Re: Testing automake on Cygwin 1.7

[Stefano Lattarini]

On Saturday 26 November 2011, Stefano Lattarini wrote:
>  
> Attached are the two patches I had promised.  I'll wait 72 hours or your
> ACK (whichever comes first) before pushing.
> 
> Regards,
>   Stefano
> 
Hmm... on a second thought, the check in txinfo21.test that sets
`$have_correct_dir_timestamp' makes little sense -- it fails also
on Linux, and *that's correct* in fact, because it's normal that
a directory timestamp is not updated when a pre-existing file in
it is touched!  The timestamp should be updated if a new file is
created in the directory, or if the directory is removed and
re-created.

I propose to substitute the first patch with the attached one
instead.  Peter, if you have time, could you test if it works
correctly on Cygwin?

Thanks, and sorry for the noise,
  Stefano

From 91ab9c61fa4fe7451341f1434245a5cd92936162 Mon Sep 17 00:00:00 2001
Message-Id: <address@hidden>
From: Stefano Lattarini <address@hidden>
Date: Sat, 26 Nov 2011 14:35:58 +0100
Subject: [PATCH] tests: work around a Cygwin limitation in 'txinfo21.test'

* tests/txinfo21.test: At least on Cygwin, the timestamp of a
directory might not be properly updated when its contents are
modified; we need to account for this behaviour.  Also, since
we are at it, remove a redundant `$sleep'.
---
 ChangeLog           |    9 +++++++++
 tests/txinfo21.test |   31 +++++++++++++++++++++++++++----
 2 files changed, 36 insertions(+), 4 deletions(-)

diff --git a/ChangeLog b/ChangeLog
index a8e6587..edb2f75 100644
--- a/ChangeLog
+++ b/ChangeLog
@@ -1,3 +1,12 @@
+2011-11-26  Stefano Lattarini  <address@hidden>
+           Peter Rosin  <address@hidden>
+
+       tests: work around a Cygwin limitation in 'txinfo21.test'
+       * tests/txinfo21.test: At least on Cygwin, the timestamp of a
+       directory might not be properly updated when its contents are
+       modified; we need to account for this behaviour.  Also, since
+       we are at it, remove a redundant `$sleep'.
+
 2011-11-24  Stefano Lattarini  <address@hidden>
 
        cosmetics: typofix in comments
diff --git a/tests/txinfo21.test b/tests/txinfo21.test
index ae1d985..e353a15 100755
--- a/tests/txinfo21.test
+++ b/tests/txinfo21.test
@@ -22,6 +22,20 @@ required='makeinfo-html tex texi2dvi'
 
 set -e
 
+# At least on Cygwin, the timestamp of a directory might not be
+# properly updated when the directory contents are changed, and
+# we need to account for that in later checks.
+mkdir tmp.d
+: > tmp.f1
+$sleep
+: > tmp.d/tmp.f2
+if is_newest tmp.d tmp.f1; then
+   have_correct_dir_timestamp=yes
+else
+   have_correct_dir_timestamp=no
+fi
+rm -rf tmp.*
+
 cat >>configure.in <<\EOF
 AC_CONFIG_FILES([rec/Makefile])
 
@@ -87,19 +101,28 @@ $AUTOCONF
 
 $MAKE
 
-$sleep
 # Test production of split-per-node HTML.
 $MAKE html
 test -d main.html
 test -d sub/main2.html
 test -d rec/main3.html
 
-# Rebuilding main.html should cause its timestamp to be updated.
-is_newest main.html main.texi
+# Rebuilding main.html should cause the timestamp of the regular
+# file `main.html/index.html' to be updated.
+is_newest main.html/index.html main.texi
+# Rebuilding main.html should cause its timestamp to be updated
+# (but main.html is a directory, so be prepared to account for the
+# Cygwin limitation described above).
+if test $have_correct_dir_timestamp = yes; then
+  is_newest main.html main.texi || Exit 1
+fi
 $sleep
 touch main.texi
 $MAKE html
-is_newest main.html main.texi
+is_newest main.html/index.html main.texi
+if test $have_correct_dir_timestamp = yes; then
+  is_newest main.html main.texi || Exit 1
+fi
 
 $MAKE clean
 test ! -d main.html
-- 
1.7.2.3