bug#9306: Java build errors due to trailing colon in CLASSPATH

2011-08-16 Thread Daniel Richard G.
I am using Automake 1.11.1, and have recently encountered a subtle issue that causes compilation of some Java programs to fail. Here is a typical build failure, produced from a minimal test case I've assembled, using OpenJDK on Ubuntu's Natty release: make[3]: Entering directory `/tmp/automake-ja

bug#9306: Java build errors due to trailing colon in CLASSPATH

2011-08-16 Thread Stefano Lattarini
Hi Daniel, thanks for the report. On Monday 15 August 2011, Daniel Richard wrote: > I am using Automake 1.11.1, and have recently encountered a subtle issue > that causes compilation of some Java programs to fail. > > Here is a typical build failure, produced from a minimal test case I've > assem

bug#9213: aclocal7.test of automake 1.8.3 fails on s390/s390x

2011-08-16 Thread Stefano Lattarini
On Monday 15 August 2011, Philipp Thomas wrote: > * Stefano Lattarini (stefano.lattar...@gmail.com) [20110808 23:53]: > > > May I ask you whether you've managed to solve your problem? If yes, I > > could close this bug report then. > > Sorry for answering so late. It disappeared without me doing

bug#9213: aclocal7.test of automake 1.8.3 fails on s390/s390x

2011-08-16 Thread Stefano Lattarini
On Tuesday 16 August 2011, Stefano Lattarini wrote: > On Monday 15 August 2011, Philipp Thomas wrote: > > * Stefano Lattarini (stefano.lattar...@gmail.com) [20110808 23:53]: > > > > > May I ask you whether you've managed to solve your problem? If yes, I > > > could close this bug report then. > >

bug#9245: FreeBSD make in concurrent mode report spurious success in automake-generated tests harness

2011-08-16 Thread Stefano Lattarini
On Friday 12 August 2011, Stefano Lattarini wrote: > OK, the attached hacky patch seems to fix the bug. > LOL no, there was an embarassingly stupid error in my patch: diff --git a/lib/am/check.am b/lib/am/check.am index 0b54312..75f8bba 100644 --- a/lib/am/check.am +++ b/lib/am/check.am @@ -2

bug#9245: FreeBSD make in concurrent mode report spurious success in automake-generated tests harness

2011-08-16 Thread Eric Blake
On 08/16/2011 10:04 AM, Stefano Lattarini wrote: The "proper fix" is even easier BTW: just use "$$exit || exit 1" instead. The updated patch should now work (and I've tested it properly this time). This new patch hasn't been derived by "cargo-cult guessing", but from the explanation of the relev

bug#9245: FreeBSD make in concurrent mode report spurious success in automake-generated tests harness

2011-08-16 Thread Stefano Lattarini
Hi Eric. On Tuesday 16 August 2011, Eric Blake wrote: > On 08/16/2011 10:04 AM, Stefano Lattarini wrote: > > The "proper fix" is even easier BTW: just use "$$exit || exit 1" instead. > > > > The updated patch should now work (and I've tested it properly this time). > > This new patch hasn't been d

bug#9242: distcheck fails when having TEST in sub-directory

2011-08-16 Thread Stefano Lattarini
On Friday 05 August 2011, Stefano Lattarini wrote: > On Thursday 04 August 2011, Peter Johansson wrote: > > > cat > Makefile.am < > TEST_EXTENSIONS = .test > > TESTS = test/foo.test > > EOF > > > You've forgotten to add $(TESTS) to $(EXTRA_DIST) here: > > EXTRA_DIST = $(TESTS) > > If you add t

bug#9306: Java build errors due to trailing colon in CLASSPATH

2011-08-16 Thread Daniel Richard G.
On Tue, 2011 Aug 16 15:30+0200, Stefano Lattarini wrote: > > Luckily, as far as I know, the above construct should be portable to > all non-museum shells. So I've used it with only a minor modification > (with the only purpose of making the contruct slighlty clearer): > > CLASSPATH=mumblemumble$