freenode/#lisp - IRC Chatlog
Search
3:49:48
drmeister
I posted an issue and I wanted to pick your brain. https://github.com/slime/slime/issues/530
3:50:49
drmeister
When using C-c C-c to compile a form slime has the line number of the start of the code - but for some reason doesn't pass it to swank-compile-string.
5:59:26
drmeister
Hi - yeah - I dropped a note in there. I haven't seen any responses yet. But it's only been about 12 hours. No worries.
6:01:23
drmeister
It's a small change that we are looking for that may be a big change if adding another keyword argument to swank-compile-string requires every implementation to handle that.
6:21:44
pjb
drmeister: you can also use C-c C-k to compile the whole file and have the source location information kept.
6:22:56
drmeister
Yes - that's what I do - but every now and then I C-c C-c and it all gets confused.
8:45:03
thijso
he was just checking if anyone was here, so if not, he could steal the silverware... ;)
9:03:33
beach
Ae_Mc: Several of us meet once a year at the European Lisp Symposium, so many of us know each other. There are usually nearly 100 participants at that conference.
9:05:31
beach
Ae_Mc: ELS2020 has not been announced yet, but we happen to know that it is going to be held in Zürich, probably at the end of March.
9:09:20
beach
... and the steering committee. And especially by the program chair and the local chair, that are also different each year.
9:19:04
ck_
Can anyone help me with an ABCL question? I think I'm missing something very basic on the java side.
9:21:02
ck_
The problem is "Don't know how to REQUIRE ABCL-CONTRIB." -- I'm trying to run this code: https://gist.github.com/fsmunoz/5850260
9:21:45
ck_
I think I set the classpath correctly (tried both the -cp argument and the environment variable), but the messages "Added jar:file:..." that those files comments talk about don't happen
9:26:46
jackdaniel
could you paste somewhere exact invocation (copy paste from the terminal) and contents of: a) abcl directory, b) directory with abcl-contrib (possibly the same directory)
9:28:02
ck_
java -cp ~/opt/abcl-bin-1.5.0/abcl.jar:~/opt/abcl-bin-1.5.0/abcl-contrib.jar:org.eclipse.paho.client.mqttv3.jar org.armedbear.lisp.Main --noinform --noinit --nosystem --load abcl-mqtt.lisp
9:29:41
ck_
the directory contents of ~/opt/abcl-bin-1.5.0 are the following 6 files: "CHANGES" "README" "abcl-1.5.0.pdf" "abcl-contrib.jar" "abcl.jar" "asdf.pdf"
9:32:28
ck_
jackdaniel: okay I'll try (and I think I did before) -- but the "Don't know how to ... -CONTRIB" messages is produced from the abcl repl, so abcl.jar is found correctly
9:33:08
jackdaniel
I gather that when you invoke repl and type (require 'abcl-contrib) the problem is the same?
9:33:35
ck_
yes -- well, not exactly, because the quote syntax produces a different error message -- but it fails either way
9:35:56
jackdaniel
I think that abcl does not work *at all* on some recent Java versions because bytecode has changed and compiler has not been adapted? I'm not certain, maybe interpreter works there
10:20:58
ghard
Gday! Prior to pulling the CFFI sock up libcrypto, I'd like to check if anybody's seen a working ECDSA signing library out in the wild.
10:22:12
ghard
Asking for a friend who's got to be verifying the darned Apple Sign On JWS pronto. ;)
10:24:39
no-defun-allowed
https://gitlab.com/cal-coop/netfarm/netfarm/blob/master/Code/Crypto/keys.lisp might be good as an example of how to use it.
11:39:18
ym
Can't find mention of ChrysaLisp in logs. Wasn't discussed yet or I'm doing something wrong?
11:45:27
no-defun-allowed
https://github.com/vygr/ChrysaLisp/blob/master/apps/films/app.lisp This indentation scares me.
11:46:15
ck_
loading cl-async through quicklisp fails with "#:libuv-grovel is unbound" -- is that expected?
11:55:07
ym
no-defun-allowed, yes, but I'm more interested in bare-metal implementations. Mezzano seems cool though.
11:58:33
no-defun-allowed
It runs on an x86_64 machine, the only non Lisp part being the bootloader.
12:02:07
no-defun-allowed
And the CADR runs Lisp Machine Lisp, which is not awfully far from Common Lisp as things go.
12:04:41
ym
CADR is for retro fanboys. Variation of RISC-V and SBCL optimized for each other - that's perfect case.
12:10:29
no-defun-allowed
So you want a C machine that has been adjusted slightly to run SBCL, which in turn is adjusted to run on that C machine better?
12:12:10
no-defun-allowed
Maybe I am too pessimistic, but if I was able to fab my own processor, I would definitely have it run a more Lisp-friendly instruction set.
12:12:29
no-defun-allowed
And I am never going to be able to fab my own processor, so we have to settle on running Lisp well on C machines.
12:19:40
jackdaniel
praise the stock hardware: it gives you a reasonable price which individual can afford with good performance disregarding the language you chose
12:22:07
no-defun-allowed
It wouldn't be anywhere close to a proper fab job in terms of efficiency or performance.
12:24:52
aap
there's a project to make a reincarnation of the lambda that executes macrocode directly from C instead of emulating the microcode
12:25:09
aap
once that is running lisp code, maybe the macrocode could be executed directly from an fpga perhaps
12:33:16
jackdaniel
behold the prophecy: one day we'll have stock hardware with powerful fpga boards being their peripherals (similar to zynq) and we'll synthesize hardware at real time