IRC channel logs
2013-09-25.log
back to list of logs
*civodul just filled out the GSoC forms <sneek>I last saw bipt on Sep 19 at 09:44 pm UTC, saying: so for CL, for example, i wouldn't find it too inconvenient to require a modules.scm packages or similar, vs. having a module definition written in portable CL. i also think ML-style module features are more natural with separate configuration, at least based on my limited experience with scheme48. <sneek>add^_ was here Sep 25 at 12:53 pm UTC, saying: sneek: seen add^_ . <add^_>That's actually a kinda convinient way of checking what time it is in UTC. <add^_>Hm, maybe sneek have a time command? <wingo>we don't provide pre-built binaries <wingo>guile does build on windows, in various ways (cygwin and mingw) <wingo>you choose whether you want to do it using cygwin or mingw <wingo>you choose whether you want to build it on windows or cross-compile from linux <wingo>then you set up your build environment with the right tools, and from there it's just a normal autotools project <wingo>it can be a bit frustrating, admittedly... <wingo>that warrants a mail to guile-user so you can get the right people to respond :) <sneek> 09:49:44 up 29 days, 1:58, 0 users, load average: 0.01, 0.05, 0.05 <add^_>Maybe that's something for ijp and aidalgol to add ;-) <add^_>Unless ofcource they already done that <ijp>adding it would be trivial, but better still would be adding a time command like lambdabot has, that can give localtime for a particular user <ijp>so then you can do "time add^_" and get whatever time it is where add^_ is <ijp>that would be one way to do it <ijp>I forget what lambdabot does <civodul>just in case someone is look for a place to hack :-) <ijp>d'oh, I see untranslated markup <davexunit>civodul: so it's an init system written in guile? <davexunit>everyone seems to be using systemd these days <foeniks>hi, I have been using port->string while writing my clojure parser since I were annoyed by the statefulnes of read-char. <dsmith-work>civodul: That's the init manager for hurd that Wolfgang somename wrote in guile? <sneek>foeniks, you have 1 message. <foeniks>I do wonder what your opinion is on that approach <foeniks>for now its seems that it is more problematic to get line number and character number that way <ijp>okay, post should be fixed nwo <ijp>many of you have seen the code already, but I promised shanecelis I'd write it up <ijp>it also means I am desperately in need of something utterly ridiculous to write <foeniks>is there a way to get all available identifiers in the current namespace? <bananagram>can you sleep for a fraction of a second somehow? <ijp>all the ones bound in a module is easy, all the imports is a little trickier <ijp>the former is (hash-map->list (lambda (k v) k) (module-obarray (current-module))) <ijp>to get imports, you'll also need to scan the uses list <ijp>foeniks: out of curiousity, what are you doing? <ijp>usleep isleep weallsleep <mark_weaver>bananagram: usleep accepts only integers, but those integers are in units of microseconds. <add^_>Uh, I just accidentally forked cunning-bot when I wanted to copy the copying text.. lol <add^_>Hm, renamed the project, but now I've named it back again, might as well keep the original name if I want to help out :-P *add^_ is feeling silly, as if he had too much whisky <foeniks>ijp, filling up a vim guile mode's function list <civodul>davexunit, dsmith-work: yes, it the init system writte 10 years ago (!) by Wolfgang Jährling <civodul>it's very GOOPSy, but pretty cool (nonetheless ;-)) <davexunit>civodul: do you plan on getting it up to snuff to be a real contender? <civodul>it's not new, it's 10 years old! :-) <civodul>it's definitely better than most init systems out there <civodul>but not as fancy as systemd in some ways <davexunit>I just frequently see flame wars about systemd <mark_weaver>systemd has some great features, but its decision to exclude support for all kernels other than Linux is a deal breaker for me. <mark_weaver>I also dislike that so many changes require recompiling from C. <mark_weaver>IMO, a better long-term strategy is to incorporate the good aspects of systemd into dmd. *civodul is consensual ;-) <civodul>not but seriously, that's the thing: there are good ideas, but it relies on Linux-only hacks, and it's C, which makes it not-so-much hackable <civodul>but systemd wants to eat your whole system ;-) <add^_>If I do this: guile --listen -L blah -L blah2 <add^_>will it include both of those directories? Or should I do it some other way? <add^_>like -L "blah:blah2" or something? <dsmith-work>add^_: try guile -L /foo/bar -L /baz/zap -c '(display %load-path)(newline)' <mark_weaver>add^_: yes, it will include both. -L blah:blah2 won't work. <mark_weaver>unlike GUILE_LOAD_PATH, which gets parsed into colon-separated components, -L does _not_ get parsed. <mark_weaver>-L simply gets prepended to the %load-path as-is, regardless of what characters are in it. <mark_weaver>(assuming of course that you've properly quoted it in your shell) *add^_ is trying to load everything in the spells project but isn't really succeding.. <add^_>Because something needs something else and that something else needs yet another something.. <add^_>Actually, thinking about it, it *should* work.. <add^_>But it gives me a warning about network.sls can't compile because there is no code in the "network compat" module.. <add^_>Actually, I didn't solve the problem, ARGH.... Still, sorry for the noice... <add^_>Oh.. For some reason it doesn't find the file because it's called something else than the module is? But why does it work for ijp and not me? <add^_>ijp: are you using your fork of spells? <add^_>No, can't be, since it's no commits after or before the master one.. <add^_>Not in the repo on github at least <add^_>And I think it should load .sls files on this setup, because I should have explicitly told guile to read those.. <mark_weaver>add^_: to clarify, ijp renamed the files to .scm in guildhall <add^_>For some reason guildhall doesn't work for me anymore.. <add^_>didn't work with the -x .sls <add^_>I guess I can try to install it again, but if it messes up my guile again I'm not going to be happy ;-) <add^_>There is an error saying "R6RS exception: &assertion" <add^_>I dunno what happened, if guile was updated in some way that made it break <add^_>I'm just following the wiki :-/ <add^_>Well, I added it now, still the same error. <add^_>reran autogen, configure and make (with a clean) <tupi>add^_: are you trying to install guildhall or developping something that you'd like to guild host ? <mark_weaver>add^_: maybe you have something from spells in your path, that's overriding something important from guildhall? or maybe if you installed spells, you overwrote some important part of guildhall? <add^_>I don't have spells installed <add^_>I have both spells and guildhall in the directory (I didn't use to though) <civodul>mark_weaver: guildhall derives from Dorodango, which uses Spells <foeniks>is guildhall something like ruby gem for guile? <foeniks>is there a central or default repository <foeniks>ah, I mean for the package list, not the source ;) <add^_>It's in filesys it goes wrong at least *civodul discovers the blog *tupi think that guild rocks and hoped guildhall be further developed, so it could be used to install guile S/W that depends on C libraries, not just scheme, like guile-gnome, guile-clutter, kisê ... but ijp got the Tree-IL, compiler ... fever, no one knows if there is a cure :) let's just hope... <add^_>I'll try again with the "normal" way to do it.. *civodul points Gnus at gwene.com.shift-reset <add^_>I just need spells, not anything else. Why is it so hard to get it to work? <add^_>I'll try the thing you said mark_weaver <add^_>Is there a faster way of doing that? <add^_>Tried with some files (the needed ones) but didn't work <add^_>Nevermind, I'll continue tomorrow