[Top][All Lists]

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

Re: [Bug-gnulib] Re: CVS Bug? or User error?

From: Alexandre Duret-Lutz
Subject: Re: [Bug-gnulib] Re: CVS Bug? or User error?
Date: Fri, 04 Jun 2004 09:13:37 +0200
User-agent: Gnus/5.1003 (Gnus v5.10.3) Emacs/21.3.50 (gnu/linux)

>>> "Paul" == Paul Eggert <address@hidden> writes:

 Paul> Derek Robert Price <address@hidden> writes:
 >> Paul, are you willing to concede this issue?

 Paul> Sure.  However, I just thought of one more problem that you should
 Paul> fix.  Suppose we have a software distribution that ships foo.y but not
 Paul> foo.c, or (worse) builds foo.y on the fly and then invokes yacc.  You
 Paul> need to detect this situation and have Automake report an error, since
 Paul> such software distributions are no longer supported by Automake.  (Or
 Paul> at the very least, you need to document this problem.)

I think the only way it can happen is if the parser.y is listed
in a nodist_prog_SOURCES variable.  Then neither parse.c nor
parse.y will be distributed and we should arrange that the rule
is always defined unconditionally in this case.  

(Unfortunately any associated parse.h would still be
distributed, causing distcheck failures.  I'll try to look at
this separate issue this weekend or next week.)
Alexandre Duret-Lutz

reply via email to

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