Squeak-dev August 2007

squeak-dev@lists.squeakfoundation.org

132 participants
163 discussions

Fix OB refactoring: Creating accessor change the instance/class variable
by Mathieu Suen 11 Aug '07

11 Aug '07

Hi, If you use the refactoring create accessor for instance or class variable some time fails: If you have a class define as: Object subclass: #A instanceVariableNames: 'c b' classVariableNames: '' poolDictionaries: '' category: 'Dummy' And add A>>b: by hand. After create the accessor for c with the refactoring in omnibrowser. "A new b: 3" will affect c and not b. This is because ob-refactoring sort the instance variable to display in the alphabetical order: ORCmdRefactoring>>instanceVariables ^ (target isKindOf: OBInstanceVariableNode) ifTrue: [ Array with: target name ] ifFalse: [ target theClass instVarNames asArray sort ] I have patched this method with the #asSortedArray. See the attached file I think is the method #asArray who make the mess if you look at the implementation of Array>>#asArray. I would rather expect a copy of self. I have open an issue: http://bugs.squeak.org/view.php?id=6602 Mth

2 1

Doc team June/July report
by Matthew Fulmer 11 Aug '07

11 Aug '07

I never sent a monthly report for the month of June, so this is a two-month report. I spent a fair bit of time in July on the beginners list helping people out with Stephan Wessels' sweet new tutorial, and providing several enhancements to it. I also re-organized http://www.squeak.org/Documentation so that it is easier to find the most useful documents -- Matthew Fulmer -- http://mtfulmer.wordpress.com/ Help improve Squeak Documentation: http://wiki.squeak.org/squeak/808

2 1

OT: Dolphin smalltalk giving up
by Chris Kassopulo 11 Aug '07

11 Aug '07

http://www.object-arts.com/content/blog/2007Aug10.html

3 2

OT: Blog: Smalltalk Is Dying
by Brad Fuller 11 Aug '07

11 Aug '07

It troubles me when I read things such as this: ...."and the only open source option is Squeak, which is not production-ready and never will be...." Don't know how many people read this blog, not many comments: http://factor-language.blogspot.com/2007/08/smalltalk-is-dying.html Thanks to Ramon for commenting on the blog.

2 1

Re: [Newbies] setting the font in a System Browser
by Brad Fuller 10 Aug '07

10 Aug '07

On Fri August 10 2007 12:19 pm, Mark Volkmann wrote: > The font dialog only lists a small number of fonts starting with > Accuat and ending with ComicSansMS. How can I get to fonts that come > after that alphabetically? Good question, the 3.10 dev image (Beta #7130), does the same thing. This is similar to what I reported with Andrew's FT2Plugin. Is this a bug? brad

1 0

versionuploader: filename.st is not in a known format
by Klaus D. Witzel 10 Aug '07

10 Aug '07

Hrm, versionuploader doesn't want stinkin' normal .st files :( Someone have a workaroud? I have no GUI on the remote server (it's in a service provider's data centre) other than Seaside's screenshot. ADVtanksANCE /Klaus

2 1

Re: Looking for TinyPaint (was: Re: PDAMorph)
by Brad Fuller 10 Aug '07

10 Aug '07

On Thu August 9 2007 4:35 pm, you wrote: > From Squeak 3.0: Javier sent me an .st of TinyPaint. Thanks Javier! (what a great community)

1 0

PDAMorph
by Brad Fuller 10 Aug '07

10 Aug '07

At one time in the recent past Tim R. suggested looking at the PDA project that he and Dan did. I did. but, now I can't find where I put it! There is a package called PDAMorph on squeaksource but it requires "WatchMorph". Is that available? Or, where can I find the PDA source? Or, is PDAMorph even the right package? Tim? thanks, brad

