#lisp - Thu 3 May 2007 between 00:01 and 04:23

NY Lost Funds



lemonodorthere used to be some at international-lisp-conference.org/speakers.html
robsynnottso where's the next one on?
Xachcambridge
tudor_i get this error when i try to make a connection using cl-xmpp: "Document not well-formed: Undeclared namespace prefix: stream"
robsynnottcambridge again?
slyrus_robsynnott: the other one
robsynnottah, I see
alastair37Is it normal for a CL program not to export symbols that are required to actually use said program? Because apparently NO ONE has EVER used cl-ode to do ANYTHING nontrivial, because otherwise they would have noticed that said symbols need to be exported.
(for instance, in ode, when making a contact joint, you need to make a struct called surface-parameters, and use a bunch of its slots; however, in cl-ode, neither the struct name nor the slot names are exported, so you have to modify cl-ode.lisp to export them.)
I just had to get that off my chest.
faxhm odd
do you think its possible the implementers wanted to put a higher level of abstraction over the function
so didnt expose it directly?
(but of course, didnt get round to doing it yet)
alastair37i can't find any evidence for that in the code. of course, if you were a real person, and you wanted to lispify things, you wouldn't expose exactly this interface, but cl-ode is not like that.
there's basically no code in cl-ode other than defcfuns and defctypes and so on.
it's basically as far from cl-opengl as you can get.
faxoh I see
yet cl-opengl is really nice
yes*
mm
alastair37at some point i am going to try to release my lispifications and modifications to cl-ode as a fork
rudican't you get upstream to merge your patches?
alastair37i already tried once with a much more trivial fix to get it to work with cffi-newtypes; the guy never responded.
faxalastair37: yeah please do I want to use cl-ode
alastair37basically i've made a wrapper around a BUNCH of the cl-ode functions that make them act like real lisp stuff
the only problem is that i named them badly
e.g. if there's a cl-ode function collide, i named mine my-collide
because collide was already taken
faxalastair37: please come to #lispcafe for a sec
alastair37will do
rudialastair37: personally, I'd try again and again, reworking the patch and making it as pretty as possible - IMO a fork needs a visible history of rejected / ignored patches and a friendly, concerned forker to succeed. (IOW, try to draw other developers with you through visible competence and friendliness :)
alastair37yeah, but that's a lot of work
also, cl-ode is really inactive
there's nothing on cl-ode-devel
jsnellin that case you might ask for commit access
alastair37and the last entry in cl-ode-announce was march 2006
_deepfirehmm, it'd be nice if macroexpanded versions containing GENSYMmed bindings could be evaluable
Zhivagowhy can't they be?
_deepfireer, macroexpanded versions, as in code transpermed by slime-macroexpand-all-inplace

Page: 2 9 16 23 30 

IrcArchive

NY Lost Funds