| |
| |
| |
|
Page: 1 2 3 4 5 6 7
Comments:
<0> nyef: hmm, what kind of non-lisp code and what for?
<1> Cin: yeah, just curious. Not being a Win32 hacker, there's not much I can do to help either.
<1> "Works for me" on XP MCE with SP2.
<2> luis: One scenario was something like a debug event monitor, since that basically has to have a separate thread, and we don't have threading in SBCL/Win32 yet.
<2> (To those who claim that you don't need a separate thread for running the debug API: You do if you want to do it right.)
<3> chandler, hehe
<2> luis: Another scenario is using SBCL as a host compilation environment for standalone Win32 executables using a subset lisp or non-lisp target environment.
<0> nyef: do you mean that compilation would make more efficient FFI calls or is this unrelated to FFI?
<2> luis: Umm... Neither?
<4> What do you call to an instance (an application, an use) of a data structure? :)
<2> luis: Basically, I already know how to find my way around sb-alien a bit, and have a good idea of how to do what I want with it.
<5> structured data
<4> Suppose I have a variable QUANTS that is a list. List is a data structure and QUANTS is a...
<6> variable
<2> masm: Variable?
<6> "quants is bound to a list"?
<5> in lisp, a variable is a variable.
<5> It may be bound to an instance of list.
<5> instances of type are values.
<4> ...
<2> masm: Getting the feeling that you're not asking your question in a way that the rest of us can understand?
<4> I guess...
<7> masm: this is not java
<1> "value"?
<7> its not an instance of a list, it is a list
<7> (if it is, a list)
<7> you even said it yourself..
<5> this is equivalent. when you say a value is a list, it's just shorthand for saying the value is an instance of list.
<3> faxD
<8> Anyone here familiar with ASDF? Trying to get UFFI acknowledged with (asdf:operate 'asdf:load-op :uffi)
<7> :p
<8> Central registry: (#P"F:/Prog/Lisp/Libraries/uffi-1.5.17"
<0> Modius: put a / at the end
<8> It says it can't find uffi, and I don't know anything about asdf
<1> what luis says :-)
<0> Modius: #P"F:/Prog/Lisp/Libraries/uffi-1.5.17/"
<8> Bingo - thanks!
<8> noob error. . . .
<4> pjb: You are right, but I would like to distinguish the use... I'm now using "data structure instance", but I don't like it.
<1> masm: "object which is TYPEP LIST"
<1> masm: otherwise, just "value"
<9> luis: hi!
<0> jamesjb: hello :)
<9> luis: i keep forgetting to reply to your mail, but i'm okay with your preferred syntax for options, keep it consistent
<9> i'm prepared to accept that my idea is too radical and progressive
<0> eheh
<1> masm: http://www.lispworks.com/documentation/HyperSpec/Body/26_glo_v.htm#value and http://www.lispworks.com/documentation/HyperSpec/Body/26_glo_o.htm#object
<4> chandler: Ignore use in programming languages. I'm thinking more in abstract terms.
<1> I'm not sure how to get more abstract than "value" or "object".
<2> chandler: "thingie" is pretty abstract. ^_-
<1> I think "whatchamacalit" is probably more abstract, since it implies that even the name of the thingie is unknown.
<3> chandler, or indeed, the er, er, um
<3> :D
<7> meta-whatchamacallit :D
<2> The whoosiewhatsit?
<7> hehe
<2> ("Whatchamacalit", to me, implies that something is of the candybar nature.)
<1> "Whatsit" has a whole different set of implications to me.
<4> I suppose that the question didn't make much sense. :)
<2> "Thingum" and "thingummie" are also options...
<8> CLSQL issue: (connect "Driver={SQL Server};Server=PSNetSQL1;Database=Consol_HamiltonSundstrand_copy_prod_Jn;Uid=ConsolSystemUser;Pwd=dotnet;" :database-type :mssqlserver)
<8> :mssqlserver not recognized?
<8> Thing is, *INITIALIZED-DATABASE-TYPES* is nil
<8> (initialize-database-type :database-type :mssqlserver) fails - where can I find a valid list?
<10> Modius: "The database type for the MySQL back-end is :mysql"
<8> Just occurred to me I need to use :odbc - have a problem with the first param now but I have to diagnose that for a while now
<10> anyway, CLSQL manual, appendix A. Database Back-ends is the source of that kind of information
<8> Thanks
<8> I need to read this a bit.
<10> There's a school of thought that hold that you should read the manual before using the sofware :)
<11> Weenies.
<10> hmmm... how about those who read the docs when they have problems?
<12> but, I thought asking on irc was the new reading the docs
<0> slyrus_: asking on web forums is the new asking on irc
<10> Answering an irc question by googing for 10 seconds make one feel a bit like failing a Turing test
<13> hi, this question is harder (I hope).
<13> is there a macro around defsystem that let you define systems simpler way, like (def-my-system "file1" "file2"....)
