+#define _LARGEFILE64_SOURCE 1
+#undef NDEBUG
+#include "../../config/config.h"
#include "bupsplit.h"
#include <Python.h>
#include <assert.h>
-#include <stdint.h>
+#include <errno.h>
#include <fcntl.h>
#include <arpa/inet.h>
-#include <unistd.h>
+#include <stdint.h>
#include <stdlib.h>
#include <stdio.h>
-static int istty = 0;
+#ifdef HAVE_SYS_TYPES_H
+#include <sys/types.h>
+#endif
+#ifdef HAVE_SYS_STAT_H
+#include <sys/stat.h>
+#endif
+#ifdef HAVE_UNISTD_H
+#include <unistd.h>
+#endif
+
+#ifdef HAVE_LINUX_FS_H
+#include <linux/fs.h>
+#endif
+#ifdef HAVE_SYS_IOCTL_H
+#include <sys/ioctl.h>
+#endif
+
+#if defined(FS_IOC_GETFLAGS) && defined(FS_IOC_SETFLAGS)
+#define BUP_HAVE_FILE_ATTRS 1
+#endif
+
+#ifndef FS_NOCOW_FL
+// Of course, this assumes it's a bitfield value.
+#define FS_NOCOW_FL 0
+#endif
+
+static int istty2 = 0;
// Probably we should use autoconf or something and set HAVE_PY_GETARGCARGV...
#if __WIN32__ || __CYGWIN__
if (!PyArg_ParseTuple(args, "t#", &buf, &len))
return NULL;
out = bupsplit_find_ofs(buf, len, &bits);
+ if (out) assert(bits >= BUP_BLOBBITS);
return Py_BuildValue("ii", out, bits);
}
}
+#define BLOOM2_HEADERLEN 16
+
typedef struct {
uint32_t high;
unsigned char low;
} bits40_t;
-
static void to_bloom_address_bitmask4(const bits40_t *buf,
const int nbits, uint64_t *v, unsigned char *bitmask)
{
*bitmask = 1 << bit;
}
-
#define BLOOM_SET_BIT(name, address, itype, otype) \
static void name(unsigned char *bloom, const void *buf, const int nbits)\
{\
unsigned char bitmask;\
otype v;\
address((itype *)buf, nbits, &v, &bitmask);\
- bloom[16+v] |= bitmask;\
+ bloom[BLOOM2_HEADERLEN+v] |= bitmask;\
}
BLOOM_SET_BIT(bloom_set_bit4, to_bloom_address_bitmask4, bits40_t, uint64_t)
BLOOM_SET_BIT(bloom_set_bit5, to_bloom_address_bitmask5, uint32_t, uint32_t)
unsigned char bitmask;\
otype v;\
address((itype *)buf, nbits, &v, &bitmask);\
- return bloom[16+v] & bitmask;\
+ return bloom[BLOOM2_HEADERLEN+v] & bitmask;\
}
BLOOM_GET_BIT(bloom_get_bit4, to_bloom_address_bitmask4, bits40_t, uint64_t)
BLOOM_GET_BIT(bloom_get_bit5, to_bloom_address_bitmask5, uint32_t, uint32_t)
else
return NULL;
- return Py_BuildValue("Oi", Py_True, k);
+ return Py_BuildValue("ii", 1, k);
}
{
PyObject *ilist = NULL;
unsigned char *fmap = NULL;
- struct sha *sha_ptr, *last = NULL;
- uint32_t *table_ptr, *name_ptr;
+ struct sha *sha_ptr, *sha_start = NULL;
+ uint32_t *table_ptr, *name_ptr, *name_start;
struct idx **idxs = NULL;
int flen = 0, bits = 0, i;
uint32_t total, count, prefix;
idxs[i]->cur_name = NULL;
}
table_ptr = (uint32_t *)&fmap[MIDX4_HEADERLEN];
- sha_ptr = (struct sha *)&table_ptr[1<<bits];
- name_ptr = (uint32_t *)&sha_ptr[total];
+ sha_start = sha_ptr = (struct sha *)&table_ptr[1<<bits];
+ name_start = name_ptr = (uint32_t *)&sha_ptr[total];
last_i = num_i-1;
count = 0;
{
struct idx *idx;
uint32_t new_prefix;
- if (count % 102424 == 0 && istty)
+ if (count % 102424 == 0 && istty2)
fprintf(stderr, "midx: writing %.2f%% (%d/%d)\r",
count*100.0/total, count, total);
idx = idxs[last_i];
new_prefix = _extract_bits((unsigned char *)idx->cur, bits);
