habamax has quit [Quit: ERC 5.4 (IRC client for GNU Emacs 28.2)]
shka has quit [Read error: Connection reset by peer]
shka has joined #commonlisp
soundmodel has joined #commonlisp
soundmodel has quit [Client Quit]
soundmodel has joined #commonlisp
soundmodel has quit [Ping timeout: 246 seconds]
Lord_of_Life_ has joined #commonlisp
Lord_of_Life has quit [Ping timeout: 246 seconds]
Lord_of_Life_ is now known as Lord_of_Life
pranavats has joined #commonlisp
dra has joined #commonlisp
dra has joined #commonlisp
dra has quit [Changing host]
habamax has joined #commonlisp
thollief has joined #commonlisp
awkravchuk has joined #commonlisp
dre has joined #commonlisp
occ has quit [Ping timeout: 246 seconds]
Gleefre has joined #commonlisp
dre has quit [Quit: Leaving]
MajorBiscuit has quit [Ping timeout: 240 seconds]
MajorBiscuit has joined #commonlisp
Brucio-61 has quit [Quit: Application exit]
random-nick has joined #commonlisp
kenran has joined #commonlisp
Brucio-61 has joined #commonlisp
mgl has quit [Ping timeout: 246 seconds]
tyson2 has joined #commonlisp
bird_ has quit [Ping timeout: 240 seconds]
euandreh has quit [Quit: euandreh]
jrx has joined #commonlisp
<jrx>
Hi. Is it possible to see the current values of declarations set by PROCLAIM, in a portable way ? I know there is (describe-compiler-policy) in SBCL. But I'm using Clozure common lisp...
<ogamita>
jrx: some implementations may have something (perhaps sbcl has some API to do that).
<Nilby>
jrx: you can use (uiop:get-optimization-settings) but it's not standard
<ogamita>
And if you have a look at the source (type proclaim M-. in the REPL or a lisp buffer), you can see that no bookkeeping is performed by ccl.
<ogamita>
and it's only for optimization.
doulos05 has joined #commonlisp
<ogamita>
But it seems to work at least on ccl.
<jrx>
ok thanks.
<doulos05>
Hello! It's been a while. I thought I remembered there being a channel specifically for McCLIM, but I didn't see it. Is here also a good place to ask about that?
<jackdaniel>
doulos05: #clim
<doulos05>
Ah, I had #clim. So close!
<jackdaniel>
:)
<jrx>
Nilby: (uiop:get-optimization-settings) works great on CCL ! thx.
<jrx>
Nilby: on SBCL too !
<jrx>
I consider it "standard" now :)
<Nilby>
if it weren't for many such defacto standands we'd be in big trouble
alendvai__ has joined #commonlisp
euandreh has joined #commonlisp
kenran has quit [Remote host closed the connection]
mgl has joined #commonlisp
sloanr has joined #commonlisp
habamax has quit [Quit: ERC 5.5.0.29.1 (IRC client for GNU Emacs 29.0.92)]
tyson2 has quit [Remote host closed the connection]
<ixelp>
[Sbcl-bugs] SBCL seems to alter (PROCLAIM (OPTIMIZE)) declaration values after LOAD | Steel Bank Common Lisp
jeffrey has joined #commonlisp
tyson2 has joined #commonlisp
<bike>
jrx: that's normal and expected. it's not "implicitly" affecting optimize, since you wrote the proclaim right there. and if you look on the clhs page for the related declaim, it says it's undefined whether it persists after compilation
ufi has joined #commonlisp
markb1 has joined #commonlisp
<|3b|>
DECLAIM is talking about compile-time side effects, which PROCLAIM doesn't have, and the example uses LOAD anyway
<ixelp>
[Sbcl-devel] Optimization policy and LOAD | Steel Bank Common Lisp
<|3b|>
apparently inherited from cmucl
<bike>
i know, i just meant it was analogous
* |3b|
was arguing that it is not :)
<|3b|>
DECLAIM adds compile-time side-effects to the normal effects, which like other compile-time side-effects might disappear after compilation, while the normal ones remain
<bike>
what i mean is, if even compile-file isn't guaranteed to not persist the special compile time effects, why would load be guaranteed to not persist the normal effects (of proclaim)
<bike>
proclaim is just a side effect to the compiler controls.
<bike>
you certainly wouldn't expect a special proclamation to be undone after load, so why optimize?
* |3b|
is confused if you are saying sbcl's behavior is ok or non-conformant now
<bike>
i'm saying it's okay.
