Squeak-dev September 1999

squeak-dev@lists.squeakfoundation.org

133 participants
357 discussions

RE: GradientFillStyle oddity
by Raab, Andreas 03 Sep '99

03 Sep '99

> > Congratulations! You just discovered one of the most serious bugs in the > > Balloon engine. I'm amazed nobody stumbled across that earlier ;-) > > Gee, I feel honored. :) > And rightly so ;-) > Okay, it's supposed to work for ovals, but the interpolation doesn't > quite work. Now, is it also supposed to work when the normal isn't > normal to the fill's direction? > No - for the time being this works neither for gradient nor bitmap fills. > > The only way to work around the problem is using a bitmap fill generated > > from a radial gradient. This should work until I've fixed the problem > (which > > will take some time). > > Thanks for the suggestion. Now, is there a way I can make the bitmap > fill use translucent colors? Aw, what am I asking for, I was going to > play around with those soon, anyway! :) > Just use a 32bit deep form and make sure the alpha channel is okay. This should work (I hope ;-) Andreas -- +===== Andreas Raab ========= (andreasr(a)wdi.disney.com) ==+ | Walt Disney Imagineering Phone: +1 818 544 5016 I I Glendale, CA Fax: +1 818 544 4544 I +======< http://isgwww.cs.uni-magdeburg.de/~raab >========+

1 0

Re: GradientFillStyle oddity
by Jesse Welton 03 Sep '99

03 Sep '99

Raab, Andreas wrote: > > Congratulations! You just discovered one of the most serious bugs in the > Balloon engine. I'm amazed nobody stumbled across that earlier ;-) Gee, I feel honored. :) > Seriously, what happens is that for the sake of speed I'm doing an extremely > ugly (but very fast!) quadrant interpolation scheme for gradient fills. > *Usually* this works just nice (as you noticed for everything that is > symmetric) but for the particular case you've shown it does not. Okay, it's supposed to work for ovals, but the interpolation doesn't quite work. Now, is it also supposed to work when the normal isn't normal to the fill's direction? I think that should be equivalent to some rotated oval, and it does look much the same, but it seems wrong to violate the implied constraint. (There were a couple of places where I got some nice effects without having to do much calculation by setting the normal at some non-right angle to the fill direction.) > The only way to work around the problem is using a bitmap fill generated > from a radial gradient. This should work until I've fixed the problem (which > will take some time). Thanks for the suggestion. Now, is there a way I can make the bitmap fill use translucent colors? Aw, what am I asking for, I was going to play around with those soon, anyway! :) -Jesse

1 0

GradientFillStyle oddity
by Jesse Welton 03 Sep '99

03 Sep '99

I was playing around with eliptical GradientFillStyles recently, and found there's an oddity in the display of two octants when the axis lies on a diagonal. This is well illustrated by the following snippet of code: | rect fill | rect _ (RectangleMorph new) extent: 200@200. fill _ (GradientFillStyle ramp: {0.0 -> (Color blue). 0.75 -> (Color red). 1.0 -> (Color yellow)}). fill origin: rect center; direction: 100@100; normal: -50@50; radial: true. rect fillStyle: fill; openInWorld. One lower right octant and one upper left octant have straight gradients, not rounded gradients. I hesitate to call this a bug instead of an oddity, because I think perhaps the radial gradient is intended to be used in radially symmetric situations, such as if I'd put normal: -100@100 above. But I would like to use it in ovals, since the effect can look extremely slick (especially with translucency - woo!). One other thing I noticed about it is that if you drag another morph a few pixels at a time* across one of the above morphs, from left to right, the aberrent octants are fixed a few columns of pixels at a time. So apparently the code to render these octants correctly does exist somewhere; it just isn't used when filling the whole region at once. I have no idea where to look to fix this to behave as I'd like. (* Click, drag, and release, a few pixels at a time, until you get across the morph. It also worked to continuously drag a morph I had created that changed itself slightly with every mouseMove.) Any suggestions? -Jesse

1 0

help with flaps in morphic
by Jay Hardesty 03 Sep '99

03 Sep '99

Hello - I used to be able to park objects on flaps by choosing 'suspend parts bin behavior' from the menu and then dragging a window over the flap. The preferences 'use global flaps' is true, and 'project viewers onto flaps' is true. But now when I drag say a morphic workspace over the exposed flap, the morphic window won't attach itself to the flap. In the past a shrunken version of the window would appear on the flap. Can anyone tell me what I'm doing wrong? I'm using version 2.5 (mac-ppc vm) Thanks for any help. Jay Hardesty jayh(a)panix.com

2 1

Re: [UPDATES] 19 fixes and enhancements
by Bruce ONeel 03 Sep '99

03 Sep '99

Hi, I've updated all of the packages on ftp://st.cs.uiuc.edu/Smalltalk/Squeak/2.5/ to have a copy of Squeak2.5.[image|changes] with updates through 1407. Of course for those of you who already have downloaded it it's way faster to just to update from server rather than download 5meg or so again :-) cheers bruce Dan Ingalls <Dan.Ingalls(a)disney.com> wrote: > Folks - > > I have just released 19 updates which address most of the outstanding bugs in 2.5, as well as several other features.

