-----BEGIN PGP SIGNED MESSAGE-----
Moin,
On Sunday 15 August 2004 15:04, muppet wrote:
> On Aug 15, 2004, at 1:08 AM, Tassilo von Parseval wrote:
> > On Sat, Aug 14, 2004 at 05:50:20PM -0400 muppet wrote:
> >> On Aug 14, 2004, at 3:28 AM, Tels wrote:
> >>> FastCalc.c: In function `XS_Math__BigInt__FastCalc__is_ten':
> >>> FastCalc.c:480: warning: unused variable `class'
> >>> FastCalc.c:482: warning: unused variable `RETVAL'
> >>
> >> you've used CODE without setting RETVAL. if you're not going to use
> >> RETVAL, don't use CODE. see below.
> >
> > I don't see why that should be so. The main difference between CODE
> > and PPCODE is that the latter will adjust SP for you, whereas CODE
> > leaves it untouched. Furthermore, PPCODE can't be used with an OUTPUT
> > section. However, it's totally ok to have CODE without OUTPUT/RETVAL.
>
> that's true. i misspoke earlier, as void return without OUTPUT is
> quite okay with CODE (remembered that within seconds of hitting send
> ;-). but the idiom of XS is to use PPCODE when you're returning more
> than one value. when people read your xsubs and see CODE they're going
> to be surprised when they find multiple values returned.
>
> > Most of the time, my XSUBs follow this scheme:
> >
> > void
> > function (arg1, arg2)
> > type1 arg1;
> > type2 arg2;
> > CODE:
> > {
> > ...
> > EXTEND(SP, 3);
> > ST(0) = ...
> > ST(1) = ...
> > ST(2) = ...
> > XSRETURN(3);
> > }
>
> if you're going to write it that way, why use XS at all? just write
> the perlapi C code directly.
>
> you could just as easily have written
>
> void function (type1 arg1, type2 arg2)
> PPCODE:
> ...
> EXTEND (SP, 3);
> PUSHs (...);
> PUSHs (...);
> PUSHs (...);
>
> >> you could much more easily use int for a bool than an SV.
> >
> > For returning one boolean value, I'd always use one of
> > XSRETURN_(YES|NO):
> >
> > if (SvIV(temp) == 10)
> > XSRETURN_YES;
> > XSRETURN_NO;
> >
> > I find this more convenient than assigning to ST(0) plus doing an
> > XSRETURN(1).
>
> forgive me, but i don't understand why that's preferable to
>
> int
> ....
> RETVAL = some boolean condition;
> OUTPUT:
> RETVAL
I know this :)
First, thank you all for your time and tricks and tipps. Not declaring class
gets rid of the warning, very cool. Also, I got segfaults because I had
return type SV* and RETVAL, but forgot the OUTPUT: statement. Ouch.
Back to the difference between void and SV*:
The last lines of my original routine, with a void return type:
temp = *av_fetch(a, 0, 0); /* fetch first element */
ST(0) = boolSV((SvIV(temp) == 0));
#line 554 "FastCalc.c"
}
XSRETURN(1);
}
The same with RETVAL and OUTPUT: section:
temp = *av_fetch(a, 0, 0); /* fetch first element */
RETVAL = boolSV((SvIV(temp) == 0));
#line 582 "FastCalc.c"
ST(0) = RETVAL;
sv_2mortal(ST(0));
}
XSRETURN(1);
}
See the sv2mortal(ST(0)); there? :) I think it doesn't hurt, but it
certainly is not neccessary, since everything works fine without it.
Best wishes,
Tels
- --
Signed on Sun Aug 15 22:11:32 2004 with key 0x93B84C15.
Visit my photo gallery at http://bloodgate.com/photos/
PGP key on http://bloodgate.com/tels.asc or per email.
"Retsina?" - "Ja, Papa?" - "Warp 3." - "Is gut, Papa."
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.4 (GNU/Linux)
Comment: When cryptography is outlawed, bayl bhgynjf jvyy unir cevinpl.
iQEVAwUBQR/EJncLPEOTuEwVAQEjjwf8CCBlnzo8PjFSckeLmaNZs5uc0OoBZwNs
cVkQTD25DEuPMAmtLqonsfFBdJCdvWripUsy2Ofb864EcdNeTKsRp8m408SXPi9/
cKI07iDFI0HuEsmbhgp5YORpBlXNZsjOOaZDWtyxbTHyoaUSmn43SBI20hQq9pDS
2a0uESsOHGFlediY8IzfhOUFmEkU624T4E2u+eLi2aJ1/voJwam93K1lGf224gxa
U2o/57cO/p24l0AZerj/UCrGK3+gwp/DvDWwgNcymkoKr5dtWwwLNKQ8uuc6X9+L
R1DsMvdFE3z3C0QqzGCq6NvgEcLkfq7YNdJHTx4pfRH+vCJI0AU30A==
=jAGv
-----END PGP SIGNATURE-----