<|3b|>
so you wouldn't expect optimize to be undone after a load, but it is ok that it is?
<|3b|>
it is a somewhat nice behavior, since it keeps random libs from making global changes to optimize settings, but is also pretty easy to argue is non-conformant
ElKowar has quit [Server closed connection]
ElKowar has joined #commonlisp
<|3b|>
less nice when it makes it more difficult for people to make those changes for themselves though
benderst` has joined #commonlisp
bendersteed has quit [Ping timeout: 240 seconds]
benderst` has quit [Client Quit]
zxcvz has joined #commonlisp
zxcvz has quit [Client Quit]
jonatack has quit [Ping timeout: 240 seconds]
jonatack has joined #commonlisp
tyson2 has quit [Remote host closed the connection]
cage has joined #commonlisp
dino_tutter has quit [Ping timeout: 250 seconds]
ufi has quit [Quit: Leaving]
Noisytoot has quit [Ping timeout: 240 seconds]
pranavats has left #commonlisp [Error from remote client]
Noisytoot has joined #commonlisp
alendvai__ has quit [Ping timeout: 246 seconds]
thmprover has joined #commonlisp
ufi has joined #commonlisp
pranavats has joined #commonlisp
occ has joined #commonlisp
jrx has quit [Quit: See You !]
ufi has quit [Quit: Leaving]
ufi has joined #commonlisp
karlosz has quit [Read error: Connection reset by peer]
<bmansurov>
Hey all, I'm unable to login at https://bugs.launchpad.net/sbcl to report a bug (I'm getting "Sorry, something just went wrong in Launchpad."). I think I've found a bug with SBCL. Here's a simple use case (the directory /home/user/trash exists and is empty) :
<bmansurov>
The error I'm getting is this:
<bmansurov>
Failed to find the TRUENAME of /home/user/a.txt:
<bmansurov>
No such file or directory
<ixelp>
Bugs : SBCL
<bmansurov>
The file /home/user/a.txt also exists.
<bmansurov>
Can someone please report it if they can reproduce?
azimut has joined #commonlisp
<bmansurov>
The same code works on ECL.
occ has quit [Ping timeout: 246 seconds]
tyson2 has quit [Remote host closed the connection]
dcb has joined #commonlisp
pve has joined #commonlisp
Oladon has joined #commonlisp
<Shinmera>
I can't reproduce. Does /home/user/trash/ exist?
<Shinmera>
and does the program have access rights to it?
<Shinmera>
is there symlink stuff involved somehow? are those the *actual* pathnames you're using, or have you substituted stuff
<Shinmera>
because things like [], {}, etc in pathnames will do different things.
<Gleefre>
I can't reproduce it either. The error suggests that a.txt doesn't exists, are you sure it is still there?..
<Shinmera>
eg in bcl #p"[ab].txt" is a wild pathname that can refer to a.txt and b.txt
<Shinmera>
*sbcl
<bmansurov>
Yes, both a.txt and trash exist. Renaming works if I just change the filename and not the directory.
<bmansurov>
This works: (rename-file #P"/home/user/a.txt" #P"/home/user/b.txt")
<Shinmera>
is trash a mount to a different device?
<Shinmera>
because moving across devices is not permitted by the standard
<bmansurov>
no, just a folder, I created it with mkdir ~/trash
<bmansurov>
Fyi, SBCL 2.3.5
<Shinmera>
no idea then.
<bmansurov>
What is your SBCL version, Shinmera?
<Shinmera>
same
<bmansurov>
OK, thanks for debugging. Something must be off with my setup.
<Shinmera>
Probably. I don't think this part of SBCL has changed recently, either.
sloanr has quit [Remote host closed the connection]
<Shinmera>
All it does is basically just TRUENAME on the pathname, and then syscall to rename.
<bmansurov>
I see.
<Shinmera>
well, sorta. rename-file has weird semantics where it first needs to merge-pathnames on both arguments
<Shinmera>
which annoys me a lot because it makes it basically impossible to strip the type
<AkashaPeppermin4>
for proclaim workaround you can try
<AkashaPeppermin4>
it does break some libraries though i think notably ppcre and some of Shinmera's libs
<Shinmera>
Really?
