Re: [Nasm-users] New Re: problem with installation

[Frank K. <fbk...@ve...>]

Rod Pemberton wrote:
> Posted to GMANE's nntp version of  nas...@li... i.e.,
> gmane.comp.lang.nasm.general:
> 
> "Frank Kotler" <fbk...@ve...> wrote in message
> news:470...@ve......
> 
>>RG Nurrahmat wrote:
>>
>>>I am having problem with installing .bz2 file (NASM source) from
>>>ftp://ftp.kernel.org/pub/software/devel/nasm/source/nasm-0.98.39.tar.bz2
>>>, however, during installation with (MINGW32) MSys, I had this error
>>>configure: error: NASM requires ISO C99 (specifically, "snprintf")
>>>after I typed the following,
> 
> 
> Anyone know if there a way to disable the snprintf check during the NASM
> configure?  For example, DJGPP v2.03 uses GCC but doesn't use GLIBC.  v2.03
> uses a custom C library which doesn't have snprintf.  One could build
> w/DJGPP v2.04 which has snprintf(), but if one needed, wanted, or prefers
> v2.03, they'd implement their own snprintf() equivalent which won't pass the
> snprintf check, correct?
> 
> Personally, I'd prefer that NASM be ISO C90 or be able to fallback to ISO
> C90 if ISO C99 isn't available, for the reason you've mentioned:
> 
> 
>>>I was under the impression that "almost any" compiler would conform to
> 
> C99 by now, but it turns out "almost none" of 'em do.

It's a work in progress. HPA checked in some stuff overnight which I 
believe addresses this issue. The lastest 20071003 should have it.

ftp://ftp.zytor.com/pub/nasm/snapshots/

Folks who prefer to use obsolete tools - and I'm generally one of 'em - 
do so at their own risk. I'm sure Nasm isn't the only project that uses 
snprintf! I'm sorry so many people are having trouble with it since 
0.98.39, but we wanted to fix that buffer overflow, and snprintf seemed 
like the "modern" way to fix it. If available compilers don't have that 
capability, perhaps *that* should be considered a "problem".

Best,
Frank