| |
| |
| |
|
Page: 1 2 3 4 5 6 7 8
Comments:
<0> vali: either you have a broken sbcl installation or you're using a custom core and haven't set SBCL_HOME
<1> http://parenscript.org/
<2> kpreid: hmm, i got sbcl 1.0.1 from freebsd ports
<3> of lispworks, clisp, cmucl, sbcl, and allegro cl, only allegro cl complains about destructuring &whole
<3> defcon8: funny that parenscript.org has nothing to do, as far as i can tell, with the authors of parenscript
<3> it is quite pretty though
<1> oh, didn't know that
<3> Does anyone mind if I do a bit of ban clearing? Depending on your client, it may clutter your view with some mode messages.
<0> vali: then the 'port' is broken
<3> iirc there's some test failure that the port works around
<2> sigh, so sbcl doesn't work on either of the two operating systems i use
<4> vali: Which two?
<3> vali: it works pretty well on freebsd.
<3> vali: if the port is broken, don't use the port.
<3> don't be a port prisoner!
<2> i get a whole bunch of "pkg_delete: file '/usr/local/lib/sbcl/sb-bsd-sockets/<blahblah>.fasl' doesn't exist" when i delete it as well
<2> obviously something with the bsd-socket stuff
<3> the port is broken.
<2> all right, i'll get source
<3> well, if you alert the port maintainer (with a fix, even), you might help future freebsd users. but that is a bit of a meta-problem.
<2> Xach: i will alert him with the error messages, but i'm not sure how to fix it
<3> That's it! We're ban-free at the moment.
<4> Wow. A clear banlist?
<5> http://www-jcsu.jesus.cam.ac.uk/~csr21/lilyish.png
<3> I don't expect it to last more than 24 hours.
<6> btw, am I not supposed to get something newer than 0.8.16 after apt-get upgrade sbcl ?
<7> frodef: unstable?
<2> Xach: it worked when i got it directly from sbcl.org ... i installed the binary
<4> 0.8.16? That's an amusing version to upgrade to these days.
<3> vali: excellent
<6> pkhuong: I don't actually know..
<6> /etc/apt/sources.list seems to hold entries both stable and unstable.
<5> http://www-jcsu.jesus.cam.ac.uk/~csr21/lilyish1.png
<5> this application is 86 lines of code so far
<5> admittedly, typing anything but #\Space or #\Newline gives an error
<8> frodef: i thought that was the philosophy of debian. Only well fermented software.
<9> Krystof: intriguing
<7> frodef: unstable tracks the monthly releases pretty closely.
<10> frodef: You can check with "apt-cache policy sbcl".
<6> thanks.. I'm moving to unstable now.
<5> I WIN
<5> http://www-jcsu.jesus.cam.ac.uk/~csr21/lilyish2.png
<5> 91 lines of code
<4> Krystof: Heh. And here I thought for a second that you'd beat nethack again.
<11> Krystof pasted "toy embedded drei/gsharp" at http://paste.lisp.org/display/35661
<5> nyef: heh
<5> pls fix all bgs n tht cod k thx
<2> does sbcl come with asdf?
<12> vali: yes
<2> excellent
<13> foom, Xach: there's already one patch for adding a --script argument floating around. by kpreid, I think
<0> yes
<3> cool
<3> i would like to read it and subscribe to your newsletter
<0> it's not "floating around"
<0> it was sent to sbcl-devel and utterly ignored.
<0> </grump>
<13> it's something that I personally find utterly useless
<13> so I'm not horribly interested in looking at it
<0> here it is: http://article.gmane.org/gmane.lisp.steel-bank.devel/5837/match=kpreid+unix+script
<13> (besides filing it away under "maybe look at this sometime")
<13> now, there are obviously people who actually care about this. and since they care about it, they probably also have some opinion on how it should work
<13> so it'd be rather useful if those people tested / looked at the patch
<13> and reported their results on the list
<13> </grump>
<14> well, one thing pops out at me: it seems to me like it should disable-debugger and (quit) at the end.
<0> I don't recall my thoughts at the time, but it seems to me now that quitting, at least, is an application thing
<0> maybe my 'script' starts a tcp server and drops to the repl for maintenance, for example
<14> my thoughts are that it should work as close as possible to how it works in other unix scripting langauges
<0> the patch only addresses getting started, and your code can do whatever it likes
<0> my thoughts are that it should be minimally unusual
<0> ...for sbcl
<0> but, well, there's no serious downside to your suggestions
<0> so, to go with what jsnell said, try it!
