IRC channel logs
2025-01-13.log
back to list of logs
<carloratm>hi all, at 40yo I finally decided to learn a lisp, I opted for guile. Am I too old? <UncleRRR>You only need 10 minutes to start coding in scheme. Several hours later you would be able to play in guile (with ai coding assistant or google search available by hand)! <sneek>Welcome back UncleRRR, you have 1 message! <sneek>UncleRRR, mdj says: There *is* core support for "void" in Guile. You can do (lambda () *unspecified*). However. the if construct is more portable, as rlb said (and the compiler makes sure nothing is "done" at run time). <UncleRRR>I even heard a 70 years old man in Japan learnt to draw in Excel. Ten years later he became a master! <dsmith>carloratm, Scheme is one of the most beautifully simple and elegant langages. <UncleRRR>For game dev, checkout chickadee. For html site generating, checkout haunt. For hosting simple site, checkout artinis. Don't miss guix the package manager as well! These are the most interesting and simple stuff in my eyes, for anyone who would like to start without a determined mind for digging into macro or standing at the very front line exploring. <lechner>Hi, may I call Glibc's malloc(3) and free(3) from Guile code when a shared library interface requires me to do so? <carloratm>is the repl I get from Debian the right place to start? I am indeed to old for emacs <carloratm>or is there a better repl around? Asking because it doesn't seem to support readline? <stevelitt>carloratm, there's a gnu program that gives readline capability to any software. I forgot the name offhand --- let me try to find it... <sneek>dthompson, you have 1 message! <sneek>dthompson, ArneBab says: can the checks in define-typed be stripped by the compiler if a procedure is only used in places where the types have already been checked? Or does that require inlining? <dthompson>because readline is gpl and guile is lgpl (ugh) <dthompson>the manual tells you how I'm pretty sure but can't look it up rn <lechner>ACTION always thought people are too young for Emacs <stevelitt>(define x 1)(define x 2)(display x)(newline) ; prints 2, why? <stevelitt>I thought if I change a variable with an existing value to another value I had to use set! <ArneBab>stevelitt: because on top level you can redefine x. <stevelitt>I also verified that set! for the redefinition works just fine, so I think I'll use set! for any redefinition, regardless of level. <dsmith>I've heard in some scheme's, you can set! a definition into existance without define <lilyp>in emacs lisp you can certainly setq variables into existence <rgherdt>dsmith: yes, that's the case of chicken <rlb>Looks like test-out-of-memory no longer crashes with -fstack-protector-strong on some architectures (e.g. ppc64el). No idea how long that's been true, but it's true now. <rlb>I've pulled the 32-bit related commits after v3.0.10 on main up to and including "Run sigbits fixpoint based on use/def graph, not cfg" (almost all from wingo) onto v3.0.10, and that does get 3.0.10 working on armhf, so I think I might upload that to debian experimental, and if it builds everywhere, restore 3.0.10 to debian unstable. <rlb>Please let me know if you think that might be a bad idea somehow. <rlb>civodul: (wrt earlier comments) ^ <rlb>If not, then suppose we'll just see how it goes. <civodul>rlb: 32-bit was one issue in 3.0.10, but there was another one related to dealing with .go files produced by 3.0.9 <civodul>(the ideal option would be to release 3.0.11)