111a5d7c78
the -n option is specified and the destination file exists. PR: bin/174489 Approved by: rstone (co-mentor) MFC after: 2 weeks
543 lines
14 KiB
C
543 lines
14 KiB
C
/*-
|
|
* Copyright (c) 1991, 1993, 1994
|
|
* The Regents of the University of California. All rights reserved.
|
|
*
|
|
* Redistribution and use in source and binary forms, with or without
|
|
* modification, are permitted provided that the following conditions
|
|
* are met:
|
|
* 1. Redistributions of source code must retain the above copyright
|
|
* notice, this list of conditions and the following disclaimer.
|
|
* 2. Redistributions in binary form must reproduce the above copyright
|
|
* notice, this list of conditions and the following disclaimer in the
|
|
* documentation and/or other materials provided with the distribution.
|
|
* 4. Neither the name of the University nor the names of its contributors
|
|
* may be used to endorse or promote products derived from this software
|
|
* without specific prior written permission.
|
|
*
|
|
* THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
|
|
* ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
|
|
* IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
|
|
* ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
|
|
* FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
|
|
* DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
|
|
* OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
|
|
* HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
|
|
* LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
|
|
* OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
|
|
* SUCH DAMAGE.
|
|
*/
|
|
|
|
#ifndef lint
|
|
#if 0
|
|
static char sccsid[] = "@(#)utils.c 8.3 (Berkeley) 4/1/94";
|
|
#endif
|
|
#endif /* not lint */
|
|
#include <sys/cdefs.h>
|
|
__FBSDID("$FreeBSD$");
|
|
|
|
#include <sys/types.h>
|
|
#include <sys/acl.h>
|
|
#include <sys/param.h>
|
|
#include <sys/stat.h>
|
|
#ifdef VM_AND_BUFFER_CACHE_SYNCHRONIZED
|
|
#include <sys/mman.h>
|
|
#endif
|
|
|
|
#include <err.h>
|
|
#include <errno.h>
|
|
#include <fcntl.h>
|
|
#include <fts.h>
|
|
#include <limits.h>
|
|
#include <stdio.h>
|
|
#include <stdlib.h>
|
|
#include <sysexits.h>
|
|
#include <unistd.h>
|
|
|
|
#include "extern.h"
|
|
|
|
#define cp_pct(x, y) ((y == 0) ? 0 : (int)(100.0 * (x) / (y)))
|
|
|
|
/* Memory strategy threshold, in pages: if physmem is larger then this, use a
|
|
* large buffer */
|
|
#define PHYSPAGES_THRESHOLD (32*1024)
|
|
|
|
/* Maximum buffer size in bytes - do not allow it to grow larger than this */
|
|
#define BUFSIZE_MAX (2*1024*1024)
|
|
|
|
/* Small (default) buffer size in bytes. It's inefficient for this to be
|
|
* smaller than MAXPHYS */
|
|
#define BUFSIZE_SMALL (MAXPHYS)
|
|
|
|
int
|
|
copy_file(const FTSENT *entp, int dne)
|
|
{
|
|
static char *buf = NULL;
|
|
static size_t bufsize;
|
|
struct stat *fs;
|
|
ssize_t wcount;
|
|
size_t wresid;
|
|
off_t wtotal;
|
|
int ch, checkch, from_fd = 0, rcount, rval, to_fd = 0;
|
|
char *bufp;
|
|
#ifdef VM_AND_BUFFER_CACHE_SYNCHRONIZED
|
|
char *p;
|
|
#endif
|
|
|
|
if ((from_fd = open(entp->fts_path, O_RDONLY, 0)) == -1) {
|
|
warn("%s", entp->fts_path);
|
|
return (1);
|
|
}
|
|
|
|
fs = entp->fts_statp;
|
|
|
|
/*
|
|
* If the file exists and we're interactive, verify with the user.
|
|
* If the file DNE, set the mode to be the from file, minus setuid
|
|
* bits, modified by the umask; arguably wrong, but it makes copying
|
|
* executables work right and it's been that way forever. (The
|
|
* other choice is 666 or'ed with the execute bits on the from file
|
|
* modified by the umask.)