while (prefix < new_prefix)
table_ptr[prefix++] = htonl(count);
- if (last == NULL || _cmp_sha(last, idx->cur) != 0)
- {
- memcpy(sha_ptr++, idx->cur, sizeof(struct sha));
- *name_ptr++ = htonl(_get_idx_i(idx));
- last = idx->cur;
- }
+ memcpy(sha_ptr++, idx->cur, sizeof(struct sha));
+ *name_ptr++ = htonl(_get_idx_i(idx));
++idx->cur;
if (idx->cur_name != NULL)
++idx->cur_name;
_fix_idx_order(idxs, &last_i);
++count;
}
- table_ptr[prefix] = htonl(count);
+ while (prefix < (1<<bits))
+ table_ptr[prefix++] = htonl(count);
+ assert(count == total);
+ assert(prefix == (1<<bits));
+ assert(sha_ptr == sha_start+count);
+ assert(name_ptr == name_start+count);
PyMem_Free(idxs);
return PyLong_FromUnsignedLong(count);
}
-// This function should techniclly be macro'd out if it's going to be used
+// This function should technically be macro'd out if it's going to be used
// more than ocasionally. As of this writing, it'll actually never be called
// in real world bup scenarios (because our packs are < MAX_INT bytes).
static uint64_t htonll(uint64_t value)
}
-static PyObject *open_noatime(PyObject *self, PyObject *args)
+static int _open_noatime(const char *filename, int attrs)
{
- char *filename = NULL;
- int attrs, attrs_noatime, fd;
- if (!PyArg_ParseTuple(args, "s", &filename))
- return NULL;
- attrs = O_RDONLY;
+ int attrs_noatime, fd;
+ attrs |= O_RDONLY;
#ifdef O_NOFOLLOW
attrs |= O_NOFOLLOW;
#endif
// just harmlessly ignore it, so this branch won't trigger)
fd = open(filename, attrs);
}
+ return fd;
+}
+
+
+static PyObject *open_noatime(PyObject *self, PyObject *args)
+{
+ char *filename = NULL;
+ int fd;
+ if (!PyArg_ParseTuple(args, "s", &filename))
+ return NULL;
+ fd = _open_noatime(filename, 0);
if (fd < 0)
- return PyErr_SetFromErrnoWithFilename(PyExc_IOError, filename);
+ return PyErr_SetFromErrnoWithFilename(PyExc_OSError, filename);
return Py_BuildValue("i", fd);
}
}
-static PyMethodDef faster_methods[] = {
+#ifdef BUP_HAVE_FILE_ATTRS
+static PyObject *bup_get_linux_file_attr(PyObject *self, PyObject *args)
+{
+ int rc;
+ unsigned long attr;
+ char *path;
+ int fd;
+
+ if (!PyArg_ParseTuple(args, "s", &path))
+ return NULL;
+
+ fd = _open_noatime(path, O_NONBLOCK);
+ if (fd == -1)
+ return PyErr_SetFromErrnoWithFilename(PyExc_OSError, path);
+
+ attr = 0;
+ rc = ioctl(fd, FS_IOC_GETFLAGS, &attr);
+ if (rc == -1)
+ {
+ close(fd);
+ return PyErr_SetFromErrnoWithFilename(PyExc_OSError, path);
+ }
+
+ close(fd);
+ return Py_BuildValue("k", attr);
+}
+#endif /* def BUP_HAVE_FILE_ATTRS */
+
+
+#ifdef BUP_HAVE_FILE_ATTRS
+static PyObject *bup_set_linux_file_attr(PyObject *self, PyObject *args)
+{
+ int rc;
+ unsigned long orig_attr, attr;
+ char *path;
+ int fd;
+
+ if (!PyArg_ParseTuple(args, "sk", &path, &attr))
+ return NULL;
+
+ fd = open(path, O_RDONLY | O_NONBLOCK | O_LARGEFILE | O_NOFOLLOW);
+ if (fd == -1)
+ return PyErr_SetFromErrnoWithFilename(PyExc_OSError, path);
+
+ // Restrict attr to modifiable flags acdeijstuADST -- see
+ // chattr(1) and the e2fsprogs source. Letter to flag mapping is
+ // in pf.c flags_array[].
