<two-face>
malc: has ocaml team already shown some interests for it?
<malc>
two-face: i dont know if have seen it, but there was pretty longish disucssion on the mailing list about this sort of stuff
<malc>
two-face: yep, Xavier made it quite clear, that its not wanted
<Dybbuk>
malc: Why is it not wanted?
<two-face>
malc: i guess he wants some runtime checks
<malc>
runtime checks are(i think it was general consensus on the before mentioned thread) impossible (or at the very least very hard)
<malc>
Dybbuk: because quote: 'dynamic linking is evil'
<Dybbuk>
Oh man.
<two-face>
what about loading bytecode within native code ? :)
<Dybbuk>
Dynamic linking isn't evil, it's the standard. :)
<Dybbuk>
Is it possible to create static libraries with OCaml?
<two-face>
cmxa are static libraries
<malc>
static libraries? ahem.. -a
<Dybbuk>
Yeah, .a files. :)
<Dybbuk>
To link against C or something.
<malc>
-a creates .cmxa and .a
<two-face>
but, in CDK, there is a DLOpen wrapper IIRC
<malc>
.cmxa is just a collection of .cmx'es. .a is, well... usual .a
<Dybbuk>
Ahh, neat.
<malc>
yes, to open C porduced DLLs and call exported functions
<Dybbuk>
Huh.
gl has quit [No route to host]
Segora has quit [Remote closed the connection]
Segora has joined #ocaml
<Segora>
re
<two-face>
re
* Segora
sighs.
malc has quit [Read error: 110 (Connection timed out)]
<Segora>
got a little further with the backup system prototype (in OCaml). Unfortunately, the final version is in C++.
<two-face>
what a shame
<Segora>
.. but type inference really helps to think about the design. ;)
<two-face>
how dare you reimplement ocaml progs in C++ !
<Segora>
yeah. wanted to hack it together at the weekend to show that it can be done in two days in a decent language instead of several months(!) for almost worthless c++ code from an external contractor. a pity that i got ill.
<two-face>
bwa ah ah
<Segora>
never saw that much confusion carved in code for a long time.
<two-face>
how dirty is your c++ code ? :)
<Segora>
that depends on whom you ask.
<two-face>
heh
<Segora>
I believe most people would say it's quite clean. Only one colleage still gives me good advice on how to improve it.
<two-face>
ok
<two-face>
got to go now
<two-face>
bye
<Segora>
bye
two-face has left #ocaml []
<Dybbuk>
Segora: I did something similar with Common Lisp (improving on a crappy Perl program), but my boss didn't seem to care.
<Segora>
Dybbuk: well, I want to reach a point where the decision is between using the usable ocaml version or putting more time in the not-yet-functional C++ version. ;)
karryall has quit [Read error: 104 (Connection reset by peer)]
<Segora>
after all, not all hope's lost over here. there's even software written in ocaml and erlang in production use.