AuroraMiddleware/glibc
1
0
Fork 0
mirror of https://sourceware.org/git/glibc.git synced 2024-12-27 21:20:18 +00:00
Code Issues Packages Projects Releases Wiki Activity
a93d9e03a3
glibc/io/ftw.c
Siddhesh Poyarekar 30891f35fa Remove "Contributed by" lines 
We stopped adding "Contributed by" or similar lines in sources in 2012
in favour of git logs and keeping the Contributors section of the
glibc manual up to date.  Removing these lines makes the license
header a bit more consistent across files and also removes the
possibility of error in attribution when license blocks or files are
copied across since the contributed-by lines don't actually reflect
reality in those cases.

Move all "Contributed by" and similar lines (Written by, Test by,
etc.) into a new file CONTRIBUTED-BY to retain record of these
contributions.  These contributors are also mentioned in
manual/contrib.texi, so we just maintain this additional record as a
courtesy to the earlier developers.

The following scripts were used to filter a list of files to edit in
place and to clean up the CONTRIBUTED-BY file respectively.  These
were not added to the glibc sources because they're not expected to be
of any use in future given that this is a one time task:

https://gist.github.com/siddhesh/b5ecac94eabfd72ed2916d6d8157e7dc
https://gist.github.com/siddhesh/15ea1f5e435ace9774f485030695ee02

Reviewed-by: Carlos O'Donell <carlos@redhat.com>
2021-09-03 22:06:44 +05:30

865 lines
21 KiB
C
Raw Blame History

/* File tree walker functions.
Copyright (C) 1996-2021 Free Software Foundation, Inc.
This file is part of the GNU C Library.
The GNU C Library is free software; you can redistribute it and/or
modify it under the terms of the GNU Lesser General Public
License as published by the Free Software Foundation; either
version 2.1 of the License, or (at your option) any later version.
The GNU C Library is distributed in the hope that it will be useful,
but WITHOUT ANY WARRANTY; without even the implied warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public
License along with the GNU C Library; if not, see
<https://www.gnu.org/licenses/>. */
#ifdef HAVE_CONFIG_H
# include <config.h>
#endif
#if __GNUC__
# define alloca __builtin_alloca
#else
# if HAVE_ALLOCA_H
# include <alloca.h>
# else
# ifdef _AIX
# pragma alloca
# else
char *alloca ();
# endif
# endif
#endif
#ifdef _LIBC
# include <dirent.h>
# define NAMLEN(dirent) _D_EXACT_NAMLEN (dirent)
#else
# if HAVE_DIRENT_H
# include <dirent.h>
# define NAMLEN(dirent) strlen ((dirent)->d_name)
# else
# define dirent direct
# define NAMLEN(dirent) (dirent)->d_namlen
# if HAVE_SYS_NDIR_H
# include <sys/ndir.h>
# endif
# if HAVE_SYS_DIR_H
# include <sys/dir.h>
# endif
# if HAVE_NDIR_H
# include <ndir.h>
# endif
# endif
#endif
#include <errno.h>
#include <fcntl.h>
#include <ftw.h>
#include <limits.h>
#include <search.h>
#include <stdlib.h>
#include <string.h>
#include <unistd.h>
#include <not-cancel.h>
#include <sys/param.h>
#ifdef _LIBC
# include <include/sys/stat.h>
#else
# include <sys/stat.h>
#endif
#if ! _LIBC && !HAVE_DECL_STPCPY && !defined stpcpy
char *stpcpy ();
#endif
#if ! _LIBC && ! defined HAVE_MEMPCPY && ! defined mempcpy
/* Be CAREFUL that there are no side effects in N. */
# define mempcpy(D, S, N) ((void *) ((char *) memcpy (D, S, N) + (N)))
#endif
/* #define NDEBUG 1 */
#include <assert.h>
#ifndef _LIBC
# undef __chdir
# define __chdir chdir
# undef __closedir
# define __closedir closedir
# undef __fchdir
# define __fchdir fchdir
# undef __getcwd
# define __getcwd(P, N) xgetcwd ()
extern char *xgetcwd (void);
# undef __mempcpy
# define __mempcpy mempcpy
# undef __opendir
# define __opendir opendir
# undef __readdir64
# define __readdir64 readdir
# undef __stpcpy
# define __stpcpy stpcpy
# undef __tdestroy
# define __tdestroy tdestroy
# undef __tfind
# define __tfind tfind
# undef __tsearch
# define __tsearch tsearch
# undef dirent64
# define dirent64 dirent
# undef MAX
# define MAX(a, b) ((a) > (b) ? (a) : (b))
#endif
/* Arrange to make lstat calls go through the wrapper function
on systems with an lstat function that does not dereference symlinks
that are specified with a trailing slash. */
#if ! _LIBC && ! LSTAT_FOLLOWS_SLASHED_SYMLINK
int rpl_lstat (const char *, struct stat *);
# undef lstat
# define lstat(Name, Stat_buf) rpl_lstat(Name, Stat_buf)
#endif
#ifndef __set_errno
# define __set_errno(Val) errno = (Val)
#endif
/* Support for the LFS API version. */
#ifndef FTW_NAME
# define FTW_NAME ftw
# define NFTW_NAME nftw
# define NFTW_OLD_NAME __old_nftw
# define NFTW_NEW_NAME __new_nftw
# define INO_T ino_t
# define STRUCT_STAT stat
# ifdef _LIBC
# define LSTAT __lstat
# define STAT __stat
# define FSTATAT __fstatat
# else
# define LSTAT lstat
# define XTAT stat
# define FSTATAT fstatat
# endif
# define FTW_FUNC_T __ftw_func_t
# define NFTW_FUNC_T __nftw_func_t
#endif
/* We define PATH_MAX if the system does not provide a definition.
This does not artificially limit any operation. PATH_MAX is simply
used as a guesstimate for the expected maximal path length.
Buffers will be enlarged if necessary. */
#ifndef PATH_MAX
# define PATH_MAX 1024
#endif
struct dir_data
{
DIR *stream;
int streamfd;
char *content;
};
struct known_object
{
dev_t dev;
INO_T ino;
};
struct ftw_data
{
/* Array with pointers to open directory streams. */
struct dir_data **dirstreams;
size_t actdir;
size_t maxdir;
/* Buffer containing name of currently processed object. */
char *dirbuf;
size_t dirbufsize;
/* Passed as fourth argument to `nftw' callback. The `base' member
tracks the content of the `dirbuf'. */
struct FTW ftw;
/* Flags passed to `nftw' function. 0 for `ftw'. */
int flags;
/* Conversion array for flag values. It is the identity mapping for
`nftw' calls, otherwise it maps the values to those known by
`ftw'. */
const int *cvt_arr;
/* Callback function. We always use the `nftw' form. */
NFTW_FUNC_T func;
/* Device of starting point. Needed for FTW_MOUNT. */
dev_t dev;
/* Data structure for keeping fingerprints of already processed
object. This is needed when not using FTW_PHYS. */
void *known_objects;
};
/* Internally we use the FTW_* constants used for `nftw'. When invoked
as `ftw', map each flag to the subset of values used by `ftw'. */
static const int nftw_arr[] =
{
FTW_F, FTW_D, FTW_DNR, FTW_NS, FTW_SL, FTW_DP, FTW_SLN
};
static const int ftw_arr[] =
{
FTW_F, FTW_D, FTW_DNR, FTW_NS, FTW_F, FTW_D, FTW_NS
};
/* Forward declarations of local functions. */
static int ftw_dir (struct ftw_data *data, struct STRUCT_STAT *st,
struct dir_data *old_dir);
static int
object_compare (const void *p1, const void *p2)
{
/* We don't need a sophisticated and useful comparison. We are only
interested in equality. However, we must be careful not to
accidentally compare `holes' in the structure. */
const struct known_object *kp1 = p1, *kp2 = p2;
int cmp1;
cmp1 = (kp1->ino > kp2->ino) - (kp1->ino < kp2->ino);
if (cmp1 != 0)
return cmp1;
return (kp1->dev > kp2->dev) - (kp1->dev < kp2->dev);
}
static int
add_object (struct ftw_data *data, struct STRUCT_STAT *st)
{
struct known_object *newp = malloc (sizeof (struct known_object));
if (newp == NULL)
return -1;
newp->dev = st->st_dev;
newp->ino = st->st_ino;
return __tsearch (newp, &data->known_objects, object_compare) ? 0 : -1;
}
static inline int
find_object (struct ftw_data *data, struct STRUCT_STAT *st)
{
struct known_object obj;
obj.dev = st->st_dev;
obj.ino = st->st_ino;
return __tfind (&obj, &data->known_objects, object_compare) != NULL;
}
static inline int
__attribute ((always_inline))
open_dir_stream (int *dfdp, struct ftw_data *data, struct dir_data *dirp)
{
int result = 0;
if (data->dirstreams[data->actdir] != NULL)
{
/* Oh, oh. We must close this stream. Get all remaining
entries and store them as a list in the `content' member of
the `struct dir_data' variable. */
size_t bufsize = 1024;
char *buf = malloc (bufsize);
if (buf == NULL)
result = -1;
else
{
DIR *st = data->dirstreams[data->actdir]->stream;
struct dirent64 *d;
size_t actsize = 0;
while ((d = __readdir64 (st)) != NULL)
{
size_t this_len = NAMLEN (d);
if (actsize + this_len + 2 >= bufsize)
{
char *newp;
bufsize += MAX (1024, 2 * this_len);
newp = (char *) realloc (buf, bufsize);
if (newp == NULL)
{
/* No more memory. */
int save_err = errno;
free (buf);
__set_errno (save_err);
return -1;
}
buf = newp;
}
*((char *) __mempcpy (buf + actsize, d->d_name, this_len))
= '\0';
actsize += this_len + 1;
}
/* Terminate the list with an additional NUL byte. */
buf[actsize++] = '\0';
/* Shrink the buffer to what we actually need. */
data->dirstreams[data->actdir]->content = realloc (buf, actsize);
if (data->dirstreams[data->actdir]->content == NULL)
{
int save_err = errno;
free (buf);
__set_errno (save_err);
result = -1;
}
else
{
__closedir (st);
data->dirstreams[data->actdir]->stream = NULL;
data->dirstreams[data->actdir]->streamfd = -1;
data->dirstreams[data->actdir] = NULL;
}
}
}
/* Open the new stream. */
if (result == 0)
{
assert (data->dirstreams[data->actdir] == NULL);
if (dfdp != NULL && *dfdp != -1)
{
int fd = __openat64_nocancel (*dfdp, data->dirbuf + data->ftw.base,
O_RDONLY | O_DIRECTORY | O_NDELAY);
dirp->stream = NULL;
if (fd != -1 && (dirp->stream = __fdopendir (fd)) == NULL)
__close_nocancel_nostatus (fd);
}
else
{
const char *name;
if (data->flags & FTW_CHDIR)
{
name = data->dirbuf + data->ftw.base;
if (name[0] == '\0')
name = ".";
}
else
name = data->dirbuf;
dirp->stream = __opendir (name);
}
if (dirp->stream == NULL)
result = -1;
else
{
dirp->streamfd = __dirfd (dirp->stream);
dirp->content = NULL;
data->dirstreams[data->actdir] = dirp;
if (++data->actdir == data->maxdir)
data->actdir = 0;
}
}
return result;
}
static int
process_entry (struct ftw_data *data, struct dir_data *dir, const char *name,
size_t namlen, int d_type)
{
struct STRUCT_STAT st;
int result = 0;
int flag = 0;
size_t new_buflen;
if (name[0] == '.' && (name[1] == '\0'
|| (name[1] == '.' && name[2] == '\0')))
/* Don't process the "." and ".." entries. */
return 0;
new_buflen = data->ftw.base + namlen + 2;
if (data->dirbufsize < new_buflen)
{
/* Enlarge the buffer. */
char *newp;
data->dirbufsize = 2 * new_buflen;
newp = (char *) realloc (data->dirbuf, data->dirbufsize);
if (newp == NULL)
return -1;
data->dirbuf = newp;
}
*((char *) __mempcpy (data->dirbuf + data->ftw.base, name, namlen)) = '\0';
int statres;
if (dir->streamfd != -1)
statres = FSTATAT (dir->streamfd, name, &st,
(data->flags & FTW_PHYS) ? AT_SYMLINK_NOFOLLOW : 0);
else
{
if ((data->flags & FTW_CHDIR) == 0)
name = data->dirbuf;
statres = ((data->flags & FTW_PHYS)
? LSTAT (name, &st)
: STAT (name, &st));
}
if (statres < 0)
{
if (errno != EACCES && errno != ENOENT)
result = -1;
else if (data->flags & FTW_PHYS)
flag = FTW_NS;
else
{
/* Old code left ST undefined for dangling DT_LNK without
FTW_PHYS set; a clarification at the POSIX level suggests
it should contain information about the link (ala lstat).
We do our best to fill in what data we can. */
if (dir->streamfd != -1)
statres = FSTATAT (dir->streamfd, name, &st,
AT_SYMLINK_NOFOLLOW);
else
statres = LSTAT (name, &st);
if (statres == 0 && S_ISLNK (st.st_mode))
flag = FTW_SLN;
else
flag = FTW_NS;
}
}
else
{
if (S_ISDIR (st.st_mode))
flag = FTW_D;
else if (S_ISLNK (st.st_mode))
flag = FTW_SL;
else
flag = FTW_F;
}
if (result == 0
&& (flag == FTW_NS
|| !(data->flags & FTW_MOUNT) || st.st_dev == data->dev))
{
if (flag == FTW_D)
{
if ((data->flags & FTW_PHYS)
|| (!find_object (data, &st)
/* Remember the object. */
&& (result = add_object (data, &st)) == 0))
result = ftw_dir (data, &st, dir);
}
else
result = (*data->func) (data->dirbuf, &st, data->cvt_arr[flag],
&data->ftw);
}
if ((data->flags & FTW_ACTIONRETVAL) && result == FTW_SKIP_SUBTREE)
result = 0;
return result;
}
static int
__attribute ((noinline))
ftw_dir (struct ftw_data *data, struct STRUCT_STAT *st, struct dir_data *old_dir)
{
struct dir_data dir;
struct dirent64 *d;
int previous_base = data->ftw.base;
int result;
char *startp;
/* Open the stream for this directory. This might require that
another stream has to be closed. */
result = open_dir_stream (old_dir == NULL ? NULL : &old_dir->streamfd,
data, &dir);
if (result != 0)
{
if (errno == EACCES)
/* We cannot read the directory. Signal this with a special flag. */
result = (*data->func) (data->dirbuf, st, FTW_DNR, &data->ftw);
return result;
}
/* First, report the directory (if not depth-first). */
if (!(data->flags & FTW_DEPTH))
{
result = (*data->func) (data->dirbuf, st, FTW_D, &data->ftw);
if (result != 0)
{
int save_err;
fail:
save_err = errno;
__closedir (dir.stream);
dir.streamfd = -1;
__set_errno (save_err);
if (data->actdir-- == 0)
data->actdir = data->maxdir - 1;
data->dirstreams[data->actdir] = NULL;
return result;
}
}
/* If necessary, change to this directory. */
if (data->flags & FTW_CHDIR)
{
if (__fchdir (__dirfd (dir.stream)) < 0)
{
result = -1;
goto fail;
}
}
/* Next, update the `struct FTW' information. */
++data->ftw.level;
startp = __rawmemchr (data->dirbuf, '\0');
/* There always must be a directory name. */
assert (startp != data->dirbuf);
if (startp[-1] != '/')
*startp++ = '/';
data->ftw.base = startp - data->dirbuf;
while (dir.stream != NULL && (d = __readdir64 (dir.stream)) != NULL)
{
int d_type = DT_UNKNOWN;
#ifdef _DIRENT_HAVE_D_TYPE
d_type = d->d_type;
#endif
result = process_entry (data, &dir, d->d_name, NAMLEN (d), d_type);
if (result != 0)
break;
}
if (dir.stream != NULL)
{
/* The stream is still open. I.e., we did not need more
descriptors. Simply close the stream now. */
int save_err = errno;
assert (dir.content == NULL);
__closedir (dir.stream);
dir.streamfd = -1;
__set_errno (save_err);
if (data->actdir-- == 0)
data->actdir = data->maxdir - 1;
data->dirstreams[data->actdir] = NULL;
}
else
{
int save_err;
char *runp = dir.content;
while (result == 0 && *runp != '\0')
{
char *endp = strchr (runp, '\0');
// XXX Should store the d_type values as well?!
