Use os.lstat() instead of os.stat().

I think this aligns better with how redo works.  Otherwise, if a.do
creates a as a symlink, then changes to the symlink's *target* will
change a's stat/stamp information without re-running a.do, which looks
to redo like you modified a by hand, which causes it to stop running
a.do altogether.

With this change, modifications to a's target are okay, but they don't
trigger any redo dependency changes.  If you want that, then a.do
should redo-ifchange on its symlink target explicitly.

t/201-fail/all.do

@@ -10,4 +10,5 @@ rm -f fail
 
 touch fail
 ../flush-cache
+# since we created this file by hand, fail.do won't run, so it won't fail.
 redo-ifchange fail >&/dev/null || exit 55  # expected to pass

t/360-symlinks/.gitignore

@@ -0,0 +1,4 @@
+*.did
+*.extra
+a
+b

t/360-symlinks/a.do

@@ -0,0 +1,4 @@
+echo x >>a.did
+rm -f $2.extra
+echo foo >$2.extra
+ln -s $2.extra $3

t/360-symlinks/all.do

@@ -0,0 +1,32 @@
+rm -f a a.extra b b.did
+d0=""
+redo a
+redo-ifchange b
+d1=$(cat b.did)
+[ "$d0" != "$d1" ] || exit 11
+
+# b only rebuilds if a changes
+../flush-cache
+redo-ifchange b
+d2=$(cat b.did)
+[ "$d1" = "$d2" ] || exit 12
+
+# forcibly changing a should rebuild b.
+# a is already symlink to a.extra, but redo shouldn't care about the
+# target of symlinks, so it shouldn't freak out that a.extra has changed.
+# Anyway, b should still rebuild because a was rebuilt.
+../flush-cache
+redo a
+redo-ifchange b
+d3=$(cat b.did)
+[ "$d2" != "$d3" ] || exit 13
+
+# Explicitly check that changing a's symlink target (a.extra) does *not*
+# trigger a rebuild of b, because b depends on the stamp of the symlink,
+# not what the symlink points to.  In redo, you declare dependencies on
+# specific filenames, not the things they happen to refer to.
+../flush-cache
+touch a.extra
+redo-ifchange b
+d4=$(cat b.did)
+[ "$d3" = "$d4" ] || exit 14

t/360-symlinks/b.do

@@ -0,0 +1,3 @@
+echo x >>b.did
+redo-ifchange a
+cat a >$3

t/360-symlinks/clean.do

@@ -0,0 +1 @@
+rm -f *~ .*~ a b *.extra *.did