<AkashaPeppermin4>
i don't know, it's off a hazy memory
ufi has quit [Quit: Leaving]
<Shinmera>
I know it makes some stuff so slow it's practically unusable (cl-opengl) but I'd be surprised if it had any semantic differences.
ufi has joined #commonlisp
ufi_ has joined #commonlisp
<AkashaPeppermin4>
very likely wasn't your libs specifically but a dependency i just wasn't very proficient at pinpointing at that time
<Shinmera>
hm
<|3b|>
might break things that depend on tail-call elimination
<Shinmera>
ah, true.
<AkashaPeppermin4>
also i just realised i have it in my sbcl-init.lisp and ppcre is loading so maybe it was fixed already
<|3b|>
and probably stresses stack a lot more in general
ufi has quit [Client Quit]
ufi_ has quit [Client Quit]
ufi has joined #commonlisp
ufi_ has joined #commonlisp
<AkashaPeppermin4>
yes definitely removes tail call optimization though it's pretty useful seeing tons of backtrace info when something goes wrong
<AkashaPeppermin4>
also i think you can call sb-ext:restrict-compiler-policy again to change it to a different setting
<AkashaPeppermin4>
i think at one point I deleted them from my .sbclrc and put them in my project's ASD with #+sbcl
<Shinmera>
I'm usually fine enough with just C-u C-c C-c when I need it.
<AkashaPeppermin4>
me too when making my own code but lately i've been looking at some legacy code not written by myself and i have no idea how anything works and there's still bugs so i just leave debug at max by default
attila_lendvai_ has quit [Read error: Connection reset by peer]
Catie` has joined #commonlisp
attila_lendvai_ has joined #commonlisp
<waleee>
nice thanks for the tip
Catie has quit [Ping timeout: 246 seconds]
shka has quit [Ping timeout: 260 seconds]
Oladon has joined #commonlisp
alcor has quit [Remote host closed the connection]
dra has quit [Quit: Leaving]
dra has joined #commonlisp
dra has quit [Changing host]
dra has joined #commonlisp
Jach has quit [Ping timeout: 245 seconds]
White_Flame has quit [Remote host closed the connection]
White_Flame has joined #commonlisp
Jach has joined #commonlisp
attila_lendvai_ has quit [Ping timeout: 252 seconds]
<Oladon>
What does it mean when I run (sb-ext:gc :full t) and the program's memory usage goes down by about 50%?
<Oladon>
I.e. what does it indicate about what I'm doing, or not doing, that I shouldn't (or should) be?
dino_tutter has quit [Ping timeout: 246 seconds]
m5zs7k has quit [Ping timeout: 246 seconds]
m5zs7k has joined #commonlisp
<_death>
it means there was garbage to collect.. it indicates that your program allocated memory and didn't hold on to all of it.. on its own it doesn't say anything about what you should or should not do
NotThatRPG has joined #commonlisp
pve has quit [Quit: leaving]
habamax has joined #commonlisp
<NotThatRPG>
I don't *think* there's an answer to this question, but are the NAMESTRING functions specified to yield a string that does *not* need quoting to the host OS?
jmdaemon has joined #commonlisp
<NotThatRPG>
I can create cases where (run-program (format nil "mkdir -p ~a" (namestring pn))) works, and (run-program (format nil "mkdir -p '~a'" pn)) but (run-program (format nil "mkdir -p '~a'" (namestring pn))) does NOT work
<_death>
it seems you conflate two issues..
prokhor has quit [Ping timeout: 246 seconds]
<NotThatRPG>
So ... should I always assume that the results of NAMESTRING are OK without quoting? I don't see anything in the spec that makes such an assertion
Catie has joined #commonlisp
<_death>
first, namestring is not guaranteed to result in a name that can be passed to your OS primitives.. for that, there's say uiop:native-namestring
rgherdt has quit [Remote host closed the connection]
<_death>
then, your example indicates that you want to use a shell to run a command.. and for that you'll also need to escape arguments for the shell
<_death>
command
<NotThatRPG>
The interesting issue here is that the escaping done by NAMESTRING actually (sometimes, on SBCL) yields a pathname that *cannot* be quoted to the shell.
Catie` has quit [Ping timeout: 260 seconds]
<NotThatRPG>
Using NATIVE-NAMESTRING and avoiding the quoting (at least on SBCL on my Mac) works. I'll see how it works on other implementations.
<NotThatRPG>
I've never quite figured out how to use RUN-PROGRAM w/o the shell -- don't I have to predict the precise location of any binary to do that?
prokhor has joined #commonlisp
NotThatRPG has quit [Remote host closed the connection]