freenode/lisp - IRC Chatlog
Search
3:06:06
torbo
Does slime have reader macro support? I notice that I'm able to use reader macros in sly but not slime. Is there something I need to configure or does slime just not support reader macros?
3:10:47
Oladon
torbo: it'd be helpful if you'd tell us what you're trying to do and what (exactly) is happening
3:12:36
torbo
If I try to run the code at the bottom of this page: https://gist.github.com/chaitanyagupta/9324402
3:13:26
torbo
By loading it directly into sbcl, and type {"a": 1} into the console, I get a hashtable object back.
3:17:30
torbo
Here's a direct link to the code in json-reader.lisp: https://gist.githubusercontent.com/chaitanyagupta/9324402/raw/54359d3f2d19b78c603cf011304a741410a873ae/json-reader.lisp
3:18:07
torbo
Nice. Maybe my configuration is broken somewhere. I'll have to investigate. Thank you for checking for me, I really appreciate it.
3:18:40
Bike
the slime repl just passes things to sbcl's read and eval. it should behave almost exactly the same as sbcl at a terminal, other than like formatting stuff.
3:24:37
torbo
Ah, yes, you're right, it does work. The reason it's broken is because I've loaded up another package before running that command and I called (in-package :other-package) first, and I think that in that package, I'm loading up other reader macros that may be affecting the readtable in some way.
3:26:55
torbo
I've been able to replicate the problem, and the expressions that cause the inconsistent state are the following:
3:32:26
loke
I was sure there was a way to export a symbol from one package into another package and a different name... I.e. so that (eq 'foo:a 'bar:b) ⇒ T
3:37:52
mfiano
Well you could use package-local nicknames, but it's non-conforming and only available on 5 implementations.
3:46:27
mfiano
Right, he's looking to map different symbol-names for a symbol internal to one package to another. READ would need to be modified for that behavior as to know which package's LUT to reference for the mapping.
4:03:47
loke
mfiano: The question originates from a larger question as to whether it's possible to create consistent aliases in Common Lisp. I.e. something where (fpp
4:04:17
LdBeth
But it might be more useful to let the compiler/eval handle that, if aliasing/renaming is desired
4:06:26
LdBeth
For example a macro involves the aliased function has been loaded before the aliasing definition
4:11:02
loke
LdBeth: If the same symbol could have different names, that would have solved that problem. Then symbols FOO and BAR would literally _be_ the same thing.
4:12:16
loke
I thought some variant of symbol aliasing was used as an argument why some people use keywords for loop commands?
4:13:16
loke
I'm OK with SYMBOL-NAME returning a name that is not consistent with its name in a package, and I thought that was possible. But apparently not.
6:07:34
afidegnum
i m trying to build a drop-n-down html builder a la dreamweaver with Js support, but i am confused with what GUI toolkit should i use, i read about McCLIM, Slip but i am a bit perplexed,
6:08:03
afidegnum
the GUI will have an internal browser engine to preview the page intead of opening an external browser,
6:09:40
loke
afidegnum: CLIM is quite opinionated in how you build your applications. If your design matches CLIM's idea of interaction (or if you're willing to adjust your ideas to match) then it's a good choice.
6:10:25
loke
You _can_ build anything of course. It's just that CLIM won't help you as much if you choose to go certain routes.
6:10:59
loke
I'm comitted to CLIM right now myself, as I help out with maintaining (and evolving) it.
6:11:49
afidegnum
Drag-DOwn feature of vectors which will generate DOM nodes which can be previewed in a customed web browser, so will be the Js
6:16:53
LdBeth
afidegnum: afaik qtwebengine is licensed under a BSD-3 clause style license which doesn’t prohibit either modify or redistribute
7:39:14
p_l
afidegnum: you might be also interested in running chrome or firefox in debugger-driven mode with external debugger (your application)
8:09:13
jackdaniel
fun stuff, one asdf function has 184 arguments (combineds, with default initargs included)
8:15:14
jackdaniel
I don't know, I'm working with ECL's argument stack now and it is one of diagnostic messages done with "printf"
8:16:08
jackdaniel
if I had to guess I'd bet on one of make-instance methods from asdf's mixin-spaghetti
8:17:44
no-defun-allowed
oh well, nice to hear that there aren't any functions that require 184 arguments at least