[Top][All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: to conditionally test, or not to conditionally test?

From: Keith MARSHALL
Subject: Re: to conditionally test, or not to conditionally test?
Date: Fri, 27 May 2005 16:05:42 +0100

Stepan Kasal wrote, quoting me:
> On Fri, May 27, 2005 at 01:38:28PM +0100, Keith MARSHALL wrote:
>>>      #! /usr/bin/perl
>>>   If you omit the space before the path, then 4.2BSD based systems
>>>   (such as DYNIX) will ignore the line, because they interpret
>>>   `#! /' as a 4-byte magic number.
>> I recently queried this advice, in respect of its application in the
>> groff project, and was advised that it is nonsense.  I posted that
>> response here, but as yet have seen no comment.
> I saw your mail and I want to comment out this paragraph, if Paul
> agrees.
> But let's define what ``nonsense'' means:
> The advice perhaps used to be true, prehaps 4.2BSD and DYNIX really
> had this limitation, I don't know.  But the reason to delete the
> advice is that these systems are now so rare that Autoconf doesn't
> want to target them.

But, if that is the case, would it not be better to retain the comment,
qualifying it to note that such systems are so rare today, that perhaps
strict conformance isn't really necessary?

> Autoconf-2.x supports sustems with shells without functions.  That's
> why autoconf generated macros cannot use functions.
> If I understood correctly what Paul said some time ago, we'll probably
> drop this support in Autoconf-3.  That means that Autoconf-3 will be
> able to use functions internally.
> But you can use functions in your with Autoconf-2.x, if
> _you_ don't need support for platforms without shell functions.
> (Ralf has explained this much better.)
> Hope this explains it.

Yes, thank you.  And thanks also to Ralf, for his explanation.  Perhaps
again, some qualification of the statement in the documemntation would
be helpful.

Best regards,

reply via email to

[Prev in Thread] Current Thread [Next in Thread]