freenode/lisp - IRC Chatlog
Search
20:36:29
rumbler31_
it goes all the way down to two with-open-file macros and just copies the a buffer at a time into the output stream, and if
20:58:42
rumbler31_
derp tried to make a bundle that didn't include quicklisp that includes a function to call quicklisp bundle.... dummy <-----
21:00:02
rumbler31_
I want to make a bundle of a project, I guess none of the functions I wrote depend on the original package
21:00:22
rumbler31_
seems kinda hacky to just make another package that assumes things about another, like file paths and project names
21:06:05
Nilby
Since Quicklisp can't really be an asdf or quicklisp dependency, then unless you always install it as part of your own installation, things have to check for it manually before using it.
21:23:36
Nilby
Yes, but I'm pretty sure one might get different results loading it in it's recommended way, vs. just asdf loading it.
21:46:56
monaliza
Heyo! I've recently learned common lisp and i've read a book about it, I've been kind of indecisive for which other vlafor of lisp i should start using next, because of how many flavours there are. I'm mostly really interested in functional programming, and I'm a real fan of haskell-ish syntax, where it's short, makes sense, and works rather well. I've heard about Racket lisp, Arc lisp, and Scheme,
21:46:59
monaliza
and I'm getting mixed signals on which one I should try to extensively dive into first
21:49:16
moon-child
monaliza: you are unlikely to get very many responses here, because this channel is specifically about common lisp. That being said: I would avoid arc. Racket and scheme are both nice enough, and you may also want to look at coalton
21:51:05
monaliza
My fault for not reading the first word of the topic, I'm very sorry! And, thank you very much!
21:57:40
kagevf
use CL so you have the flexibility to take advantage of other styles when FP isn't a good fit
21:59:00
kagevf
also, if you want FP only no matter what, maybe consider clojure ... with the caveat that it's on the jvm
22:35:15
no-defun-allowed
m00natic: IIRC {} and [] in Scheme produce equivalent syntax to (), so you could write (let ([x 1]) x) in Scheme or even [let [[x 1]] x]
22:38:06
moon-child
seems chicken does exactly as you say, guile doesn'tlike {}, and s7 doesn't like either
23:01:14
edgar-rft
the beatiful thing with Scheme is that there are so many *different* syntax schemes you can choose from
23:35:47
kagevf
in CL could you write a reader macro to do the same thing? to make [] and {} act like ()?
23:36:40
kagevf
ok, that's what I thought ... do they exist as part of the standard, or in a library?
23:37:28
fiddlerwoaroof
There's also things like this that port the good bits of Clojure's syntax: https://github.com/joinr/clclojure/blob/master/reader.lisp
23:38:11
fiddlerwoaroof
So, technically, libraries aren't supposed to use these characters in the standard readtable
23:38:38
fiddlerwoaroof
However, libraries are free to provide their own readtables, and named-readtables exists to make that process nicer
23:39:06
fiddlerwoaroof
Also, personally, I tend to think "whatever a library can do should be left out of the standard as much as possible"
0:53:42
Bike
{} and [] are reserved in the sense that the implementation can't use them. there's no restrictions on programmer users, library writers or not. named readtables is probably still a good idea tho.
1:24:33
aeth
You'd probably want #[...] and #{...} instead of just [...] and {...} to make it really clear that it's a reader macro
5:45:33
fiddlerwoaroof
Bike, I'm confusing two things, it's interesting that #{} and #[] aredescribed this way: "The combinations marked by an asterisk (*) are explicitly reserved to the user"
5:46:32
fiddlerwoaroof
While [] and {} are defined this way: "The characters marked with an asterisk (*) are initially constituents, but they are not used in any standard Common Lisp notations. These characters are explicitly reserved to the programmer."
5:47:47
fiddlerwoaroof
It seems to me that there's an attempt here to distinguish characters that end-users might define reader macros on from characters that non-end-user programmers might define reader macros on
5:48:19
fiddlerwoaroof
Of course, in retrospect, it seems to me that NAMED-READTABLES is the right solution to this sort of problems