|
|
*/
|
|
if (!dne) {
|
|
#define YESNO "(y/n [n]) "
|
|
if (nflag) {
|
|
if (vflag)
|
|
printf("%s not overwritten\n", to.p_path);
|
|
(void)close(from_fd);
|
|
return (1);
|
|
} else if (iflag) {
|
|
(void)fprintf(stderr, "overwrite %s? %s",
|
|
to.p_path, YESNO);
|
|
checkch = ch = getchar();
|
|
while (ch != '\n' && ch != EOF)
|
|
ch = getchar();
|
|
if (checkch != 'y' && checkch != 'Y') {
|
|
(void)close(from_fd);
|
|
(void)fprintf(stderr, "not overwritten\n");
|
|
return (1);
|
|
}
|
|
}
|
|
|
|
if (fflag) {
|
|
/* remove existing destination file name,
|
|
* create a new file */
|
|
(void)unlink(to.p_path);
|
|
if (!lflag)
|
|
to_fd = open(to.p_path, O_WRONLY | O_TRUNC | O_CREAT,
|
|
fs->st_mode & ~(S_ISUID | S_ISGID));
|
|
} else {
|
|
if (!lflag)
|
|
/* overwrite existing destination file name */
|
|
to_fd = open(to.p_path, O_WRONLY | O_TRUNC, 0);
|
|
}
|
|
} else {
|
|
if (!lflag)
|
|
to_fd = open(to.p_path, O_WRONLY | O_TRUNC | O_CREAT,
|
|
fs->st_mode & ~(S_ISUID | S_ISGID));
|
|
}
|
|
|
|
if (to_fd == -1) {
|
|
warn("%s", to.p_path);
|
|
(void)close(from_fd);
|
|
return (1);
|
|
}
|
|
|
|
rval = 0;
|
|
|
|
if (!lflag) {
|
|
/*
|
|
* Mmap and write if less than 8M (the limit is so we don't totally
|
|
* trash memory on big files. This is really a minor hack, but it
|
|
* wins some CPU back.
|
|
* Some filesystems, such as smbnetfs, don't support mmap,
|
|
* so this is a best-effort attempt.
|
|
*/
|
|
#ifdef VM_AND_BUFFER_CACHE_SYNCHRONIZED
|
|
if (S_ISREG(fs->st_mode) && fs->st_size > 0 &&
|
|
fs->st_size <= 8 * 1024 * 1024 &&
|
|
(p = mmap(NULL, (size_t)fs->st_size, PROT_READ,
|
|
MAP_SHARED, from_fd, (off_t)0)) != MAP_FAILED) {
|
|
wtotal = 0;
|
|
for (bufp = p, wresid = fs->st_size; ;
|
|
bufp += wcount, wresid -= (size_t)wcount) {
|
|
wcount = write(to_fd, bufp, wresid);
|
|
if (wcount <= 0)
|
|
break;
|
|
wtotal += wcount;
|
|
if (info) {
|
|
info = 0;
|
|
(void)fprintf(stderr,
|
|
"%s -> %s %3d%%\n",
|
|
entp->fts_path, to.p_path,
|
|
cp_pct(wtotal, fs->st_size));
|
|
}
|
|
if (wcount >= (ssize_t)wresid)
|
|
break;
|
|
}
|
|
if (wcount != (ssize_t)wresid) {
|
|
warn("%s", to.p_path);
|
|
rval = 1;
|
|
}
|
|
/* Some systems don't unmap on close(2). */
|
|
if (munmap(p, fs->st_size) < 0) {
|
|
warn("%s", entp->fts_path);
|
|
rval = 1;
|
|
}
|
|
} else
|
|
#endif
|
|
{
|
|
if (buf == NULL) {
|
|
/*
|
|
* Note that buf and bufsize are static. If
|
|
* malloc() fails, it will fail at the start
|
|
* and not copy only some files.
|
|
*/
|
|
if (sysconf(_SC_PHYS_PAGES) >
|
|
PHYSPAGES_THRESHOLD)
|
|
bufsize = MIN(BUFSIZE_MAX, MAXPHYS * 8);
|
|
else
|
|
bufsize = BUFSIZE_SMALL;
|
|
buf = malloc(bufsize);
|
|
if (buf == NULL)
|
|
err(1, "Not enough memory");
|
|
}
|
|
wtotal = 0;
|
|
while ((rcount = read(from_fd, buf, bufsize)) > 0) {
|
|
for (bufp = buf, wresid = rcount; ;
|
|
bufp += wcount, wresid -= wcount) {
|
|
wcount = write(to_fd, bufp, wresid);
|
|
if (wcount <= 0)
|
|
break;
|
|
wtotal += wcount;
|
|
if (info) {
|
|
info = 0;
|
|
(void)fprintf(stderr,
|
|
"%s -> %s %3d%%\n",
|
|
entp->fts_path, to.p_path,
|
|
cp_pct(wtotal, fs->st_size));
|
|
}
|
|
if (wcount >= (ssize_t)wresid)
|
|
break;
|
|
}
|
|
if (wcount != (ssize_t)wresid) {
|
|
warn("%s", to.p_path);
|
|
rval = 1;
|
|
break;
|
|
}
|
|
}
|
|
if (rcount < 0) {
|
|
warn("%s", entp->fts_path);
|
|
rval = 1;
|
|
}
|
|
}
|
|
} else {
|
|
if (link(entp->fts_path, to.p_path)) {
|
|
warn("%s", to.p_path);
|
|
rval = 1;
|
|
}
|
|
}
|
|
|
|
/*
|
|
* Don't remove the target even after an error. The target might
|
|
* not be a regular file, or its attributes might be important,
|
|
* or its contents might be irreplaceable. It would only be safe
|
|
* to remove it if we created it and its length is 0.
|
|
*/
|
|
|
|
if (!lflag) {
|
|
if (pflag && setfile(fs, to_fd))
|
|
rval = 1;
|
|
if (pflag && preserve_fd_acls(from_fd, to_fd) != 0)
|
|
rval = 1;
|
|
if (close(to_fd)) {
|
|
warn("%s", to.p_path);
|
|
rval = 1;
|
|
}
|
|
}
|
|
|
|
(void)close(from_fd);
|
|
|
|
return (rval);
|
|
}
|
|
|
|
int
|
|
copy_link(const FTSENT *p, int exists)
|
|
{
|
|
int len;
|
|
char llink[PATH_MAX];
|
|
|
|
if (exists && nflag) {
|
|
if (vflag)
|
|
printf("%s not overwritten\n", to.p_path);
|
|
return (1);
|
|
}
|
|
if ((len = readlink(p->fts_path, llink, sizeof(llink) - 1)) == -1) {
|
|
warn("readlink: %s", p->fts_path);
|
|
return (1);
|
|
}
|
|
llink[len] = '\0';
|
|
if (exists && unlink(to.p_path)) {
|
|
warn("unlink: %s", to.p_path);
|
|
return (1);
|
|
}
|
|
if (symlink(llink, to.p_path)) {
|
|
warn("symlink: %s", llink);
|
|
return (1);
|
|
}
|
|
return (pflag ? setfile(p->fts_statp, -1) : 0);
|
|
}
|
|
|
|
int
|
|
copy_fifo(struct stat *from_stat, int exists)
|
|
{
|
|
|
|
if (exists && nflag) {
|
|
if (vflag)
|
|
printf("%s not overwritten\n", to.p_path);
|
|
return (1);
|
|
}
|
|
if (exists && unlink(to.p_path)) {
|
|
warn("unlink: %s", to.p_path);
|
|
return (1);
|
|
}
|
|
if (mkfifo(to.p_path, from_stat->st_mode)) {
|
|
warn("mkfifo: %s", to.p_path);
|
|
return (1);
|
|
}
|
|
return (pflag ? setfile(from_stat, -1) : 0);
|
|
}
|
|
|
|
int
|
|
copy_special(struct stat *from_stat, int exists)
|
|
{
|
|
|
|
if (exists && nflag) {
|
|
if (vflag)
|
|
printf("%s not overwritten\n", to.p_path);
|
|
return (1);
|
|
}
|
|
if (exists && unlink(to.p_path)) {
|
|
warn("unlink: %s", to.p_path);
|
|
return (1);
|
|
}
|
|
if (mknod(to.p_path, from_stat->st_mode, from_stat->st_rdev)) {
|
|
warn("mknod: %s", to.p_path);
|
|
return (1);
|
|
}
|
|
return (pflag ? setfile(from_stat, -1) : 0);
|
|
}
|
|
|
|
int
|
|
setfile(struct stat *fs, int fd)
|
|
{
|
|
static struct timeval tv[2];
|
|
struct stat ts;
|
|
int rval, gotstat, islink, fdval;
|
|
|
|
rval = 0;
|
|
fdval = fd != -1;
|
|
islink = !fdval && S_ISLNK(fs->st_mode);
|
|
fs->st_mode &= S_ISUID | S_ISGID | S_ISVTX |
|
|
S_IRWXU | S_IRWXG | S_IRWXO;
|
|
|
|
TIMESPEC_TO_TIMEVAL(&tv[0], &fs->st_atim);
|
|
TIMESPEC_TO_TIMEVAL(&tv[1], &fs->st_mtim);
|
|
if (islink ? lutimes(to.p_path, tv) : utimes(to.p_path, tv)) {
|
|
warn("%sutimes: %s", islink ? "l" : "", to.p_path);
|
|
rval = 1;
|
|
}
|
|
if (fdval ? fstat(fd, &ts) :
|
|
(islink ? lstat(to.p_path, &ts) : stat(to.p_path, &ts)))
|
|
gotstat = 0;
|
|
else {
|
|
gotstat = 1;
|
|
ts.st_mode &= S_ISUID | S_ISGID | S_ISVTX |
|
|
S_IRWXU | S_IRWXG | S_IRWXO;
|
|
}
|
|
/*
|
|
* Changing the ownership probably won't succeed, unless we're root
|
|
* or POSIX_CHOWN_RESTRICTED is not set. Set uid/gid before setting
|
|
* the mode; current BSD behavior is to remove all setuid bits on
|
|
* chown. If chown fails, lose setuid/setgid bits.
