<daviid>i think these 2 modules should merged, wdt? <daviid>ok i'll try to work in that, as time allows <daviid>we have to release g-wrap, guile-cairo, guile-gnome and guile-clutter stricktly depending on guile-2.0, debian wihdraw guile-1.8 from unstable and soon testing ... <daviid>mark_weaver: i1ll try, i like the idea to merge these 2 <mark_weaver>janneke: it would be helpful to profile it, and find out which operations have gotten slower. <sneek>wingo was here Apr 20 at 12:04 am UTC, saying: yeah. <sneek>dsmith-work was here Apr 21 at 05:36 am UTC, saying: I think blue is best.. <sneek>I last saw dsmith-work on Apr 21 at 05:36 am UTC, saying: I think blue is best.. <sneek>dsmith was here Sep 14 at 10:33 pm UTC, saying: sneek, seen dsmith. <mark_weaver>dsmith-work: I see in my logs that wingo was here two days ago. ***jjmarin_ is now known as jjmarin
<artyom-poptsov>"mornoon" sounds like a combination of "morning" and "afternoon" <lloda>question: is it possible to use the repl debugger on the compiler itself? I get While executing meta-command: ERROR ... and no new prompt. <wingo>lloda: the debugger isn't enabled for meta-commands; perhaps we should add a mode <wingo>anyway, you can call (compile FOO) <wingo>or (compile FOO #:env (current-module)) <wingo>where FOO is a quoted expression <nalaginrut>wingo: it's better to have compile-from-string for multi-lang debugging <wingo>i think you might be able to (read-and-compile (open-input-string STR) #:to 'value) <wingo>if you use (system base compile) anyway <nalaginrut>hmm..it's a little bit circle, but yes it works ;-P <lloda>wingo, (compile FOO) doesn't give me a prompt. While compiling expression: ERROR ... and no prompt. <wingo>lloda: if you've really broken the compiler so that it can't compile the (compile ...) expression then you can ,o interp #t <lloda>ah, that works. very good, thanks. <lloda>btw (read-and-compile (open-input-string "99") #:to 'value) -> $2 = ERROR: In procedure simple-format: <lloda>ERROR: In procedure simple-format: FORMAT: Unsupported format option ~4 - use (ice-9 format) instead <lloda>c5ea75534c683638fb1f3b328efe6ba0d242e485 ? ***Guest54372 is now known as micro
***wlkely is now known as wkelly
<sneek>wingo was here Sep 15 at 09:13 am UTC, saying: strange. <sneek>ijp was here Sep 15 at 02:15 pm UTC, saying: sneek: seen ijp. <ijp>a simple mistake to make <taylanub>fangism: not in my logs. I think there were also public logs somewhere, or was that only #guix... <stis__>heh saw mark_weaver warn against guile-cl! <stis__>his right, the good part is based on guile-syntax-parse, which is not gnu <stis__>The good one is the loop macro implementation <stis__>but who wants cl's loop macro :-) <stis__>on the other hand the idea of gule-cl of implementing missing parts of cl in a schemy way is not a bad idea <stis__>for example what is a good schemy way of attaching properties to syntactic objects like in cl's get and put <stis__>my version in guile-cl is not good atm. PLease implement the correct one :-) *stis__ will maybe implement a gnu syntax-parse later on. <stis__>dsmith-works: thanks for the link, really interesting. *stis__ should also make sure that the guile-log doc becomes better *daviid still write (call-next-method) [instead of (next-method)], even after 15y of using goops, i wonder how much control we have on our oan brain :) <janneke>daviid: that's an interesting question <daviid>janneke: while you are here :), i saw on your gdk related example you used . for you accessors to preceed slot names... I'd like to suggest ! instead, wdt? it may sound a little thing but (1) i don't think we want people to think goops comes after and inspires itself from java :), (2) there i a tradition in scheme that mutable procedures end with !, since an accessor can access or used to mutate, set!, hence the preceeding ! [instead <daviid>of following] (3) it would be nice to uniform our examples, I think