1 0

Re: anAnyClass scaleBy:<foo>. What should <foo> be?
by Reinier van Loon 03 Sep '99

03 Sep '99

John-Reed, >I am making modifications to the scaleBy: method in >class Path. That method is scaling points but does not use the >class Point scaleBy: method and I think that it should. [snip] >checkScale: aScale > "Convert aScale to the internal format of a floating-point Point." > > | checkedScale | > checkedScale _ aScale asPoint. > ^checkedScale x asFloat @ checkedScale y asFloat > >--- end quote > >it seems to me that scaleBy: aFloatingPointPoint should be the >standard. Yes? No? Maybe? > If you're going to make modifications why not rename scaleBy: to scaledBy: as the receiver does not get scaled but answers itself scaled by aScale. Same for transformBy:. That would be nice standards. Just this week a trainee was completely confused by this. She expected the object to scale but found nothing happened. After I explained the 'name' bug she looked a bit funny at me... Regards, Reinier

1 0

Re: [BUG]Stream class>on:from:to: & reset
by Alan Lovejoy 03 Sep '99

03 Sep '99

> ** Original Sender: Tim Rowledge <rowledge(a)interval.com> > > You are right, this would foul up the nextPut primitive for WriteStream, so > you would > > have to add the new field separately after those defined in ReadStream and > WriteStream > > rather than doing it in one place in PositionableStream. > Duh, talk about missing the obvious one! Still, the prim recovery code > handles it just fine.... > > > Otherwise, it should be str > > aightforward, but a bit of a shame to add something that's hardly ever > > used (though, I admit, frustrating when it's not there). > Well, a similar argument could be made about a lot of stuff in Smalltalk; > how often does anyone actually use metaprogramming, reified Contexts, etc > etc? Yet it would be a much lesser system if they were not there. I find > considerable value in (for want ofa better term) correctness, not to > mention pragmatic value in minimizing nasty surprises. Well, Tim: Looks like you and I are in violent agreement on this issue! Hope that doesn't cause you to reconsider... :-). --Alan

1 0

Lampson/Thacker to do Dynabook?
by Tim Rowledge 03 Sep '99

03 Sep '99

Yet another project to do a Dynabook. Maybe they'll be able to get further than the ones I've been involved with? (Annoyingly, this requires a (free) subscription.) WILL THE FUTURE OF COMPUTING BE WRITTEN ON A TABLET? Enlisting the talents of two computer designers who were part of Alan Kay's legendary team at Xerox's Palo Alto Research Center in the 1970s, Microsoft has plans for developing a portable, wireless, keyboardless "tablet computer." Although generally similar devices developed by Apple, Go Corp., and AT&T were commercial failures in the past decade, Microsoft is betting that the current rapid convergence of display, processing, storage and other technologies is finally setting the stage for commercial tablet computers using handwriting or speech input. (One recent development has been Microsoft's introduction of its Cleartype software that improves the readability of fonts on flat panel computer displays.) The two designers who will lead the new Microsoft tablet computer effort are Chuck Thacker and Butler Lampson. Lampson says, "I think this will be the way most people interact with the Net and the rest of the computing universe as well." Thacker injects a personal note: "I've always wanted this kind of device, and in systems research one of the most motivating things is that you want the device yourself." (New York Times 30 Aug 99) http://search.nytimes.com/search/daily/bin/fastweb?getdoc+site+site+82844+0… -- New: It comes in different colors from the previous version. Tim Rowledge: rowledge(a)interval.com (w) +1 (650) 842-6110 (w) tim(a)sumeru.stanford.edu (h) <http://sumeru.stanford.edu/tim>

5 5

RE: [ENH] Image based session management
by Stephen Pair 03 Sep '99

03 Sep '99

Tim Rowledge wrote: > Nice idea. I _think_ that with just a little rearrangement of code (in > ioExit & main) and adding a new exit prim we could actually make it > possible for you to call exitAndRestart: anImageName. That would be great. It's a pain to have to manually restart, but one that I'm willing to put up with for the moment. - Stephen

1 0

StreamSequence
by Andres Valloud 03 Sep '99

03 Sep '99

Hi. What things should this object support? Consider that lots of features will mean lots of index and position management overhead... 1. On the fly removeAt: and add:readFrom:to: without losing position. 2. On the fly resizing of stream segments sequenced. 3. Support for nextPut:, enlarging the last stream sequenced when appropriate. 4. Reverse next. 5. Access by substream. 6. Any other? 7. A better name? Andres. PS: note that the method removeAt: has a little bug in the last changeset, it should include the line size _ nil. somewhere.

2 1

Jump to page: