[racket-dev] [plt] Push #26344: master branch updated
Well, I didn't try to explain _that_ part of the issue here (what I called
"b" in my message on the audit trail). It is much more complex and based
more on conjecture, so I'm hesitant to even try especially when the
existing discussion covers a more general, related problem that is likely
to satisfy people. (I hope.)
Robby
On Fri, Feb 22, 2013 at 3:56 PM, Sam Tobin-Hochstadt <samth at ccs.neu.edu>wrote:
> Ok, that makes sense.
>
> I think you should maybe add the word "floor" in the nitty-gritty
> documentation. You might also link from that discussion in the docs
> to the Cairo docs here: http://cairographics.org/operators/ .
> However, I can't find anything where they say they round down rather
> than to nearest. Grrr. Makes me glad for the Racket documentation
> (such as all the nitty-gritty sections you've written).
>
> On Fri, Feb 22, 2013 at 4:49 PM, Robby Findler
> <robby at eecs.northwestern.edu> wrote:
> > No, I don't think it does. The drawing operation is called "over" and it
> > takes two colors (rgb, say), and an alpha and produces a color (rgb). So
> you
> > always start with some solid color in the background and draw the shape
> on
> > top of it, resulting in another color that some other shape can be drawn
> on
> > top of.
> >
> >
> > On Fri, Feb 22, 2013 at 3:45 PM, Sam Tobin-Hochstadt <samth at ccs.neu.edu>
> > wrote:
> >>
> >> On Fri, Feb 22, 2013 at 4:40 PM, <robby at racket-lang.org> wrote:
> >> >
> >> > + The list of colors is obtained by drawing the image on a white
> >> > + background and then reading off the colors of the pixels that were
> >> > drawn.
> >>
> >> Is there a particular reason to choose white, as opposed to clear, for
> >> the background in this test? Does drawing on a "clear" background
> >> even make any sense in this case?
> >>
> >> Sam
> >
> >
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.racket-lang.org/dev/archive/attachments/20130222/c6fa637d/attachment-0001.html>