+ attr &= FS_APPEND_FL | FS_COMPR_FL | FS_NODUMP_FL | FS_EXTENT_FL
+ | FS_IMMUTABLE_FL | FS_JOURNAL_DATA_FL | FS_SECRM_FL | FS_NOTAIL_FL
+ | FS_UNRM_FL | FS_NOATIME_FL | FS_DIRSYNC_FL | FS_SYNC_FL
+ | FS_TOPDIR_FL | FS_NOCOW_FL;
+
+ // The extents flag can't be removed, so don't (see chattr(1) and chattr.c).
+ rc = ioctl(fd, FS_IOC_GETFLAGS, &orig_attr);
+ if (rc == -1)
+ {
+ close(fd);
+ return PyErr_SetFromErrnoWithFilename(PyExc_OSError, path);
+ }
+ attr |= (orig_attr & FS_EXTENT_FL);
+
+ rc = ioctl(fd, FS_IOC_SETFLAGS, &attr);
+ if (rc == -1)
+ {
+ close(fd);
+ return PyErr_SetFromErrnoWithFilename(PyExc_OSError, path);
+ }
+
+ close(fd);
+ return Py_BuildValue("O", Py_None);
+}
+#endif /* def BUP_HAVE_FILE_ATTRS */
+
+
+#if defined(HAVE_UTIMENSAT) || defined(HAVE_FUTIMES) || defined(HAVE_LUTIMES)
+
+static int bup_parse_xutime_args(char **path,
+ long *access,
+ long *access_ns,
+ long *modification,
+ long *modification_ns,
+ PyObject *self, PyObject *args)
+{
+ if (!PyArg_ParseTuple(args, "s((ll)(ll))",
+ path,
+ access, access_ns,
+ modification, modification_ns))
+ return 0;
+
+ if (isnan(*access))
+ {
+ PyErr_SetString(PyExc_ValueError, "access time is NaN");
+ return 0;
+ }
+ else if (isinf(*access))
+ {
+ PyErr_SetString(PyExc_ValueError, "access time is infinite");
+ return 0;
+ }
+ else if (isnan(*modification))
+ {
+ PyErr_SetString(PyExc_ValueError, "modification time is NaN");
+ return 0;
+ }
+ else if (isinf(*modification))
+ {
+ PyErr_SetString(PyExc_ValueError, "modification time is infinite");
+ return 0;
+ }
+
+ if (isnan(*access_ns))
+ {
+ PyErr_SetString(PyExc_ValueError, "access time ns is NaN");
+ return 0;
+ }
+ else if (isinf(*access_ns))
+ {
+ PyErr_SetString(PyExc_ValueError, "access time ns is infinite");
+ return 0;
+ }
+ else if (isnan(*modification_ns))
+ {
+ PyErr_SetString(PyExc_ValueError, "modification time ns is NaN");
+ return 0;
+ }
+ else if (isinf(*modification_ns))
+ {
+ PyErr_SetString(PyExc_ValueError, "modification time ns is infinite");
+ return 0;
+ }
+
+ return 1;
+}
+
+#endif /* defined(HAVE_UTIMENSAT) || defined(HAVE_FUTIMES)
+ || defined(HAVE_LUTIMES) */
+
+
+#ifdef HAVE_UTIMENSAT
+
+static PyObject *bup_xutime_ns(PyObject *self, PyObject *args,
+ int follow_symlinks)
+{
+ int rc;
+ char *path;
+ long access, access_ns, modification, modification_ns;
+ struct timespec ts[2];
+
+ if (!bup_parse_xutime_args(&path, &access, &access_ns,
+ &modification, &modification_ns,
+ self, args))
+ return NULL;
+
+ ts[0].tv_sec = access;
+ ts[0].tv_nsec = access_ns;
+ ts[1].tv_sec = modification;
+ ts[1].tv_nsec = modification_ns;
+ rc = utimensat(AT_FDCWD, path, ts,
+ follow_symlinks ? 0 : AT_SYMLINK_NOFOLLOW);
+ if (rc != 0)
+ return PyErr_SetFromErrnoWithFilename(PyExc_OSError, path);
+
+ return Py_BuildValue("O", Py_None);
+}
+
+
+#define BUP_HAVE_BUP_UTIME_NS 1
+static PyObject *bup_utime_ns(PyObject *self, PyObject *args)
