freenode/#lisp - IRC Chatlog
Search
22:28:31
ZigPaw
aeth you can run X Server on windows and have WSL apps using this (MobaXTerm works best with latest emacs snapshot).
22:54:20
drunk_foxx[m]
But honestly this intro video tells almost nothing about the language, I bet it could have been done better. But whatever
23:35:20
didi
Heh, I'm having a brain fart. So (loop for fn in (loop for i to 3 collect (lambda () i)) collect (funcall fn)) evals to (4 4 4 4), but I want (0 1 2 3), i.e. I want the thunk to have the value of 'i' when it was created, not a reference to 'i'. What to do?
0:28:05
stylewarning
drunk_foxx[m]: the first take of the video did everything from defining the syntax of S-expressions and etc etc
0:54:40
Xach
didi: https://www.youtube.com/watch?v=svmPz5oxMlI - i'm not familiar with the channel but it has a big audience
2:40:51
ealfonso
is there something better than: (format log-fh "~A~A~A~A~A~A~A~%" ip #\tab date #\tab user-agent #\tab path)
2:49:10
ealfonso
I heard of this http://www.sbcl.org/manual/Statistical-Profiler.html which gives me 404
3:03:04
PuercoPop
additionally when you install SBCL you can build the docs as texinfo, which can be read from inside Emacs (C-h i)
3:09:58
pierpa
ealfonso: you could abstract and give a name to the action "print a few things separated by tabs and then swithch to a new line"
3:16:58
ealfonso
pierpa yeah, I was just wondering if there's a built-in way with format or another standard way
3:20:58
pierpa
Not sure if there's a format trick for doing it in an obfuscated way saving one or two keystrokes. But even if, don't. :)
3:23:25
pierpa
you surely won't like it, but FWIW, here's how I would do it: (defun print-line-tab-separated (where &rest what) (princ (first what) where) (loop for item in (rest what) do (princ #\tab where) (princ item where)) (terpri))
3:34:04
mange
ealfonso: Would (format log-fh "~@{~a~^ ~}" ip date user-agent path) do what you want? (Note the literal tab in the format string.)
4:55:30
jasom
on_ion: Other than initialization and garbage collection, all of sbcl is written in CL.
5:09:40
pillton
I think it would be difficult to integrate an independent implementation of CLOS with the host's implementation of typep; assuming the host doesn't have CLOS already.
5:15:33
jackdaniel
pillton: CLOS is integral part of ECL now, but years ago (before I even knew about the project) it had option to build it without CLOS – type implementation there is independent
5:32:03
theemacsshibe[m]
it's all good but to handle improper lists like `(x . xs)` i need to add a conditional setf which i don't like very much
5:38:43
mange
theemacsshibe[m]: At the moment if you call (p-match 'a 10) you get back '(a . 10), but if you call (p-match '(a) '(10)) you get '((a . 10)). That's surprising to me.
5:39:07
mange
If you make it so p-match always returns a list of matches, I think you can also clean up your conditional setf, and the conditional after it.
5:41:27
theemacsshibe[m]
aaaaactually given i want to put this on an environment alist there could be a better solution
5:48:13
theemacsshibe[m]
my end goal is to port it to cee to get it working in my interpreter for lambdas