4 7

Morphic! (ans to Jason from previous thread)
by Jerome Peace 09 Aug '07

09 Aug '07

Morphic! (ans to Jason from previous thread) My previous attempt to post this was only partially successful. This is the full post. -Jer Hi Jason, Thanks for your interest. Spec means speculation and doing something on spec means you take the responsibility that it might or might not be accepted by the publisher. The other way to do something is on commission where others specify what you are to do but give a garantee that the work will be of use to them and you will be compensated. As for motivation for morphic work, for me it means the work must be interesting (as defined by my curiosity) lead to something I think elegant in the code and the behavior of the morphs and allow me to learn and grow in my own understanding of squeak, morphic and things larger. You apparently have the curiosity streak as well. So answer me this, and I can better advise you. What would you like to learn from reading the specifications? you thought I had mentioned? What would you like to be able to do with the knowledge? Often, instead of specifications, I find it useful to just write down what I want to see squeak do. As vaguely as my initial understanding permits. As briefly as I can fit all my ideas in. And as passionately as the forces of desire inside me wish for the feature to exist. Oddly, the rest of it is all downhill from there. The specifications become the detailing of the initial statement. The code comes from implementing pieces of the details, testing them and proving they work. The fun comes when the coding process leads to something unexpected and the detail or even the original statement becomes worth rewriting. Incremental design benifits the designer by educating him all along the way. (Look on the web for articles about McReadys gossamer condor.) Anyway thanks for your response and request. Its been fun writing this. Yours in curiosity and service, --Jerome Peace ____________________________________________________________________________________ Sick sense of humor? Visit Yahoo! TV's Comedy with an Edge to see what's on, when. http://tv.yahoo.com/collections/222

2 1

[ANN] Taming Fills
by Jerome Peace 09 Aug '07

09 Aug '07

[ANN] Taming Fills Hi Subbu, Thank you for your comments. I was hoping you would notice my work and I would get the kind of feedback you have given. Now, I would like to know what you mean by them. Can you draw me a picture of what your handles look like? My decision on four handles (an their current colors) are a first pass. And I have found better ideas come out of the woodwork in second and third passes. I was mostly aiming on getting the functionality to work on this pass. The dismiss handle is technically surperfluous (you can call up a halo and dismiss the handles that way). By anchoring the last corner of the square it helps visulization. The brown anchor is the origin and serves as host to the others. The blue and yellow handle in their shifted version are the unlimited actions. They are complimetary. Blue scaleRotate is the analog of a polar transformation (angle and radius) and works as you describe the second handle to work. while Yellow stretchReflect is the analog of a cartesian transformation ( scalling x and y separately ). And works like the yellow growth handle works on the halo. Except my yellow handle allows for reflection. The unshifted handles are constrained s.t. the blue mimics its halo handle and the yellow can only scale and reflect w/o changing the aspect ratio. The yellow halo handle behaves that way currently when it is shifted. Except my yellow handle allows for reflection. My experience is that you want the handles to be limited in their normal operation and unlimited in special operation. That provides the user a choice not to change invariants such as aspect ratio, in the case of the yellow handle, or scale, in the case of the blue. My current thought is to substitute a checkbox for the shift key. With both handles being either limited or unlimited as the check box gets toggled. I did add balloon help to the handles to indicate their purpose. I think your comments show that neither the shiftkey nor the balloon help is a good way to convey that to the hands on user. That's an insight worth having too. Yours in curiosity and service, --Jerome Peace *** >subbukk subbukk at gmail.com >Wed Aug 8 12:29:08 UTC 2007 > > >On Tuesday 07 August 2007 11:39 pm, Jerome Peace wrote: >> Solution: Have handles for the fill that allow >> manipulation of all the parameters until the user is >> satisfied. Then allow the removal of those handles. >Jerome, > >It is nice to be able to control the fills directly through handles, but the >handles are too many. >A simple 'rubber band' with two handles (anchored and >movable) is sufficient. What do you mean by rubberband? Circular gradient will need a ellipse with two >orthogonal 'rubber band' lines. Lost me here. Where do the lines go and what do they do thats different? > >Use a color dropper on a handle to change the color. Ah but fills can be much more glorious than two colors. I was thinking of something like a slider with multiple slides each one capable of defining an new color on the ramp. And of course a midpoint or two to add even more colors. Drag anchored handle to >change origin and moveable handle to scale and rotate. Shift click on the >rubber band to dismiss the handle. This would be consistent with operations on mixed curve and border width etc. Shift-move on the blue handle does scale and rotate. But try shift-moving on the yellow. That is what it is really there for. And try it on a bitmapfill. I can see the check box is really whats needed. I want to see what your rubberband looks like. Thanks again for you help. > >Regards .. Subbu *** ____________________________________________________________________________________ Got a little couch potato? Check out fun summer activities for kids. http://search.yahoo.com/search?fr=oni_on_mail&p=summer+activities+for+kids&…

2 1

← Newer
1
...
10
11
12
13
14
15
16
17
Older →

Jump to page: