Also introduce library functions for copying extended attributes from one
file to another:
- extattr_copy_file, extattr_copy_fd, extattr_copy_link, with FreeBSD style,
where a namespace is to be supplied
- cpxattr, fcpxattr, lcpxattr, with Linux style, where all namespaces
accessible to the caller are copied, and the others are silently ignored.
Write out the mmap-ed data in small chunks to avoid locking the output
file for a long time.
Suggested by David Holland on "source-changes-d" mailing list.
a very annoying property: if the source media is slow (like a slow
network), the target file will be locked for the duration of the
entire max 8MB write and cause processes attempting to e.g. stat()
it to "tstile" (for several minutes in the worst case). Revisit
this if/when vnode locking gets a little smarter.
chs ok (although he would rather see vnode locking improvements,
but that's a slightly larger project)
"-R" claims:
-R [...] Created directories have the same mode as the corre-
sponding source directory, unmodified
by the process' umask.
Make this actually true.
In addition, make '-P' (no symbolic links are followed) apply even if
'-R' is not specified. This allows users to overwrite symbolic links
with files and/or to copy symbolic links over a file without indirecting
through the link (ie a copy of a link turns the target into a link, not
a copy of the file pointed to by the source).
If stdout is a tty, use vis(3) to print any filenames to prevent garbage
from being printed if the filename contains control- or other non-printable
characters.
While here, sprinkle some EXIT_FAILURE and NOTREACHED where appropriate.
to what the manpage and POSIX say) copied as symlinks, even without -R.
Return ENOENT instead. Closes PR 6975 by Johan Danielsson <joda@pdc.kth.se>.
From FreeBSD: return correct error message if source directory is
unreadable; remove unnecessarily included headers.
It's suppsed to remove the file and then copy, which it wasn't doing.
But no wait, it turns out that the described behaviour in the manual doesn't
agree with POSIX. So we change the above fix and the manual to "try copy, and
if fail, try remove, then copy".
Fix bug where "cp -R" didn't work on read-only directories:
It would make the directory, set the mode, and not be able to write files into it.
Don't bother mmap()ing files of zero length. Was a workaround for a bug in Rhapsody
mmap(), which didn't get along with such files, but makes sense anyway.
Fix race condition where "cp -p" would set the mod time of a file before close()ing
the file, which would update the mod time and therefore screw up the "-p" idea,
except, of course, while running in gdb, which sucked.
Add -f option to usage message in binary and man page. Already documented in man page.
>NFS doesn't support chflags; ignore errors unless there's reason
>to believe we're losing bits. (Note, this still won't be right
>if the server supports flags and we were trying to *remove* flags
>on a file that we copied, i.e., that we didn't create.)
very bad if it is actually a device or fifo. POSIX.2 requires that a
diagnostic message be written, the file descriptor be closed, and for
cp to continue on with the rest of the files.
