Matthew pointed out that that's how record-dc% already works! So now the example you write below works properly.
Sorry for my confusion. Robby On Sat, May 26, 2012 at 8:14 PM, Robby Findler <[email protected]> wrote: > That could probably be added to record-dc% (which the repl uses). > > Robby > > On Sat, May 26, 2012 at 2:58 PM, Neil Toronto <[email protected]> wrote: >> On 05/26/2012 08:17 AM, Robby Findler wrote: >>> >>> On Sat, May 26, 2012 at 10:09 AM, Jens Axel Søgaard >>> <[email protected]> wrote: >>>> >>>> 2012/5/26 Robby Findler<[email protected]>: >>>>> >>>>> As for cropping, the DrRacket REPL is the thing that does the cropping >>>>> >>>>> (and, in some cases, it may not be cropping, actually, but two >>>>> different parts of the window are fighting with each other for the >>>>> pixels when something draws outside of its bounding box, as a scaled >>>>> hline does). >>>> >>>> >>>> Okay, I thought there was a guarantee that the drawing happened >>>> inside the bounding box. >>> >>> >>> That is not a guarantee and importantly not one (for picts, generally >>> speaking, that is). refocus is a common way that guarantee fails to >>> hold, for example. >> >> >> It really should be guaranteed by the REPL, though. This is awful: >> >> #lang racket >> >> (require slideshow) >> >> (dc (λ (dc x y) >> (send dc set-background "magenta") >> (send dc clear)) >> 100 100) >> >> Shouldn't snips generally not be allowed to draw outside their bounding >> boxes? >> >> Neil ⊥ >> >> ____________________ >> Racket Users list: >> http://lists.racket-lang.org/users ____________________ Racket Users list: http://lists.racket-lang.org/users
