IRC channel logs


back to list of logs

***modula is now known as defaultxr
<lampilelo>RhodiumToad: i did some more research on this match issue and submitted a bug report with your patch attached as a diff, bug 49707
<tohoyn>what's the status of "warnings as errors" in guile? it would be useful for checking large applications.
<tohoyn>andrzejku: hello
<civodul>tohoyn: currently there's no way to treat warnings as errors
<andrzejku>I am looking for guile friends here
<lampilelo>sneek: seen friends
<sneek>Not as far as I can remember.
<lampilelo>that's depressing
<tohoyn>andrzejku: Have you been programming with guile?
<andrzejku>tohoyn, just a little bit
<andrzejku>tohoyn, I did some exercises
<tohoyn>andrzejku: I'm developing an own programming language with guile
<andrzejku>tohoyn, but I have no Guile friends =/
<andrzejku>tohoyn, cool
<tohoyn>andrzejku: it's basically "Scheme with types"
<andrzejku>tohoyn, I am more interested in system programming
<tohoyn>have to leave for a while
<andrzejku>sure no problem nice to meet you tohoyn
<dsmith>sneek: botsnack
<lampilelo>andrzejku: we're all friendly here, so ask if you have any questions
<dsmith-work>Happy Friday, Guilers!!
<sneek>I've been running for one day and 16 hours
<sneek>This system has been up 1 day, 22 hours, 44 minutes
<RhodiumToad>bah :-)
<manumanumanu>lampilelo: regarding upstream match.scm: no documentation mentions several patterns of not. Only guile's, it seems.
<manumanumanu>And since andy's patch of similar patterns got merged, I don't think upstream and the guile version diverges at all currently.
<lampilelo>manumanumanu: did you read my bug report?
<manumanumanu>Yes. to me it is obviously the documentation that is wrong
<manumanumanu>unless guile should diverge from upstream
<manumanumanu>there is a bug in upstream chibi regarding it, though:
<manumanumanu>which you might have mentioned
<manumanumanu>(not 1 2) is not the same as (not (or 1 2))
<manumanumanu>try submitting it to the chibi issue tracker, maybe.
<lampilelo>i think we should change the code, not the documentation, and try to push the changes upstream, since that's how the paper on match and both srfis define it
<lampilelo>they define it as "it matches if none of the patterns match", (not (or 1 2)) = (and (not 1) (not 2))
<lampilelo>i'm paraphrasing
<lampilelo>and yes, i think i should submit it to chibi, but first i wanted to get the opinions of more people
<manumanumanu>I have no say in the matter, but discuss upstream first and then diverge from upstream if no action is taken is probably the best solution.
<manumanumanu>lampilelo: I love this message in the srfi-204 mailing list:
<lampilelo>yeah, maybe that would be the way, i'll do that, although i'd say that in guile we don't have all the changes they've made over the last few years so we diverge a bit from the upstream already