result = process_entry (data, &dir, runp, endp - runp, DT_UNKNOWN);
runp = endp + 1;
}
save_err = errno;
free (dir.content);
__set_errno (save_err);
}
if ((data->flags & FTW_ACTIONRETVAL) && result == FTW_SKIP_SIBLINGS)
result = 0;
/* Prepare the return, revert the `struct FTW' information. */
data->dirbuf[data->ftw.base - 1] = '\0';
--data->ftw.level;
data->ftw.base = previous_base;
/* Finally, if we process depth-first report the directory. */
if (result == 0 && (data->flags & FTW_DEPTH))
result = (*data->func) (data->dirbuf, st, FTW_DP, &data->ftw);
if (old_dir
&& (data->flags & FTW_CHDIR)
&& (result == 0
|| ((data->flags & FTW_ACTIONRETVAL)
&& (result != -1 && result != FTW_STOP))))
{
/* Change back to the parent directory. */
int done = 0;
if (old_dir->stream != NULL)
if (__fchdir (__dirfd (old_dir->stream)) == 0)
done = 1;
if (!done)
{
if (data->ftw.base == 1)
{
if (__chdir ("/") < 0)
result = -1;
}
else
if (__chdir ("..") < 0)
result = -1;
}
}
return result;
}
static int
__attribute ((noinline))
ftw_startup (const char *dir, int is_nftw, void *func, int descriptors,
int flags)
{
struct ftw_data data;
struct STRUCT_STAT st;
int result = 0;
int save_err;
int cwdfd = -1;
char *cwd = NULL;
char *cp;
/* First make sure the parameters are reasonable. */
if (dir[0] == '\0')
{
__set_errno (ENOENT);
return -1;
}
data.maxdir = descriptors < 1 ? 1 : descriptors;
data.actdir = 0;
/* PATH_MAX is always defined when we get here. */
data.dirbufsize = MAX (2 * strlen (dir), PATH_MAX);
data.dirstreams = malloc (data.maxdir * sizeof (struct dir_data *)
+ data.dirbufsize);
if (data.dirstreams == NULL)
return -1;
memset (data.dirstreams, '\0', data.maxdir * sizeof (struct dir_data *));
data.dirbuf = (char *) data.dirstreams
+ data.maxdir * sizeof (struct dir_data *);
cp = __stpcpy (data.dirbuf, dir);
/* Strip trailing slashes. */
while (cp > data.dirbuf + 1 && cp[-1] == '/')
--cp;
*cp = '\0';
data.ftw.level = 0;
/* Find basename. */
while (cp > data.dirbuf && cp[-1] != '/')
--cp;
data.ftw.base = cp - data.dirbuf;
data.flags = flags;
/* This assignment might seem to be strange but it is what we want.
The trick is that the first three arguments to the `ftw' and
`nftw' callback functions are equal. Therefore we can call in
every case the callback using the format of the `nftw' version
and get the correct result since the stack layout for a function
call in C allows this. */
data.func = (NFTW_FUNC_T) func;
/* Since we internally use the complete set of FTW_* values we need
to reduce the value range before calling a `ftw' callback. */
data.cvt_arr = is_nftw ? nftw_arr : ftw_arr;
/* No object known so far. */
data.known_objects = NULL;
/* Now go to the directory containing the initial file/directory. */
if (flags & FTW_CHDIR)
{
/* We have to be able to go back to the current working
directory. The best way to do this is to use a file
descriptor. */
cwdfd = __open (".", O_RDONLY | O_DIRECTORY);
if (cwdfd == -1)
{
/* Try getting the directory name. This can be needed if
the current directory is executable but not readable. */
if (errno == EACCES)
/* GNU extension ahead. */
cwd = __getcwd (NULL, 0);
if (cwd == NULL)
goto out_fail;
}
else if (data.maxdir > 1)
/* Account for the file descriptor we use here. */
--data.maxdir;
if (data.ftw.base > 0)
{
/* Change to the directory the file is in. In data.dirbuf
we have a writable copy of the file name. Just NUL
terminate it for now and change the directory. */
if (data.ftw.base == 1)
/* I.e., the file is in the root directory. */
result = __chdir ("/");
else
{
char ch = data.dirbuf[data.ftw.base - 1];
data.dirbuf[data.ftw.base - 1] = '\0';
result = __chdir (data.dirbuf);
data.dirbuf[data.ftw.base - 1] = ch;
}
}
}
/* Get stat info for start directory. */
if (result == 0)
{
const char *name;
if (data.flags & FTW_CHDIR)
{
name = data.dirbuf + data.ftw.base;
if (name[0] == '\0')
name = ".";
}
else
name = data.dirbuf;
if (((flags & FTW_PHYS)
? LSTAT (name, &st)
: STAT (name, &st)) < 0)
{
if (!(flags & FTW_PHYS)
&& errno == ENOENT
&& LSTAT (name, &st) == 0
&& S_ISLNK (st.st_mode))
result = (*data.func) (data.dirbuf, &st, data.cvt_arr[FTW_SLN],
&data.ftw);
else
/* No need to call the callback since we cannot say anything
about the object. */
result = -1;
}
else
{
if (S_ISDIR (st.st_mode))
{
/* Remember the device of the initial directory in case
FTW_MOUNT is given. */
data.dev = st.st_dev;
/* We know this directory now. */
if (!(flags & FTW_PHYS))
result = add_object (&data, &st);
if (result == 0)
result = ftw_dir (&data, &st, NULL);
}
else
{
int flag = S_ISLNK (st.st_mode) ? FTW_SL : FTW_F;
result = (*data.func) (data.dirbuf, &st, data.cvt_arr[flag],
&data.ftw);
}
}
if ((flags & FTW_ACTIONRETVAL)
&& (result == FTW_SKIP_SUBTREE || result == FTW_SKIP_SIBLINGS))
result = 0;
}
/* Return to the start directory (if necessary). */
if (cwdfd != -1)
{
int save_err = errno;
__fchdir (cwdfd);
__close_nocancel_nostatus (cwdfd);
__set_errno (save_err);
}
else if (cwd != NULL)
{
int save_err = errno;
__chdir (cwd);
free (cwd);
__set_errno (save_err);
}
/* Free all memory. */
out_fail:
save_err = errno;
__tdestroy (data.known_objects, free);
free (data.dirstreams);
__set_errno (save_err);
return result;
}
/* Entry points. */
int
FTW_NAME (const char *path, FTW_FUNC_T func, int descriptors)
{
return ftw_startup (path, 0, func, descriptors, 0);
}
#ifndef NFTW_OLD_NAME
int
NFTW_NAME (const char *path, NFTW_FUNC_T func, int descriptors, int flags)
{
return ftw_startup (path, 1, func, descriptors, flags);
}
#else
# include <shlib-compat.h>
int NFTW_NEW_NAME (const char *, NFTW_FUNC_T, int, int);
int
NFTW_NEW_NAME (const char *path, NFTW_FUNC_T func, int descriptors, int flags)
{
if (flags
& ~(FTW_PHYS | FTW_MOUNT | FTW_CHDIR | FTW_DEPTH | FTW_ACTIONRETVAL))
{
__set_errno (EINVAL);
return -1;
}
return ftw_startup (path, 1, func, descriptors, flags);
}
versioned_symbol (libc, NFTW_NEW_NAME, NFTW_NAME, GLIBC_2_3_3);
# if SHLIB_COMPAT(libc, GLIBC_2_1, GLIBC_2_3_3)
/* Older nftw* version just ignored all unknown flags. */
int NFTW_OLD_NAME (const char *, NFTW_FUNC_T, int, int);
int
attribute_compat_text_section
NFTW_OLD_NAME (const char *path, NFTW_FUNC_T func, int descriptors, int flags)
{
flags &= (FTW_PHYS | FTW_MOUNT | FTW_CHDIR | FTW_DEPTH);
return ftw_startup (path, 1, func, descriptors, flags);
}
compat_symbol (libc, NFTW_OLD_NAME, NFTW_NAME, GLIBC_2_1);
# endif
#endif /* NFTW_OLD_NAME */
Reference in New Issue View Git Blame Copy Permalink