From: "Robert I. Eachus" <rieachus@attbi.com>
Subject: Re: PL/I outflanks Fortran
Date: Tue, 01 Apr 2003 04:11:23 GMT
Date: 2003-04-01T04:11:23+00:00 [thread overview]
Message-ID: <3E89119E.4040002@attbi.com> (raw)
In-Reply-To: m1r891q5bz.gnus@usa.net
> i'm not sure it matters whether i find it useful, after all you find only 4
> out of 15 to be useful. i.e., that there is little common use for e.g base
> 9 did not prevent the language designers from providing for it, so why not
> the entire well defined range?
There is a subtle place where it matters, Ada.Text_IO.Integer_IO.Get
reads a number (without a point) according to the syntax of an integer
literal. (Ada.Text_IO.Float_IO allows a point.) This means that when
reading a string like 23#ABCEDFGHI... the read will stop immediately
after the F. So as long as we are allowing A, B, C, D, E, and F for
hexadecimal constants, there is no reason to outlaw bases lower than 16,
but higher bases would have noticable effects even if a user never
expected to use based numbers.
prev parent reply other threads:[~2003-04-01 4:11 UTC|newest]
Thread overview: 4+ messages / expand[flat|nested] mbox.gz Atom feed top
[not found] <2Q_ca.1611$dE2.3935@newsfeeds.bigpond.com>
[not found] ` <64050551.0303180643.7901fb94@posting.google.com>
[not found] ` <3e77364f$0$24810$91cee783@newsreader02.highway.telekom.at>
[not found] ` <3E78B30E.7C37E27E@adaworks.com>
2003-03-19 21:24 ` PL/I outflanks Fortran those who know me have no need of my name
2003-03-20 14:34 ` Frank J. Lhota
2003-03-20 18:13 ` those who know me have no need of my name
2003-04-01 4:11 ` Robert I. Eachus [this message]
replies disabled
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox