Email List: Xaustin-group-lX
[All Lists]

Re: AI 2000-05-034: wording for strtod()

To: Ulrich Drepper <yyyyyyy@xxxxxxxxxx>
Subject: Re: AI 2000-05-034: wording for strtod()
From: "Clive D.W. Feather" <yyyyy@xxxxxxxxx>
Date: Tue, 15 Aug 2000 18:47:43 +0100
Cc: yyyyyyyyyyyy@xxxxxxxxxxxxx
References: <m3d7jbvcgr.fsf@otr.mynet.cygnus.com> <20000815111248.N42158@demon.net> <m3aeeeqzzg.fsf@otr.mynet.cygnus.com> <20000815171140.D42158@demon.net> <m3d7japkhr.fsf@otr.mynet.cygnus.com>
Ulrich Drepper said:
>> Thus proving once again that perl is broken. :-)
> It's not proving this.

Sheesh. It was a joke.

But, since you raise the topic, how *does* perl treat "1234" versus
"0x1234". Are you saying that the former is a float but the latter an
integer ? If so, I reckon it *is* broken.

> It's proving that the ISO C99 committee did a
> lousy job by introducing changes outside those the future directions
> described.  I didn't want to write this but you forced me.

Only if you can guarantee to be in the C locale. Outside that the
implementation is free to accept anything it likes with strtod.

It's true that this is a Quiet Change. That doesn't make it a "lousy job";
sometimes that is better than the alternatives.

-- 
Clive D.W. Feather  | Work:  <yyyyy@xxxxxxxxx>   | Tel:  +44 20 8371 1138
Internet Expert     | Home:  <yyyyy@xxxxxxxxxx>  | Fax:  +44 20 8371 1037
Demon Internet      | WWW: http://www.davros.org | DFax: +44 20 8371 4037
Thus plc            |                            | Mobile: +44 7973 377646 

<Prev in Thread] Current Thread [Next in Thread>