<cojy>is guile bytecode suitable for a typed language that can erase checks? <ijp>cojy: you're not liable to get an answer to that for about 8h unless mark_weaver chips in <ijp>I don't really remember how rtl changes things, but I think so <ijp>well, 'bytecode' wasn't really usable, IIRC it was more or less a translation of the assembly language <ijp>you also won't be able to avoid tagging ***DeeEff_ is now known as DeeEff
<ozzloy>wingo, i searched for "guile exit" <ozzloy>that brought me to that page as the top hit <wleslie>'those in the know use Control-D to close REPLs' ? <ozzloy>what does (emit port "string") do exactly? i can't find documentation for it <ozzloy>is there some list of all procedures somewhere? <ecraven>wingo: you do know geiser internals, right? <ecraven>I'm playing around with writing a backend for a new scheme implementation, but see strange behaviour :-/ <wingo>my guile is usually broken so i rarely have the pleasure of using guile <stis>not sure what that means. I'm not in us, kind of strange because how easy it is to get e.g. SWI prolog <wleslie>aww. well, it's a pun on 'unification' - not only unification of prolog terms, but also unification of countries <wleslie>(eq |unification on their terms| |north korea running the show|) <stis>ahh, lol now I get it (I'm too serious I need to go out and ghave a beer) <cojy>is there a guile library out there for rope data structure or similar ones like rrb-tree? <zzh>hello. if i eval (parse-header 'content-length "text/plain;charset=utf-8") from the guile manual. i get the following error : "Bad non-negative-integer header component: text/plain;charset=utf-8" . is this a bug? <holomorph>well, isn't http content-length supposed to be a non-negative integer? <cojy>zzh: looks to me like a typo in the documentation; copy/paste that never got finished <cojy>the section is "HTTP Header: List content-type" so you want 'content-type not content-length <zzh>yes, (parse-header 'content-type "text/plain;charset=utf-8") works