18:10:21scymtymfroggey: i didn't mean to say the compiler was bad. we have to keep in mind that other implementations often fall back to libc or even gmp for certain numeric things
18:10:53scymtymand trowing around rationals in a software rasterizer is also pretty bold :)
18:17:38froggeyoh, don't worry. the compiler definitely is bad. occasionally it can be convinced to produce decent code, but it falls to bits more often than not
18:18:07froggeyand you got the profiler working, impressive!
18:25:48scymtymi'm actually hooking the profiler up with this: https://techfak.de/~jmoringe/clamegraph-rewrite-1.ogv
23:11:08jackdanielclim specification was written way before I've even starting learning common lisp, I wouldn't call it my thing; but sure, that's what we aim at -- providing usable toolkit which is a CL project (not a binding)
23:13:06jackdanieland callbacks are called in a context of frame, so *application-frame* is accessible, so you may avoid global side effects
23:13:54jackdanielredisplay frame panes is to ensure, that the application pane shows updated *xxx* and *yyy* (notice, that display function prints them)
23:15:03thrashdinand this `(value (gadget-value gadget))` is what was in the text field in the first place?
23:15:05jackdanielscymtym: I've noticed some problems with text-field gadget when writing that -- will report the ticket tomorrow. 1. it misbehaves in with-output-as-gadget (missing slot), and it always requires value -- it could provide empty adjustable vector by default i.e
23:15:26jackdanielyes, gadget-value is the accessor to the text-field text