freenode/lisp - IRC Chatlog
Search
16:41:28
beach
Anyway, time for me to go cook dinner, so I won't do anything else until tomorrow morning (UTC+2).
17:48:22
jackdaniel
how to install custom lisp build (for instance sbcl compiled by me) in it from specified path under known name
18:42:34
TruePika
I can't figure out how to print some ANSI escapes without advancing the column (for pprint etc.)
18:46:04
TruePika
shrdlu68: I don't know if screen does (I'll check in a bit), but PuTTY does for sure
18:47:48
shrdlu68
strange. So when you ouput the code for color yellow, for example, the pointer position advances?
18:49:02
TruePika
(progn (pprint #<something with escapes>) (pprint-tabular ...)) doesn't have the first column align as expected
18:52:37
TruePika
I output 16 characters which don't advance the TTY cursor (including #\Esc), with the tabbing interval at four, it looks like there are 20 extra spaces of indent
18:54:31
TruePika
I'm now also curious about how double-width characters are treated, both by screen and Lisp (I know PuTTY and IIRC Vim work with them correctly)
18:55:16
shrdlu68
TruePika: Would it make a difference if you used format or some other means to print the codes?
18:56:05
TruePika
I'm using format, actually (so I can include the ^[ without literally typing it and potentially confusing the TTY)
18:58:42
TruePika
Ideally, I should be able to say "this is zero-width, don't pprint newline or touch the column position"
18:59:49
TruePika
aside from trying some sort of printf() call which might break with some stream types
19:34:24
joe9
any recommendations for an easy-to-understand pure lisp variant, please? I found LispKit Lisp https://en.wikipedia.org/wiki/Lispkit_Lisp , but, would prefer strict evaluation.
19:35:02
Xach
This channel is for Common Lisp. I don't really know where you might ask your question, sorry.
19:36:25
joe9
Xach: oh, sorry for the bother. I will check around for any other generic lisp channel.
19:56:15
shrdlu68
Adding ocsp support to cl-tls, and I'm seeing rather strange behavior with LetsEncrypt ocsp responders.
20:22:22
mazoe_
Hmm. I’m getting lost in the complexity that is CL paths. How does one get/print the string representation of a pathname?
20:24:07
mazoe_
Xach: I’d like to pass it to a unix utility. So I need an absolute or relative unix pathname
20:24:22
mazoe_
so, e.g., from (make-pathname :directory `(:relative “dir1”) :name "file.txt”), I would get “./dir1/file.txt”
20:24:45
Xach
mazoe_: because CL and the underlying OS might disagree about the details of the name.
20:25:32
Xach
namestring is the standard function, but it is for interoperability with the lisp currently running, not for external OS purposes.
20:25:52
TruePika
"However, since the effect of the pretty printer can be customized by conforming programs, the necessary flexibility is provided for individual programs to achieve an arbitrary degree of aesthetic control."
20:35:29
shrdlu68
TruePika: Highly unlikely, but I was thinking perhaps creating a new stream using something like #'make-broadcast-stream and writing the ANSI codes to that might help.
20:55:03
TruePika
I have a couple comments on the gist; one showing SBCL outputting ultimately to PuTTY, and one showing CCL on my Windows host
21:33:44
shrdlu68
TruePika: have you tried printing a newline before calling pprint-tabular? When I do, I don't get the spaces.
22:06:41
dbjergaard
why is this construction giving me an error: (defun foobar (foo) (format t "~a" foo))
22:07:59
dbjergaard
i am a dumb, its because I'm using evil and its not putting the cursor at the end of the s-expression
22:16:35
dbjergaard
does anyone know how to extend an existing package in sbcl? I'm writing functions that should live in sb-posix
22:19:04
White_Flame
dbjergaard: (in-package #:sb-posix) (defun ...), or (defun sb-posix::myfunc ...)
22:19:55
nyef
dbjergaard: Unless you're writing stuff to be included in SBCL as part of SB-POSIX, hacking up the SB-POSIX package is not recommended.
3:58:48
beach
In case anyone missed it, I posted several very small tasks on the Lisp Guild: https://github.com/Lisp-Guild/lisp-todo/projects/1
4:06:11
beach
Several of these tasks took about as long to for me to post as they would have taken for me to just go ahead and do. But the point here is for people who might be interested in learning about some of this software to start getting involved through some simple improvements.
4:08:03
beach
Take the assembler (Cluster) for instance. It might take a few hours for someone interested to accomplish the tasks of adding the shift instructions to the instruction description. Once that is done, that someone also knows something about how the assembler is designed, so could very well get involved with more instruction definitions and more features.
4:19:02
beach
Just in case someone thinks something like "Groan! An assembler. How boring! We already have a bunch of those.", Cluster is the assembler with a difference. Rather than feeding it input in the form of text or nested lists, the input is a sequence of standard instances. This way, we avoid the entire issue of syntax, especially for complicated address modes of the x86.
4:22:54
beach
Cluster is one more software component where I want to get away from the Unix way of imposing serialization and de-serialization. Furthermore, surface syntax seems to be the first point of disagreement in many situations, so I just avoid the issue. I mean, take documentation. Whatever choice is made (Texinfo, LaTeX, Docbook, Markdown, ...) it seems 90% of Lispers disagree; probably because there are 10 or more possible choices that