On 02/14/2012 01:34 PM, Peter Rosin wrote: > Hi! > > Commit v1.11-685-g947ce0e of 07-02-2012 added code for am__make_dryrun > that triggers a false positive in makefile-deps.test. > Please see:
<http://lists.gnu.org/archive/html/automake-patches/2012-02/msg00109.html> and follow-up messages. BTW, this message of yours definitively proves that Jim was right, and that we need the KFAIL_TESTS workaround ASAP. > I see two alternative approaches, which should I commit? > > 1. Avoid the check in the test > > From: Peter Rosin <p...@lysator.liu.se> > Date: Tue, 14 Feb 2012 12:55:32 +0100 > Subject: [PATCH] fixup: do not tickle makefile-deps.test with the make_dryrun > code > > * lib/am/header-vars.am (am__make_dryrun): Avoid code matching > '^ *:' in the Makefile, as that tickles makefile-deps.test. > --- > lib/am/header-vars.am | 5 ++++- > 1 files changed, 4 insertions(+), 1 deletions(-) > > diff --git a/lib/am/header-vars.am b/lib/am/header-vars.am > index 1078a3d..97516ef 100644 > --- a/lib/am/header-vars.am > +++ b/lib/am/header-vars.am > @@ -32,12 +32,15 @@ VPATH = @srcdir@ > ## and are thus executed also with "make -n" -- either because they > ## are declared as dependencies to '.MAKE' (NetBSD make), or because > ## their recipes contain the "$(MAKE)" string (GNU and Solari make). > + > +## The case statement has [:] in order to not tickle makefile-deps.test > +## which greps for '^ *:'. > am__make_dryrun = \ > { \ > am__dry=no; \ > for am__flg in : $(MAKEFLAGS); do \ > case $$am__flg in \ > - :) ;; \ > + [:]) ;; \ > *=*|--*) ;; \ > *n*) am__dry=yes; break;; \ > esac; \ > I'd go with this FTM. Feel free to push it when you want. Thanks, and again, sorry for the confusion, Stefano