<13> ?
<10> puchacz: what's wrong with cargo-culting an existing simple system from somewhere else?
<10> no doc-reading envolved ;)
<13> Urfin: I don't like typing each file twice, once as :file and then :depends-on :)
<10> use serial t
<13> uhm?
<2> (asdf:defsystem :my-system :serial t :components ((:file "whatever") (:file "whatever-else")))
<13> possibly doc-reading will be involved then, thanks.
<13> nyef: thanks, I will cargo cult it.
<2> Enjoy.
<14> Is there a way to unbind a symbol macro? fmakunbound works for regular macros, but makunbound doesn't seem to do the trick....
<6> you might need to unintern the symbol completely
<15> or re-symbol-macro it to (symbol-value 'foo).
<2> It's things like that that make me wonder about the viability of CL for truly long-term maintainance on running systems...
<10> attila_lendvai: around?
<16> Urfin: yep
<10> I wanted to ask, is your repo GnU-Emacs friendly?
<10> GNU even
<16> Urfin: i usually test it on gnu emacs, but my primary emacs is xemacs
<10> ah ok, thanks
<16> Urfin: if you pulled it recently i've pushed some potentially fragile changes... just unpull/fix/report them
<16> i mean the bugs if any pops up
<2> Oh, look. PG added a new essay.
<10> ATM I'm unable to get it, it says "Copying patch ... of 2904" and at some random number around 500 I get "libcurl: HTTP error (404?)"
<0> nyef: nah, it's just marketing for that new book.
<16> Urfin: use darcs get --partial
<2> luis: True, but it beats "How Art Can Be Good".
<2> (A.K.A. "Why Selling Out Is What Makes Art Worth Doing.")
<0> heh
<14> Krystof & pkhuong: thanks
<10> attila_lendvai: thanks I'll give it a spin
<12> so, with mcclim, if I change the :text-style for my interactor pane, it works for things that I input, but the "Command:" bit is still in the default font. any ideas?
<17> what's the default test function for find?
<17> nevermind, eql.
<18> psi: aye. eql is the default test for everything.
<12> oh, I can set climi::+default-prompt-style+, but that seems a bogus way to do things.
<17> Xach: ok, thanks.
<19> slyrus_: also, Goatee will ignore the text style.
<12> nice :)
<20> slyrus: if that
<20> slyrus: if there's a default-prompt-style, one might imagine there's a non-default-prompt-style
<12> are you suggesting there _is_ or the _should be_ one?
<20> either!
<12> which one? :)
<20> i dunno, whichever fits reality
<12> ah, ok.
<12> I think it should pull the text-style from the interactor instead of using the default, but what do I know
<20> if I had that problem I'd look at the uses of +default-prompt-style+ to see if and when they use something other than the default
<20> if they do, then there's your hook; if not, there's your patch :)
<12> gee, thanks :)
<19> I think it's because we might want monospaced text for the input field, for easy editing, but serif (or whatever) fonts for the prompt for more pretty.
<12> alright, off to the store. more mcclim learning later.
<19> Of course, the ideal would be to define a load of generic functions to permit the programmer to customize all this, but alas, CLIM defines no such thing.
<19> I guess fonts weren't a big issue on the Lisp machine. :)
<21> IPmonger: Around?
<22> hey
<21> Hey! You rang?
<23> IPmonger pasted "strange CMUCL 19D problem" at http://paste.lisp.org/display/36686
<22> I've been doing some testing of various snapshots against ansi-tests
<22> the 19D snapshots have this wierd problem
<21> Only on ppc?
<24> Looks very similar to the error you get with time on non-64bit Solaris, which has to do with performance counters of some sort.
<22> rtoy_: i've only tried ppc
<25> what would cause this: http://rafb.net/p/rY95Lh35.html ?
<24> ("with TIME", I mean.)
<21> rydis: What error is that? You mean non-ultrasparc CPUs? Or an early Solaris that doesn't support 64-bit machines?
<24> rtoy: Solaris on U1:s run 32-bit by default, i
<24> ... since there is a hardware bug in US-I cpu:s.
<21> IPmonger: Ok. I "fixed' the FP traps on ppc so it was supposed to work. I guess I messed something up.
<21> rydis: Hmm. Ok. I don't care too much about those machines, since I don't have one. :-)
<21> IPmonger: The other divide-by-zero tests are ok? Only the divide by 0L0?
<24> rtoy: IIRC TIME ***umes it's running a 64-bit kernel, with some performance counters or something. It's been ages since I ran into it, and I know how to fix it, so it's not a big deal.
Return to
#lisp
or
Go to some related
logs:
Ubuntu Livedisk Root password
manman torrent
drjef
gentoo-livecd root-account
wifi-radar not using eth1
#gentoo
/usr/sbin/mkinitrd: Cannot determine root device
#python
#math
phprox
|
|