autoconf
[Top][All Lists]
Advanced

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

[Jason Dorje Short <address@hidden>] Bug#179086: autoconf: AC_DEFINE_UNQ


From: Ben Pfaff
Subject: [Jason Dorje Short <address@hidden>] Bug#179086: autoconf: AC_DEFINE_UNQUOTED fails when used on variables
Date: 05 Feb 2003 16:43:23 -0800
User-agent: Gnus/5.09 (Gnus v5.9.0) Emacs/21.2

The following bug was reported against the Debian GNU/Linux
packaging of Autoconf 2.57.  I am passing it along because I
suspect it is an upstream issue.

-------------------- Start of forwarded message --------------------
Subject: Bug#179086: autoconf: AC_DEFINE_UNQUOTED fails when used on variables
From: Jason Dorje Short <address@hidden>
To: Debian Bug Tracking System <address@hidden>
Date: Thu, 30 Jan 2003 15:28:47 -0500

I have code like

    ac_tr_lib=HAVE_LIB`echo $1 | sed -e 's/[^a-zA-Z0-9_]/_/g' \
    -e 'y/abcdefghijklmnopqrstuvwxyz/ABCDEFGHIJKLMNOPQRSTUVWXYZ/'`
    AC_DEFINE_UNQUOTED($ac_tr_lib)

which, is intended to define certain preprocessor variables depending on the
value it is given.  Under recent versions of autoconf, there is no acconfig.h
so the information it would have provided is now provided by autoheader.

According to the autoconf manual, AC_DEFINE_UNQUOTED is able to handle variables
as its first argument.  Yet it appears it cannot.  Perhaps autoheader must
extrapolate what possible values will be defined, and it just fails to do that
here.  Regardless, the result is that the value does not get defined, causing
(in this case) an obscure bug.

I can believe that autoheader is not capable of handling code like this.  But
what is bad is that the autoconf manual indicates this code will work -
and autoheader gives no warning that it is confused.  So what would be an
easy fix becomes a well-hidden bug.
-------------------- End of forwarded message --------------------

-- 
"The only problem with Linux for Dummies is that the advice it
 contains will result only in embarrassment and inconvenience if
 followed, not actual death."
--henke at insync dot net




reply via email to

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