Squeak-dev July 2000

squeak-dev@lists.squeakfoundation.org

121 participants
566 discussions

Re: Displaying colored text
by Jon Hylands 26 Jul '00

26 Jul '00

On Wed, 26 Jul 2000 18:05:03 +0200, Thomas Kuehne <kuehne(a)informatik.uni-kl.de> wrote: > I tried my luck with a ListComponent morph, but although the emphasis (normal/bold/italic) was respected the colors did not come through. > Any quick hint how to display a list of differently coloured text lines? If you do it by code, its easy. Just create a TextMorph, set its contents to the string you want to display, and ask for its Text object (asText). Then, for each color you want: - create a TextColor object with the appropriate color - send #addAttribute:from:to: to the Text object, with the TextColor and the range in the string you want the color applied to When you're done, send openInWorld to the text morph, and you'll see your text in colors. You can also do it by inspecting a text morph that is already opened, and sending the messages to it from inside the inspector. If you do that, you need to send #changed to the morph to make it redisplay in color. Later, Jon -------------------------------------------------------------- Jon Hylands Jon(a)huv.com http://www.huv.com/jon Project: Micro Seeker (Micro Autonomous Underwater Vehicle) http://www.huv.com

1 0

Face Animator
by ssadams＠us.ibm.com 26 Jul '00

26 Jul '00

I forgot to mention that the face animator only works in 2.8 and above. Regards, Sam ------------------------------------ Squeakers, See http://www.mindspring.com/~ssadams/squeak/ To download a standalone changeset for Luciano Notarfrancesco's really cool 3D face animation demo. All the required data files have been converted to methods within the code, so just load and go. Hey Wonderland experts, how about giving this guy some eyeballs and a mouth! Regards, Sam Sam S. Adams, IBM Distinguished Engineer, IBM Research tie line 444-0736, outside 919-254-0736, email: ssadams(a)us.ibm.com <<Hebrews 11:6, Proverbs 3:5-6, Romans 1:16-17, I Corinthians 1:10>>

2 1

[Enh] Prototypes in Squeak
by russell.allen＠firebirdmedia.com 26 Jul '00

26 Jul '00

Hi, This is just a quick change to Hans-Martin Mosner's original prototyping code to allow prototype objects in Squeak to have a 'delegate' slot, allowing delegated behaviour a la Self's 'parent' slots. This is obviously not a long term solution performance wise - did I hear that someone was working on this? Can we get an update on how it's going? :) Russell ---------------------------------------- Russell Allen russell.allen(a)firebirdmedia.com ----------------------------------------

1 0

Re: Window Frames for system windows
by Lex Spoon 26 Jul '00

26 Jul '00

Try opening a rectangle morph and an ellipse morph and putting the ellipse morph on top of the rectangle morph but extending past at least one of the rectangle's edges. Then, use the red halo to embed the ellipse into the rectangle. There you will have it. If you inspect the rectangle morph, you will see that "self fullBounds" is now larger than "self bounds". (Now the matter of drawing outside of one's *own* bounds is another matter. This is possible right now, but I believe it's done that way for performance....) -Lex Karl Ramberg <karl.ramberg(a)chello.se> wrote: > While we are at it, is there a way to draw a submorph behind it's owner? > That would make life easier. > > Karl > > Henrik Gedenryd wrote: > > > > Norton, Chris wrote: > > > > > I haven't given this much thought, but it strikes me that > > > submorphs should never be allowed to write beyond the bounds of their > > > owners. > > > > Chris, this is a crucial and explicitly intended feature--the fullBounds > > construct would for example be unnecessary without it. Besides the halo, how > > would you for instance allow a HandMorph to grab other morphs without weird > > restrictions? Drawing shadows is another issue, although that is a bit of a > > pain anyway. I think it comes down to that Morphic is not made (just) for > > drawing widgets, which are a very small subset of the interaction space. > > > > Henrik

1 0

RE: Intro To Morphic, by John Maloney
by Lex Spoon 26 Jul '00

26 Jul '00

It's a real fascist that doesn't allow web requests to port 8888. To answer your question, it really is convenient to run multiple servers on the same computer. -Lex Ross Boylan <RossBoylan(a)stanfordalumni.org> wrote: > As "Jochen F. Rick" <nadja(a)cc.gatech.edu> suggested (off list), this was a > firewall problem. I can get at the file fine from at home. The pdf link > uses a special port (is that necessary?), and the firewall wouldn't let it > through. > > The file is quite reasonable to download over a modem. Someday I may even > have a chance to read it! > > Thanks. > At 11:25 AM 7/24/2000, Ross Boylan wrote: > >The link to the pdf doesn't work for me, and the graphics for the buttons at > >the top of the page aren't there either. > > > > > -----Original Message----- > > > From: Mark Guzdial [mailto:guzdial@cc.gatech.edu] > > > Sent: Friday, July 21, 2000 9:03 AM > > > To: squeak(a)cs.uiuc.edu > > > Subject: Intro To Morphic, by John Maloney > > > > > > > > > A chapter on Morphic (for beginning-to-intermediate Morphers) by John > > > Maloney is the latest submission to the SqueakBook: > > > http://coweb.cc.gatech.edu/squeakbook/30 > > > > > > Since this is a last-minute addition (we're starting to integrate the > > > final version in AUGUST!), quick reviews would be appreciated! If > > > you'd always wanted to learn Morphic, please take a look and tell us > > > how well the chapter worked for you. > > > > > > THANKS! > > > Mark > > > -------------------------- > > > Mark Guzdial : Georgia Tech : College of Computing : Atlanta, GA > > > 30332-0280 > > > Associate Professor - Learning Sciences & Technologies. > > > Collaborative Software Lab - http://coweb.cc.gatech.edu/csl/ > > > (404) 894-5618 : Fax (404) 894-0673 : guzdial(a)cc.gatech.edu > > > http://www.cc.gatech.edu/gvu/people/Faculty/Mark.Guzdial.html > > >