+{
+ return bup_xutime_ns(self, args, 1);
+}
+
+
+#define BUP_HAVE_BUP_LUTIME_NS 1
+static PyObject *bup_lutime_ns(PyObject *self, PyObject *args)
+{
+ return bup_xutime_ns(self, args, 0);
+}
+
+
+#else /* not defined(HAVE_UTIMENSAT) */
+
+
+#ifdef HAVE_UTIMES
+#define BUP_HAVE_BUP_UTIME_NS 1
+static PyObject *bup_utime_ns(PyObject *self, PyObject *args)
+{
+ int rc;
+ char *path;
+ long access, access_ns, modification, modification_ns;
+ struct timeval tv[2];
+
+ if (!bup_parse_xutime_args(&path, &access, &access_ns,
+ &modification, &modification_ns,
+ self, args))
+ return NULL;
+
+ tv[0].tv_sec = access;
+ tv[0].tv_usec = access_ns / 1000;
+ tv[1].tv_sec = modification;
+ tv[1].tv_usec = modification_ns / 1000;
+ rc = utimes(path, tv);
+ if (rc != 0)
+ return PyErr_SetFromErrnoWithFilename(PyExc_OSError, path);
+
+ return Py_BuildValue("O", Py_None);
+}
+#endif /* def HAVE_UTIMES */
+
+
+#ifdef HAVE_LUTIMES
+#define BUP_HAVE_BUP_LUTIME_NS 1
+static PyObject *bup_lutime_ns(PyObject *self, PyObject *args)
+{
+ int rc;
+ char *path;
+ long access, access_ns, modification, modification_ns;
+ struct timeval tv[2];
+
+ if (!bup_parse_xutime_args(&path, &access, &access_ns,
+ &modification, &modification_ns,
+ self, args))
+ return NULL;
+
+ tv[0].tv_sec = access;
+ tv[0].tv_usec = access_ns / 1000;
+ tv[1].tv_sec = modification;
+ tv[1].tv_usec = modification_ns / 1000;
+ rc = lutimes(path, tv);
+ if (rc != 0)
+ return PyErr_SetFromErrnoWithFilename(PyExc_OSError, path);
+
+ return Py_BuildValue("O", Py_None);
+}
+#endif /* def HAVE_LUTIMES */
+
+
+#endif /* not defined(HAVE_UTIMENSAT) */
+
+
+#ifdef HAVE_STAT_ST_ATIM
+# define BUP_STAT_ATIME_NS(st) (st)->st_atim.tv_nsec
+# define BUP_STAT_MTIME_NS(st) (st)->st_mtim.tv_nsec
+# define BUP_STAT_CTIME_NS(st) (st)->st_ctim.tv_nsec
+#elif defined HAVE_STAT_ST_ATIMENSEC
+# define BUP_STAT_ATIME_NS(st) (st)->st_atimespec.tv_nsec
+# define BUP_STAT_MTIME_NS(st) (st)->st_mtimespec.tv_nsec
+# define BUP_STAT_CTIME_NS(st) (st)->st_ctimespec.tv_nsec
+#else
+# define BUP_STAT_ATIME_NS(st) 0
+# define BUP_STAT_MTIME_NS(st) 0
+# define BUP_STAT_CTIME_NS(st) 0
+#endif
+
+
+static PyObject *stat_struct_to_py(const struct stat *st)
+{
+ long atime_ns = BUP_STAT_ATIME_NS(st);
+ long mtime_ns = BUP_STAT_MTIME_NS(st);
+ long ctime_ns = BUP_STAT_CTIME_NS(st);
+
+ /* Enforce the current timespec nanosecond range expectations. */
+ if (atime_ns < 0 || atime_ns > 999999999)
+ {
+ PyErr_SetString(PyExc_ValueError, "invalid atime timespec nanoseconds");
+ return NULL;
+ }
+ if (mtime_ns < 0 || mtime_ns > 999999999)
+ {
+ PyErr_SetString(PyExc_ValueError, "invalid mtime timespec nanoseconds");
+ return NULL;
+ }
+ if (ctime_ns < 0 || ctime_ns > 999999999)
+ {
+ PyErr_SetString(PyExc_ValueError, "invalid ctime timespec nanoseconds");
+ return NULL;
+ }
+
+ return Py_BuildValue("kkkkkkkk(Ll)(Ll)(Ll)",
+ (unsigned long) st->st_mode,
