Search
Thursday, 25th of May 2017, 21:16:46 UTC
22:26:23
JuanDaugherty
do I correctly understand planetlisp, cliki, and whatever ql gets to be the massed lisp greatness at this point, the equiv of hackage, et al.?
22:41:42
JuanDaugherty
do you know of other repos current?
22:41:53
whoman
just quicklisp and melpa/elpa
22:49:41
JuanDaugherty
it's more of a repo than emacs is a lisp
22:51:23
JuanDaugherty
ACTION (sorry didn't see had left)
22:57:51
pillton
phoe: Have you had a look at http://ipfs.io?
22:59:41
pillton
It would be cool to have support for that.
23:00:39
pillton
It would be even better if the environment of a lisp implementation was stored in ipfs.
1:19:59
akkad
ipfs is just a transport layer.
1:20:10
akkad
like downloading over http or sctp
2:57:59
loke`
What is the "best", or most portable, way to access the home directory?
2:59:31
pillton
What do you mean by access?
2:59:39
specbot
user-homedir-pathname: http://www.lispworks.com/reference/HyperSpec/Body/f_user_h.htm
2:59:41
loke`
I want to create a file in the home directory
3:00:41
loke`
(merge-pathnames (user-homedir-pathname) #p"somename.txt")
3:01:27
|3b|
ACTION would probably put those in opposite order, not sure it matters for that though
3:02:07
loke`
is the following portable though?
3:02:22
loke`
(make-pathname :directory '(:absolute :home) :name "somename" :type "txt")
3:02:39
|3b|
yeah, that sounds reasonable too
3:03:00
loke`
I'm not sure if :HOME is portable though
3:03:28
loke`
The glossary entry doesn't say much on the subject: http://www.lispworks.com/documentation/HyperSpec/Body/26_glo_v.htm#valid_pathname_directory
3:04:37
|3b|
ACTION doesn't see it anywhere either
3:05:03
loke`
Hmm, where is the documentation on :ABSOLUTE and :RELATIVE?
3:05:07
loke`
I can't even find that.
3:05:23
specbot
http://www.lispworks.com/reference/HyperSpec/Body/f_mk_pn.htm
3:05:55
loke`
Interesting... so :UP isn't standardised either?
3:06:05
loke`
I've used that on multiple occasions.
3:06:45
|3b|
not that it actually /defines/ it, just says something else is same
3:08:47
loke`
Pathnames are the worst part of CL. It's both grossly overengineered and at the same time woefully incomplete.
3:10:08
specbot
Restrictions on Examining a Pathname Directory Component: http://www.lispworks.com/reference/HyperSpec/Body/19_bbdc.htm
3:10:53
|3b|
so :up is portable but might error on some hypothetical filesystems
3:11:55
|3b|
(well, probably real filesystems too, if you happen to be using old/odd platforms)
3:12:24
loke`
Yes. I use ITS you insensitive clod!
3:12:47
loke`
Actually, I do have an ITS system. Let's try it there
3:16:38
loke`
MacLisp doesn't even have MAKE-PATHNAME
3:16:46
loke`
I should have guessed
3:17:07
loke`
Let's see if TOPS-20 lisp has it.
3:33:00
loke`
Hmm, the living computer museum's tops-20 doesn't have a working Lisp
3:39:49
ryanbw
loke`: perhaps try http://www.twenex.org
3:40:08
ryanbw
see also: http://wiki.twenex.org/tutorials:lisp
3:40:48
loke`
I know how it works. It's jsut that the LCM installation is missingt he runtime library, so I don't even have the + function :-)
3:43:56
ryanbw
loke`: I was suggesting that twenex.org may have a working set of lisps installed.
4:17:03
loke`
ryanbw: Yeah, I tried to add an account there, but their registration doesn't work.
4:43:23
beach
Good morning everyone!
4:53:38
atgreen
I get a "Heap exhausted during garbage collection" when I try to (ql:quickload 'lisp-binary). sbcl 1.3.5 on Fedora x86_64. Any ideas? How can help debug?
4:58:22
atgreen
hmmm... I updated quicklisp and lisp-binary was removed. maybe there;s a reason!
6:16:49
phoe
atgreen: exactly because of that.
6:17:04
phoe
lisp-binary is affected by a sbcl compilation bug.
7:06:01
akkad
e.g. `sbcl --dynamic-space-size 20480`
7:11:06
phoe
akkad: no, it does not work.
7:11:18
phoe
it's a bug in SBCL, present on their bug tracker
7:12:10
|3b|
ACTION tried the bug on tracker and didn't see the problem
7:12:14
phoe
https://github.com/heegaiximephoomeeghahyaiseekh/lisp-binary/issues/3
7:12:30
phoe
https://bugs.launchpad.net/sbcl/+bug/1557590
7:15:09
phoe
Xach: ^ about lisp-binary being removed from quicklisp
7:15:52
|3b|
ACTION only tried the sbcl bug test case, not lisp-binary though
7:16:27
phoe
I think you can import lisp-binary into QL local-projects and try to load it
7:16:45
|3b|
ACTION doesn't have ql set up on the system where i was trying it
7:17:20
|3b|
ah, maybe i was running it wrong though
7:31:44
|3b|
yeah, breaks if i run it properly
8:46:52
beach
Every time I have to restart my computer after having installed software updates, I am furious that we are still using operating-system technology from half a century ago.
8:49:50
splittist
beach: as opposed to better operating-system technology from half a century ago?
8:50:38
beach
That would be better, yes, but the best thing would be to move closer to the 21st century.
8:52:56
beach
splittist: A while ago, you hinted that you might do something about my suggested documentation system, or did I misunderstand?
8:55:42
splittist
beach: yes, I'm slowly thinking about it. Or thinking about writing up some outline documentation to start with.
Friday, 26th of May 2017, 9:16:46 UTC