Here is a patch to tar that fixes the problem by using mtime instead of ctime.
Yes, it's not a problem with cygwin or tar, but I see no way for users to
figure out, under Windows, which processes are changing the status of their
files.
diff -ur tar-1.25-copy/src/extract.c tar-1.25/src/extract.c
--- tar-1.25-copy/src/extract.c 2010-11-01 15:34:59.000000000 -0500
+++ tar-1.25/src/extract.c 2011-04-27 09:12:58.941555800 -0500
@@ -115,7 +115,10 @@
other process removes the placeholder. */
dev_t dev;
ino_t ino;
- struct timespec ctime;
+ /* for cygwin, we use mtime instead of ctime of the
+ placeholder, because background processes running
+ under Windows may change it */
+ struct timespec mtime0;
/* True if the link is symbolic. */
bool is_symlink;
@@ -1064,7 +1067,7 @@
delayed_link_head = p;
p->dev = st.st_dev;
p->ino = st.st_ino;
- p->ctime = get_stat_ctime (&st);
+ p->mtime0 = get_stat_mtime (&st);
p->is_symlink = is_symlink;
if (is_symlink)
{
@@ -1122,7 +1125,7 @@
if (ds->change_dir == chdir_current
&& ds->dev == st1.st_dev
&& ds->ino == st1.st_ino
- && timespec_cmp (ds->ctime, get_stat_ctime (&st1)) == 0)
+ && timespec_cmp (ds->mtime0, get_stat_mtime (&st1)) == 0)
{
struct string_list *p = xmalloc (offsetof (struct
string_list, string)
+ strlen (file_name) + 1);
@@ -1488,7 +1491,7 @@
if (fstatat (chdir_fd, source, &st, AT_SYMLINK_NOFOLLOW) == 0
&& st.st_dev == ds->dev
&& st.st_ino == ds->ino
- && timespec_cmp (get_stat_ctime (&st), ds->ctime) == 0)
+ && timespec_cmp (get_stat_mtime (&st), ds->mtime0) == 0)
{
/* Unlink the placeholder, then create a hard link if possible,
a symbolic link otherwise. */
--
Problem reports: http://cygwin.com/problems.html
FAQ: http://cygwin.com/faq/
Documentation: http://cygwin.com/docs.html
Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