<14> the other thing I think is that sbcl should probably just ignore the #! line on load/compile always.
<14> despite that not being specified in clhs.
<14> but that's an unpopular opinion. :)
<9> where, when and how one stops does seem like an application thing (what if you want to return some sort of exit code?); and I can see the point of a "Do you want to debug ?" restart...
<14> splittist: exiting with a default value of 0 doesn't prevent you from exiting otherwise
<0> foom: I'd be annoyed if LOADing a file consisting of "#!foo" didn't invoke my custom reader macro...
<14> kpreid: yeah well, like I said, an unpopular opinion. :)
<0> but if it were the *initial* #! reader macro, and only took effect at the top of the file, it'd be pretty safe
<14> kpreid: that's what I meant: only if it's the first 2 characters of the line.
<14> er, file
<3> right. and unloaded itself.
<4> At that point, why use a reader macro instead of just hacking LOAD to check for a file starting with #!?
<0> nyef: so that you can redefine #!
<4> (Well, other than a script, but you know what I mean.)
<3> i can imagine it.
<0> gratuitously excluding particular cases from correct operation bothers me
<4> Fine, how about an extra keyword argument to LOAD for "ignore shebang header" ?
<3> i have some thoughts on it, kpreid. i'll try to put them together and talk about it.
<0> ah, I know: LOAD establishes a handler which catches the undefined reader macro error for the first line. :)
<0> nyef: that's what my patch does, iirc
<4> ... Okay.
<13> yours does the opposite, no?
<13> err... well, depends on what "ignore" means :-)
<0> oh, I see
<0> wonder what I was thinking when I did that
<15> rydis: usocket catches hostname resolution errors on cmucl now (throwing errors when there are)
<16> indeed.
<16> minion: tell fowlduck about pcl-book
<17> fowlduck: have a look at pcl-book: "Practical Common Lisp", an introduction to Common Lisp by Peter Seibel, available at http://www.gigamonkeys.com/book/ and in dead-tree form from Apress (as of 11 April 2005).
<16> don't
<18> don't, not with common lisp
<16> recursion is just a tool, and common lisp has many tools
<19> expect them pathnames
<20> oh, cool
<19> from
<16> JKnecht: ((-:
<18> I wonder if Siebel is super rich because of his book
<20> when I have some time I will, I really need to have a heavy motivating factor, one heavier than self-enrichment
<16> if he's super rich, then it surely is from contracts gained through the book's popularity (:
<4> Cowmoo: That's not likely to be why Siebel is super rich (***uming that he's super rich).
<16> fowlduck: when you come up with an idea for a project, consider writing it in common lisp. (:
<18> well it's the canonical intro to lisp book now
<4> fowlduck: I find curiousity a good motivating device.
<18> but yes, good points heh
<20> giving the book away the book for free isn't a very effective way to make money directly from the sale of the book.
<18> does anybody really sit there reading a whole book on a screen?
<20> nyef: I do to a degree, but I'm just not curious enough. I'm a python junkie :)
<16> show of hands: who has the deadtree book on the shelf?
<16> \o/
<19> not me
<18> I bought deadtree
<20> not me
<18> er, do I lose?
<20> i didn't know it was even sold at all
<16> 2:3 isn't such a bad ratio (:
<20> i thought it was just given away, no dead-tree
<16> fowlduck: see the second part of the description, then (:
<18> mind, I bought it to learn...if I already new CL then I'd just glance at the free version
<20> ahh, i see
<20> what does it mean to be a functional language?
<19> differnt things to different people
<16> nyef: haha
<4> (Yes, this means that 6502 machine code is functional to me. My ability to sight-read much of it from a hexdump notwithstanding.)
<16> nyef: that's a great definition. it puts php firmly in the "dysfunctional" camp, too (:
<19> http://meansofproduction.biz/everything/index.pl?node=fp
<4> antifuchs: Oh, no. Just most PHP software.
Return to
#lisp
or
Go to some related
logs:
perl +submit +blinken
fglrx shmget(lowmem) Function not implemented
debian pentium-m libc6-686
ubuntu is not available in any software channel
+GLIBC_PRIVATE +kernel +2.6.15
#math
numpy and hotshot
#css
dovecot-0.99.11-2.EL4.1 Red Hat
#ai
|
|
