beneroth changed the topic of #picolisp to: PicoLisp language | The scalpel of software development | Channel Log: https://libera.irclog.whitequark.org/picolisp | Check www.picolisp.com for more information
oldf8l is now known as f8l
<abu[7]> Cool tankf33der! Whas it a problem with the right library path environment?
<tankf33der> Correct, and the same relocation missing before. And correct dynamic library creations for ht and ext
<abu[7]> great
<gahr> hi - would you be interested in making it simpler to adjust the installation prefix?
<abu[7]> Hi gahr, you mean the "@" before path names?
<gahr> now /usr/lib and /usr/bin are hardcoded in the scripts shebangs
<gahr> no, the @ adjusts itself pretty nicely :)
<abu[7]> There are just a few helper scripts, not very critical I think
<gahr> ok
<abu[7]> (I don't use them ;)
<abu[7]> What is your idea to make it simpler?
<gahr> most 3rd party softwre take a PREFIX as a make variable
<gahr> and adjust stuff based on that
<gahr> we could sed the scripts and also provide an install target based on that
<gahr> I am packaging picolisp to freebsd, and the global installation method doesn't fit very well: we don't have a local installation that sticks around so we can point to it, we effectively need to copy files from the build dir to /usr/local
<abu[7]> I think the only script is bin/vip
<gahr> psh, pty, picolisp
<gahr> whatchdog
<gahr> they all point to /usr/bin and/or /usr/lib
<gahr> people.freebsd.org/~gahr/picolisp-freebsd-port-Makefile <-- this is what I have to install on FreeBSD
<gahr> I was looking to upstreaming part of the logic, if possible
<gahr> (see the do-install target)
<gahr> and the post-patch
<gahr> well, everything from line 20 on is to adjust paths and install stuff
<abu[7]> ok, let's think about it later. I have a meeting in 5 mins
<gahr> sure
<gahr> thanks
<abu[7]> thank you!
isaneran has joined #picolisp
<tankf33der> abu[7]
<tankf33der> $ pil +
<tankf33der> "Function not implemented" -- IP socket
<tankf33der> : (port 0 "L")
<tankf33der> ?
<tankf33der> got this error on muslc system, glibc is ok
<tankf33der> i am using muslc system for years, maybe rename failed somehow?!
<tankf33der> worked several days ago
<gahr> : (port 0 "L")
<gahr> !? (%@ "closeOnExec" NIL (cons T (up)) Sd)
<gahr> Bad ffi
<tankf33der> freebsd ?
<gahr> yep
<gahr> ah wait
<gahr> : (port 0 "L")
<gahr> -> 3
<gahr> works on the "installed" pil
<tankf33der> what is difference between two outputs ?
<gahr> first one was from a local installation, second one is from the global installation I'm testing out
<gahr> not sure what was the tarball for the local installation anymore, let me figure it out
<tankf33der> strange
<gahr> re-downloaded the tarball from the rolling release
<gahr> fine now
<gahr> : (port 0 "L")
<gahr> !? (set "Var" (+ (>> -8 (byte (+ Addr sin6_port))) (byte (+ Addr sin6_port 1))))
<gahr> -> 3
<gahr> : (port 0 "L")
<gahr> 40675 -- Variable expected
<gahr> ? (port 0 "L")
<gahr> -> 5
<gahr> ? (port 0 "L")
<gahr> !? (set "Var" (+ (>> -8 (byte (+ Addr sin6_port))) (byte (+ Addr sin6_port 1))))
<gahr> 33972 -- Variable expected
<gahr> is this expected?
<tankf33der> (port 0 "L")
<tankf33der> -> <number>
<tankf33der> should like this
<gahr> it works every other time it seems
<tankf33der> run tests
<tankf33der> pil @lib/test.l +
<gahr> I see now picolisp is listening on 4 ports
<tankf33der> ok
<gahr> including 40675 and 33972
<gahr> tests used to work fine
<gahr> yep, OK
<tankf33der> : (port 0 "L")
<tankf33der> "Function not implemented" -- IP socket
<tankf33der> comes from torsocks, issue closed. this is problem on my side
<gahr> :)
chexum has quit [Ping timeout: 240 seconds]
chexum has joined #picolisp
<abu[7]> ret
gahr has quit [*.net *.split]
Abhishek_ has quit [*.net *.split]
gahr has joined #picolisp
Abhishek_ has joined #picolisp
calle has joined #picolisp
calle has quit [Ping timeout: 244 seconds]
calle has joined #picolisp
<gahr> am I correct that pil doesn't use llvm at runtime, only to build itself?
<abu[7]> It may be used by loading @lib/clang.l
calle has quit [Ping timeout: 240 seconds]
<gahr> I see
<abu[7]> Seldom used I think
calle has joined #picolisp
calle has quit [Ping timeout: 260 seconds]
<gahr> yup I'm looking at the example under the (lisp) doc in the reference
<abu[7]> It is quite heavily used in the "rcsim" flight simulator
<gahr> wow impressive!"
<gahr> you don't ship rcsim in misc anymore?
<gahr> ah I see, separate distribution
<abu[7]> Yeah, it should not be in the distro. Not generally useful
<abu[7]> rigxt, rcsim.tgz
<abu[7]> But I think I should change the keyboard interface again
<abu[7]> It is now targeted for Penti
<abu[7]> Problem is that I cannot test with a normal keyboard] I don't have one
<gahr> penti is as geek as it gets :)
<gahr> do you actually use it daily?
<abu[7]> yes, always, I don't have a normal PC
<gahr> impressive
<abu[7]> All on smartphone, or the smartphone as an input for my headless server in the basement with the TV-Set as display ;)
pablo_escoberg has joined #picolisp
<gahr> wow
<pablo_escoberg> gahr agreed, that is very impressive. abu[7] about how long do you think it would take to become proficient with penti?
<abu[7]> A few hours to get the letters, a few days for all punctuation, alt an fun keys, but several weeks to get it into muscle memory
<pablo_escoberg> that's not too bad.  any idea how fast you type on it?
<abu[7]> I'm faster now than on a normal keyboard, but t is also because I got slower on normal keyboads
<abu[7]> So in reality touch typing with 10 fingers is faster
<pablo_escoberg> hmmmm... probably still worth the training for mobile at least...
<abu[7]> I think so
<abu[7]> The reason is not speed, but convenience
<abu[7]> Needs just one hand
<pablo_escoberg> right, and I imagine I'll keep using a regular keyboard for my laptop, but it's probably worth it just to avoid those awful screen keyboards on phones.
<abu[7]> T
<abu[7]> ok, reverted rcsim.tgz to the old keyboard version
pablo_escoberg has quit [Quit: Client closed]
pablo_escoberg has joined #picolisp
<abu[7]> Thanks tankf33der :)
<abu[7]> I wonder why the video in the Wiki article does not work
<abu[7]> Ah, in Tor browser it works
<abu[7]> not in Chrome it seems
pablo_escoberg has quit [Quit: Client closed]
isaneran has quit [Ping timeout: 256 seconds]
pablo_escoberg has joined #picolisp
avocadoist has joined #picolisp
pablo_escoberg has quit [Ping timeout: 246 seconds]
calle has joined #picolisp
calle has quit [Ping timeout: 245 seconds]
calle has joined #picolisp
calle has quit [Remote host closed the connection]