|
|
*/
|
|
if (!gotstat || fs->st_uid != ts.st_uid || fs->st_gid != ts.st_gid)
|
|
if (fdval ? fchown(fd, fs->st_uid, fs->st_gid) :
|
|
(islink ? lchown(to.p_path, fs->st_uid, fs->st_gid) :
|
|
chown(to.p_path, fs->st_uid, fs->st_gid))) {
|
|
if (errno != EPERM) {
|
|
warn("chown: %s", to.p_path);
|
|
rval = 1;
|
|
}
|
|
fs->st_mode &= ~(S_ISUID | S_ISGID);
|
|
}
|
|
|
|
if (!gotstat || fs->st_mode != ts.st_mode)
|
|
if (fdval ? fchmod(fd, fs->st_mode) :
|
|
(islink ? lchmod(to.p_path, fs->st_mode) :
|
|
chmod(to.p_path, fs->st_mode))) {
|
|
warn("chmod: %s", to.p_path);
|
|
rval = 1;
|
|
}
|
|
|
|
if (!gotstat || fs->st_flags != ts.st_flags)
|
|
if (fdval ?
|
|
fchflags(fd, fs->st_flags) :
|
|
(islink ? lchflags(to.p_path, fs->st_flags) :
|
|
chflags(to.p_path, fs->st_flags))) {
|
|
warn("chflags: %s", to.p_path);
|
|
rval = 1;
|
|
}
|
|
|
|
return (rval);
|
|
}
|
|
|
|
int
|
|
preserve_fd_acls(int source_fd, int dest_fd)
|
|
{
|
|
acl_t acl;
|
|
acl_type_t acl_type;
|
|
int acl_supported = 0, ret, trivial;
|
|
|
|
ret = fpathconf(source_fd, _PC_ACL_NFS4);
|
|
if (ret > 0 ) {
|
|
acl_supported = 1;
|
|
acl_type = ACL_TYPE_NFS4;
|
|
} else if (ret < 0 && errno != EINVAL) {
|
|
warn("fpathconf(..., _PC_ACL_NFS4) failed for %s", to.p_path);
|
|
return (1);
|
|
}
|
|
if (acl_supported == 0) {
|
|
ret = fpathconf(source_fd, _PC_ACL_EXTENDED);
|
|
if (ret > 0 ) {
|
|
acl_supported = 1;
|
|
acl_type = ACL_TYPE_ACCESS;
|
|
} else if (ret < 0 && errno != EINVAL) {
|
|
warn("fpathconf(..., _PC_ACL_EXTENDED) failed for %s",
|
|
to.p_path);
|
|
return (1);
|
|
}
|
|
}
|
|
if (acl_supported == 0)
|
|
return (0);
|
|
|
|
acl = acl_get_fd_np(source_fd, acl_type);
|
|
if (acl == NULL) {
|
|
warn("failed to get acl entries while setting %s", to.p_path);
|
|
return (1);
|
|
}
|
|
if (acl_is_trivial_np(acl, &trivial)) {
|
|
warn("acl_is_trivial() failed for %s", to.p_path);
|
|
acl_free(acl);
|
|
return (1);
|
|
}
|
|
if (trivial) {
|
|
acl_free(acl);
|
|
return (0);
|
|
}
|
|
if (acl_set_fd_np(dest_fd, acl, acl_type) < 0) {
|
|
warn("failed to set acl entries for %s", to.p_path);
|
|
acl_free(acl);
|
|
return (1);
|
|
}
|
|
acl_free(acl);
|
|
return (0);
|
|
}
|
|
|
|
int
|
|
preserve_dir_acls(struct stat *fs, char *source_dir, char *dest_dir)
|
|
{
|
|
acl_t (*aclgetf)(const char *, acl_type_t);
|
|
int (*aclsetf)(const char *, acl_type_t, acl_t);
|
|