1 0

Re: Buffered streams (was: RE: [Win32] VM Update for 2.7/2.8)
by Lex Spoon 26 Jul '00

26 Jul '00

Andres Valloud <sqrmax(a)prodigy.net> wrote: > Hi Lex. > > > > Let's consider this scenario too. Suppose you have a dirty buffer from 0 to > > > 100. Then you set the position 200 and send nextPutAll:. So you shouldn't > > > flush. > > It's unlikely that someone is going to position: back to spot 100 and > > start increasing the buffer again. So, most implementations would just > > do a flush whenever position: is called, anyway.... > > Just consider peekFor:. A client of it, nextChunk in PositionableStream. > Flushing or attempting to flush everytime position: is sent is not a > good idea (been there). > Well, the root problem here is that #peekFor: should really be implemented using #peek, don't you agree? #peek (and the missing #peek:) should be much easier to manage than arbitrary position changes. Anyway, a fair generalization is that buffering strategies can be as complex as one feels like making them. Even so, there are surely a few simple things that would perform well in most situations. -Lex

1 0

[FIX] Behavior>>basicNew: infinite failure loop
by Andres Valloud 26 Jul '00

26 Jul '00

Hi. If you try something like Array new: 347862478627862786278278 then Squeak enters an infinite failure loop. The problem is in Behavior>>basicNew:. After the primitive fails, Squeak is told to signal the low space alarm and, if the user proceeds, the operation is tried again. But there's plenty of memory, so the alert doesn't come up and the operation is tried again. And again. And again. And again... This is solved by assuming, as in the comment of Behavior>>basicNew:, that if the primitive fails and the argument is an integer >= 0, then the problem is that the memory is low. The new assumption is that if even after trying to get more memory the primitive still fails, the problem is that the argument is too large, so the primitive should fail. A new method, basicBasicNew:, handles this. We also noticed that #primitiveFailed was present in Object but not in ProtoObject. It turns out that under some questionable circumstances, ProtoObject might send #primitiveFailed to itself. We moved this method to ProtoObject. Andres & Luciano.

1 0

RE: #changeClassToThatOf: in Squeak?
by Brent Pinkney 25 Jul '00

25 Jul '00

I did some work in 2.7 on explicit metaclasses in Squeak. I added a #primitiveChangeClass to the VM to modify the class of an object and also Object>>#class: I have not had time to do more work on the explicit metaclass stuff due to work and have been reluctant to have a go at compiling the 2.8/2.9 VM as it sounds very volatile. Have a look at http://sites.netscape.net/pinkles/neosqueak It is still very immature and I havent't apudated the web site in months but the VM changes seem to be ok. If you are into compiling your own VM and just want the change set let me know (the .cs is much smaller than downloading th 2.7VM). I too would like to see the primitve added to the base image. It makes the SUGAR guys object\instance protocol much simpler. Groete, Brent

1 0

Re: XML-RPC in Squeak?
by Bolot Kerimbaev 25 Jul '00

25 Jul '00

A simple (see notice below) implementation is included with Comanche (web app dev framework): http://comanche.swiki.net Notice: the XML-RPC and SOAP were started, but are not entirely complete. The obvious big thing that's missing is a tighter integration with Squeak, although some starting code is there. Oh, and this includes my own XML parser - there are a few dozen more out there in the Squeakville. I'm planning to take this implementation further to completion this Fall, as it will be an essential component of another project. I might completely discard my code in favor of Camp Smalltalk work, since they've turned out tons of useful stuff (still have to dig through for relevant treasures). And I'd be interested in looking at your code. Bolot On Tue, Jul 25, 2000 at 09:29:12PM +0200, Udo Schneider wrote: > Is somebody working on a XML-RPC (http://www.xmlrpc.com) > implementation in Squeak? > > I'm just hacking together some stuff for parsing it. Basic parsing and > translation in Smalltalk Objects works right now. > > The hard part will be the validation and the server. > > If somebody is interested in the code, drop me a mail and I'll send it > to you. But warning. It's quick and dirty NEWBIE code!! > > Udo >

1 0

Re: #changeClassToThatOf: in Squeak?
by Bob Arning 25 Jul '00

25 Jul '00

On Tue, 25 Jul 2000 17:22:22 +0200 (CEST) Johan Brichau <johanb(a)asterix.rave.org> wrote: >Does anyone know how to achieve the same functionality in Squeak of the >#changeClassToThatOf: message on Object in VisualWorks Smalltalk? > >This message changes the class of the receiving object to the class of the >argument. Johan, You might try Object>>as: and see if that does what you want. Cheers, Bob

1 0

← Newer
1
...
7
8
9
10
11
12
13
...
57
Older →

Jump to page: