which is hopefully where most of the bugs are. > > Later, > Jon > > -------------------------------------------------------------- > Jon Hylands Jon(a)huv.com http://www.huv.com/jon > > Project: Micro Seeker (Micro Autonomous Underwater Vehicle) > http://www.huv.com

2 1

Re: FW: [Qsoul] FW: Bug in class message behaviour
by Mats Nygren 29 Aug '00

29 Aug '00

Johan Fabry <johan.fabry(a)vub.ac.be> wrote: > > ProtoObject "minimal amount of messages" > > Debuggable(Proto)Object "messages used by debugger (instVarAt: ..) but nothing else" > > Object > > I dont think I get the point here... What would be the behaviour of the > debugger for ProtoObject subclasses? Forward everything? So when you > need to debug the proxy, you would need to change its class? The point is that if a proxy really needs to be without #instVarAt: then it can inherit ProtoObject and will be difficult to debug. But in many cases, proxies can inherit DebuggableProtoObject instead which implement, by definition, the messages sent by the debugger but nothing else, and that should work. I think there is a distinction to be made: - some proxies like ObjectOut becomes their real objects. They inherit ProtoObject directly and will turn into their real object when debugged. (Probably before that). - others like a decorator will remain a distinct object and never become what they proxy. These would be very difficult to debug unless they handled the debugger messages themselves so should inherit DebuggableProtoObject. So it seems to me, but I'm rather new to these kinds of questions. /Mats

3 2

← Newer
1
2
3
4
5
6
7
8
...
42
Older →

Jump to page: