IRC channel logs


back to list of logs

<mwette> iff
<daviid>mwette: tx!
<roelj>Is there a standard “throw” key for missing arguments?
<foojin>roelj: 'wrong-number-of-args seems to serve this purpose.
<roelj>foojin: Thanks. Is this documented in the manual somewhere? (I couldn't find it..)
<foojin>roelj: Yep, 6.13.13 How to Handle Errors
<roelj>foojin: Oh snap. Thanks!
<foojin>You can use '/' in standalone info or 's' in Emacs to search these with a regular expression.
<foojin>Emacs info mode that is.
<foojin>You're welcome. BTW I wonder if the Emacs one can be made not to ascend above the node where you started it.
<roelj>Is it also possible to not have to go into the Guile manual first?
<roelj>Or to start info for the Guile manual specifically?
<foojin>M-x 6.13.13 How to Handle Errors
<foojin>You can either invoke info like this: "info guile" or do M-x info-display-manual and type guile there.
<roelj>info-display-manual is nice. Thanks once again!
***snap-l is now known as cmaloney
***ft_ is now known as ft
***Raimondii is now known as Raimondi
<dsmith-work>{appropriate time} Greetings, Guilers
<foojin>Does Geiser consider everything evaluated in Scheme buffers in the context of the manually specified module (,m module) or does it sometimes switch/guess automatically?
<jao>foojin, it uses the file's module, not the repl's
<jao>only when no module can be assignd to the buffer's file does it fallback to the current namespace in the repl
<foojin>So if I evaluate something in the buffer without having used ,m in the REPL and then switch using ,m module, everything would be in scope?
<foojin>Everything that I've evaluated, that is.
<OrangeShark>hello everyone
<jao>foojin, yes, it should. everything you evaluate in the buffer will be evaluated in its module.
<jao>dsmith-work, dale! :)
<dsmith-work>jao: Did you see that paper by Didir Verna on Lisp, Jazz, and Aikido?
<jao>dsmith-work, i did, but i'm afraid it's still on my TO-READ list
<dsmith-work>jao: Cool.
<jao>dsmith-work, i take it you liked it
<dsmith-work>Yeah, very interesting.
<jao>i have to read it. i even took aikido classes back in the day.
<dsmith-work>Heh. And my old co-worker, Gary Bernhardt, is mentiond!
<dsmith-work>(his famous "Wat?" lightning talk)
<jao>ah, yes
<foojin>jao: I see. By the way, these contexts have to pile up somewhere if one evaluates code in different modules. Is there a way to "clear" one of them?
<jao>foojin, they're not "context" on the emacs side, just the definitions associated to each guile module
<jao>it's actually guile who's keeping track of them
<jao>so "clearing" would mean removing all bidings from a guile module
<jao>i woudn't be surprised there's an easy way to do that, but geiser does not provide any shortcut for it
<foojin>jao: I thought as much. I'll look into it, along with the way to make the current namespace fallback the default. It just seems _too smart_ for me to be comfortable with it.
<davexunit>use geiser for long enough and the other way will seem too dumb
<davexunit>if I had to set the current module in the repl each time it would get really frustrating
<davexunit>the programs I work on consist of many modules that I switch between frequently. it's really nice that geiser knows the module I'm editing when I eval stuff
<foojin>If I'm not misunderstanding something, you'd have to switch manually each time you want to type a one-liner at the REPL anyways.
<foojin>Maybe it would become useful if I do everything in buffers.
<davexunit>if you're in a source buffer and want to switch to the repl buffer to type a one liner, you can do C-c C-a
<foojin>Ah, so there's a shortcut for that!
<daviid>ot; reading an article, I don't undertand this: the maverick company run by ... (talking avout spacex in this case, but i don't think it matter) what is a maverick company?