Path: csiph.com!news.swapon.de!eternal-september.org!feeder.eternal-september.org!mx02.eternal-september.org!.POSTED!not-for-mail
From: Keith Thompson <kst-u@mib.org>
Newsgroups: comp.lang.c
Subject: Re: snprintf alternatives in iso9899
Date: Tue, 08 Dec 2015 09:18:52 -0800
Organization: None to speak of
Lines: 45
Message-ID: <lnegewoo3n.fsf@kst-u.example.com>
References: <n45p5d$hmp$1@speranza.aioe.org> <87lh956li9.fsf@bsb.me.uk> <allocating-printf-20151208160736@ram.dialup.fu-berlin.de> <87fuzc7uq3.fsf@bsb.me.uk> <salfmt-20151208175950@ram.dialup.fu-berlin.de>
Mime-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: 8bit
Injection-Info: mx02.eternal-september.org; posting-host="945944de09706c9b4e29b53c9d2efdc2"; logging-data="15435"; mail-complaints-to="abuse@eternal-september.org";	posting-account="U2FsdGVkX1/yrZd3roy/r/6to/QwDUqi"
User-Agent: Gnus/5.13 (Gnus v5.13) Emacs/23.1 (gnu/linux)
Cancel-Lock: sha1:OYMHffmsCGHGv4COymFfieXfaWw= sha1:R0kgPCfkl6um/6NZZ5V6o+ApxJY=
Xref: csiph.com comp.lang.c:78188

ram@zedat.fu-berlin.de (Stefan Ram) writes:
> Ben Bacarisse <ben.usenet@bsb.me.uk> writes:
>>Did you ever show it to a compiler?  There's no way that code got
>
>   The last version of salfmt was not tested, indeed.
>   (I think, after I inserted »va_copy«, I did not 
>   test again.)
>
>   But previous versions were tested, and I hope to
>   find the time to one day maintain that page again. 
>
>   If you refer to a specific location in the code,
>   feel free to point out where you see the problem.

www.purl.org/stefan_ram/pub/c_faq_de

I haven't tried compiling it myself, but a couple of things jump out.

    s_type salfmt( s_type const f, ... )
    { va_list a; 
      char * b = 0;
      va_copy(a,f); { int const s = vsnprintf( 0, 0, f, a ); va_end(a); }
      if( s >= 0 )
      { size_t const k = 1 + s; if( b = malloc( k ))
        { va_start(a,f); { vsprintf( b, f, a ); va_end(a); }}} 
      return b; }

The code layout makes it difficult to read.

s_type is a typedef for char*.  (IMHO the code would be much easier to
read if it used char* directly.)

You call va_copy with an argument of type char*.  That might happen to
compile depending on how va_list is defined, but it's clearly wrong.

You define `s` inside a block and then refer to it outside the block.

(Personally I wouldn't publish code on a web page without at least
compiling it first.)

-- 
Keith Thompson (The_Other_Keith) kst-u@mib.org  <http://www.ghoti.net/~kst>
Working, but not speaking, for JetHead Development, Inc.
"We must do something.  This is something.  Therefore, we must do this."
    -- Antony Jay and Jonathan Lynn, "Yes Minister"