Re: strtonum(3) from OpenBSD?

[Marc Balmer <marc%msys.ch@localhost>]

Am 06.08.2009 um 23:31 schrieb Matt Thomas:

> On Aug 6, 2009, at 7:49 AM, Thor Lancelot Simon wrote:
>
> > On Wed, Aug 05, 2009 at 10:48:39PM +0200, Marc Balmer wrote:
> > > almost everything that has been brought up against having strtonum  
> > > in
> > > libc is present in strsuftoll(x):  it returns long long, it uses a
> > > string for error reports, it does not support i18n etc....  So all  
> > > the
> > > arguments gainst strtonum are somewhat void.
> >
> > Um, no.  As I said earlier, NetBSD's libc does not need to be the  
> > _union_
> > of every mistake anyone has ever made in libc in any other Unix-like
> > operating sytstem.
> >
> > It is unquestionably the case that there is bad code in NetBSD.   
> > That does
> > not in any way go to justify adding more bad code to NetBSD.
> >
> > There has been nearly unanimous opposition from other developers to  
> > adding
> > this particular code to NetBSD's libc.  I suggest that unless core,  
> > who are
> > the relevant technical management, say it's okay to add, it would  
> > really be
> > about the time to drop the subject, for now.
>
> I concur whole-heartedly with everything Thor said.  We are supposed  
> to learn from our mistakes, not repeat them.  If this issues needs  
> to be decided by core, you can guess which way I'm deciding.
>
> If you feel strongly about getting it added, speaking as a member of  
> core, you will need to ask for and receive core's blessing.

you missed to read my answer to Thor's mail before mailing me this, I  
guess.  if strtonum is consider a repetition of the strsuftoll  
mistake, we could think of an API that resolves the mistake.