+ (unsigned long) st->st_ino,
+ (unsigned long) st->st_dev,
+ (unsigned long) st->st_nlink,
+ (unsigned long) st->st_uid,
+ (unsigned long) st->st_gid,
+ (unsigned long) st->st_rdev,
+ (unsigned long) st->st_size,
+ (long long) st->st_atime,
+ (long) atime_ns,
+ (long long) st->st_mtime,
+ (long) mtime_ns,
+ (long long) st->st_ctime,
+ (long) ctime_ns);
+}
+
+
+static PyObject *bup_stat(PyObject *self, PyObject *args)
+{
+ int rc;
+ char *filename;
+
+ if (!PyArg_ParseTuple(args, "s", &filename))
+ return NULL;
+
+ struct stat st;
+ rc = stat(filename, &st);
+ if (rc != 0)
+ return PyErr_SetFromErrnoWithFilename(PyExc_OSError, filename);
+ return stat_struct_to_py(&st);
+}
+
+
+static PyObject *bup_lstat(PyObject *self, PyObject *args)
+{
+ int rc;
+ char *filename;
+
+ if (!PyArg_ParseTuple(args, "s", &filename))
+ return NULL;
+
+ struct stat st;
+ rc = lstat(filename, &st);
+ if (rc != 0)
+ return PyErr_SetFromErrnoWithFilename(PyExc_OSError, filename);
+ return stat_struct_to_py(&st);
+}
+
+
+static PyObject *bup_fstat(PyObject *self, PyObject *args)
+{
+ int rc, fd;
+
+ if (!PyArg_ParseTuple(args, "i", &fd))
+ return NULL;
+
+ struct stat st;
+ rc = fstat(fd, &st);
+ if (rc != 0)
+ return PyErr_SetFromErrno(PyExc_OSError);
+ return stat_struct_to_py(&st);
+}
+
+
+static PyMethodDef helper_methods[] = {
{ "selftest", selftest, METH_VARARGS,
"Check that the rolling checksum rolls correctly (for unit tests)." },
{ "blobbits", blobbits, METH_VARARGS,
"open() the given filename for read with O_NOATIME if possible" },
{ "fadvise_done", fadvise_done, METH_VARARGS,
"Inform the kernel that we're finished with earlier parts of a file" },
+#ifdef BUP_HAVE_FILE_ATTRS
+ { "get_linux_file_attr", bup_get_linux_file_attr, METH_VARARGS,
+ "Return the Linux attributes for the given file." },
+#endif
+#ifdef BUP_HAVE_FILE_ATTRS
+ { "set_linux_file_attr", bup_set_linux_file_attr, METH_VARARGS,
+ "Set the Linux attributes for the given file." },
+#endif
+#ifdef BUP_HAVE_BUP_UTIME_NS
+ { "bup_utime_ns", bup_utime_ns, METH_VARARGS,
+ "Change path timestamps with up to nanosecond precision." },
+#endif
+#ifdef BUP_HAVE_BUP_LUTIME_NS
+ { "bup_lutime_ns", bup_lutime_ns, METH_VARARGS,
+ "Change path timestamps with up to nanosecond precision;"
+ " don't follow symlinks." },
+#endif
+ { "stat", bup_stat, METH_VARARGS,
+ "Extended version of stat." },
+ { "lstat", bup_lstat, METH_VARARGS,
+ "Extended version of lstat." },
+ { "fstat", bup_fstat, METH_VARARGS,
+ "Extended version of fstat." },
{ NULL, NULL, 0, NULL }, // sentinel
};
+
PyMODINIT_FUNC init_helpers(void)
{
- Py_InitModule("_helpers", faster_methods);
- istty = isatty(2) || getenv("BUP_FORCE_TTY");
+ char *e;
+ PyObject *m = Py_InitModule("_helpers", helper_methods);
+ if (m == NULL)
+ return;
+ e = getenv("BUP_FORCE_TTY");
+ istty2 = isatty(2) || (atoi(e ? e : "0") & 2);
unpythonize_argv();
}
projects / bup.git / blobdiff