struct acl *aclp;
|
|
acl_t acl;
|
|
acl_type_t acl_type;
|
|
int acl_supported = 0, ret, trivial;
|
|
|
|
ret = pathconf(source_dir, _PC_ACL_NFS4);
|
|
if (ret > 0) {
|
|
acl_supported = 1;
|
|
acl_type = ACL_TYPE_NFS4;
|
|
} else if (ret < 0 && errno != EINVAL) {
|
|
warn("fpathconf(..., _PC_ACL_NFS4) failed for %s", source_dir);
|
|
return (1);
|
|
}
|
|
if (acl_supported == 0) {
|
|
ret = pathconf(source_dir, _PC_ACL_EXTENDED);
|
|
if (ret > 0) {
|
|
acl_supported = 1;
|
|
acl_type = ACL_TYPE_ACCESS;
|
|
} else if (ret < 0 && errno != EINVAL) {
|
|
warn("fpathconf(..., _PC_ACL_EXTENDED) failed for %s",
|
|
source_dir);
|
|
return (1);
|
|
}
|
|
}
|
|
if (acl_supported == 0)
|
|
return (0);
|
|
|
|
/*
|
|
* If the file is a link we will not follow it
|
|
*/
|
|
if (S_ISLNK(fs->st_mode)) {
|
|
aclgetf = acl_get_link_np;
|
|
aclsetf = acl_set_link_np;
|
|
} else {
|
|
aclgetf = acl_get_file;
|
|
aclsetf = acl_set_file;
|
|
}
|
|
if (acl_type == ACL_TYPE_ACCESS) {
|
|
/*
|
|
* Even if there is no ACL_TYPE_DEFAULT entry here, a zero
|
|
* size ACL will be returned. So it is not safe to simply
|
|
* check the pointer to see if the default ACL is present.
|
|
*/
|
|
acl = aclgetf(source_dir, ACL_TYPE_DEFAULT);
|
|
if (acl == NULL) {
|
|
warn("failed to get default acl entries on %s",
|
|
source_dir);
|
|
return (1);
|
|
}
|
|
aclp = &acl->ats_acl;
|
|
if (aclp->acl_cnt != 0 && aclsetf(dest_dir,
|
|
ACL_TYPE_DEFAULT, acl) < 0) {
|
|
warn("failed to set default acl entries on %s",
|
|
dest_dir);
|
|
acl_free(acl);
|
|
return (1);
|
|
}
|
|
acl_free(acl);
|
|
}
|
|
acl = aclgetf(source_dir, acl_type);
|
|
if (acl == NULL) {
|
|
warn("failed to get acl entries on %s", source_dir);
|
|
return (1);
|
|
}
|
|
if (acl_is_trivial_np(acl, &trivial)) {
|
|
warn("acl_is_trivial() failed on %s", source_dir);
|
|
acl_free(acl);
|
|
return (1);
|
|
}
|
|
if (trivial) {
|
|
acl_free(acl);
|
|
return (0);
|
|
}
|
|
if (aclsetf(dest_dir, acl_type, acl) < 0) {
|
|
warn("failed to set acl entries on %s", dest_dir);
|
|
acl_free(acl);
|
|
return (1);
|
|
}
|
|
acl_free(acl);
|
|
return (0);
|
|
}
|
|
|
|
void
|
|
usage(void)
|
|
{
|
|
|
|
(void)fprintf(stderr, "%s\n%s\n",
|
|
"usage: cp [-R [-H | -L | -P]] [-f | -i | -n] [-alpvx] source_file target_file",
|
|
" cp [-R [-H | -L | -P]] [-f | -i | -n] [-alpvx] source_file ... "
|
|
"target_directory");
|
|
exit(EX_USAGE);
|
|
}
|