[Top][All Lists]

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

Re: document AS_BASENAME

From: Stepan Kasal
Subject: Re: document AS_BASENAME
Date: Thu, 30 Mar 2006 18:18:43 +0200
User-agent: Mutt/1.4.1i


On Wed, Mar 29, 2006 at 10:47:14PM -0800, Paul Eggert wrote:
> address@hidden (Eric Blake) writes:
> > But looking at the source to AS_BASENAME, even recent Solaris
> > mishandles "basename /" with an empty line instead of "/".

> [...] the practical importance of "basename /" is pretty small.

well, the importance of "basename /" is that it is the single test
which we use to decide whether the basename implementation is trustworthy
or not.

Are there buggy implementations of basename?  If yes, how can be they

Our current approach is that if it is able to handle even the tricky "/",
we can use it, otherwise we shall use expr/sed instead.
That has some point: "If it is not fully compliant, don't bother with it."

> > However, I just noticed that AS_BASENAME does not do suffix
> > stripping [...]

> [...] To remove a suffix one should use expr, not basename.

This trivial point should be perhaps added to the docs.

When I was at it, I noticed that the description of AS_BASENAME contains

        See {Limitations of Usual Tools}, for more details about what this
        returns and why it is more portable than the @command{basename}

but the target of this reference contains:
  - a one line example which should have been cited directly
  - no actual portability problems

And likewise for the description of AS_DIRNAME.

I tried to resolve all the points above, see the attached patch.

OK to commit?

Have a nice day,

Attachment: autoconf-20060330-basename-doc.patch
Description: Text document

reply via email to

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