mirror of
https://sourceware.org/git/binutils-gdb.git
synced 2024-12-21 04:42:53 +08:00
49e4877c54
I tried building gdb with G++ 4.7 and CXX="g++ -std=gnu+11", and that tripped on a build error: In file included from build-gnulib/import/stdio.h:53:0, from /home/pedro/gdb/mygit/cxx-convertion/src/gdb/common/common-defs.h:31, from /home/pedro/gdb/mygit/cxx-convertion/src/gdb/defs.h:28, from /home/pedro/gdb/mygit/cxx-convertion/src/gdb/armobsd-tdep.c:20: build-gnulib/import/stddef.h:104:3: error: conflicting declaration ‘typedef union max_align_t max_align_t’ In file included from build-gnulib/import/stddef.h:55:0, from build-gnulib/import/stdio.h:53, from /home/pedro/gdb/mygit/cxx-convertion/src/gdb/common/common-defs.h:31, from /home/pedro/gdb/mygit/cxx-convertion/src/gdb/defs.h:28, from /home/pedro/gdb/mygit/cxx-convertion/src/gdb/armobsd-tdep.c:20: /opt/gcc-4.7/lib/gcc/x86_64-unknown-linux-gnu/4.7.4/include/stddef.h:426:3: error: ‘max_align_t’ has a previous declaration as ‘typedef struct max_align_t max_align_t’ Updating gnulib to current master fixes it, since it brings in this fix: http://lists.gnu.org/archive/html/bug-gnulib/2016-04/msg00000.html Our last update was in August 2015. This doesn't bring in much added baggage, it's mostly bug fixes. It pulled in the "limits-h" module as automatic dependency, and given that looks potentially useful I added it to the set of modules we require. Tested on x86_64 Fedora 23, with g++ 4.7 and g++ 5.3, native and gdbserver. gdb/ChangeLog: 2016-10-18 Pedro Alves <palves@redhat.com> * gnulib/update-gnulib.sh (GNULIB_COMMIT_SHA1): Set to 2692e23a48e21f6daa029e8af9f1a143b7532f47. * gnulib/configure, gnulib/config.in, gnulib/aclocal.m4: Regenerate. * gnulib/import/Makefile: Update. * gnulib/import/alloca: Update. * gnulib/import/basename-lgpl: Update. * gnulib/import/canonicalize-lgpl: Update. * gnulib/import/config: Update. * gnulib/import/dirent: Update. * gnulib/import/dirfd: Update. * gnulib/import/dirname-lgpl: Update. * gnulib/import/dirname.h: Update. * gnulib/import/dosname.h: Update. * gnulib/import/errno: Update. * gnulib/import/extra/snippet/arg-nonnull.h: Update. * gnulib/import/extra/snippet/c++defs.h: Update. * gnulib/import/extra/snippet/warn-on-use.h: Update. * gnulib/import/extra/update-copyright: Update. * gnulib/import/flexmember.h: Update. * gnulib/import/float+.h: Update. * gnulib/import/float: Update. * gnulib/import/float: Update. * gnulib/import/fnmatch: Update. * gnulib/import/fnmatch: Update. * gnulib/import/fnmatch_loop: Update. * gnulib/import/fpucw.h: Update. * gnulib/import/frexp: Update. * gnulib/import/frexpl: Update. * gnulib/import/gettimeofday: Update. * gnulib/import/hard-locale: Update. * gnulib/import/hard-locale.h: Update. * gnulib/import/inttypes: Update. * gnulib/import/isnan: Update. * gnulib/import/isnand-nolibm.h: Update. * gnulib/import/isnand: Update. * gnulib/import/isnanl-nolibm.h: Update. * gnulib/import/isnanl: Update. * gnulib/import/itold: Update. * gnulib/import/limits: Update. * gnulib/import/localcharset: Update. * gnulib/import/localcharset.h: Update. * gnulib/import/lstat: Update. * gnulib/import/m4/00gnulib: Update. * gnulib/import/m4/absolute-header: Update. * gnulib/import/m4/alloca: Update. * gnulib/import/m4/canonicalize: Update. * gnulib/import/m4/codeset: Update. * gnulib/import/m4/configmake: Update. * gnulib/import/m4/dirent_h: Update. * gnulib/import/m4/dirfd: Update. * gnulib/import/m4/dirname: Update. * gnulib/import/m4/double-slash-root: Update. * gnulib/import/m4/eealloc: Update. * gnulib/import/m4/errno_h: Update. * gnulib/import/m4/exponentd: Update. * gnulib/import/m4/exponentl: Update. * gnulib/import/m4/extensions: Update. * gnulib/import/m4/extern-inline: Update. * gnulib/import/m4/fcntl-o: Update. * gnulib/import/m4/flexmember: Update. * gnulib/import/m4/float_h: Update. * gnulib/import/m4/fnmatch: Update. * gnulib/import/m4/fpieee: Update. * gnulib/import/m4/frexp: Update. * gnulib/import/m4/frexpl: Update. * gnulib/import/m4/gettimeofday: Update. * gnulib/import/m4/glibc21: Update. * gnulib/import/m4/gnulib-cache: Update. * gnulib/import/m4/gnulib-common: Update. * gnulib/import/m4/gnulib-comp: Update. * gnulib/import/m4/gnulib-tool: Update. * gnulib/import/m4/hard-locale: Update. * gnulib/import/m4/include_next: Update. * gnulib/import/m4/inttypes-pri: Update. * gnulib/import/m4/inttypes: Update. * gnulib/import/m4/isnand: Update. * gnulib/import/m4/isnanl: Update. * gnulib/import/m4/largefile: Update. * gnulib/import/m4/limits-h: Update. * gnulib/import/m4/localcharset: Update. * gnulib/import/m4/locale-fr: Update. * gnulib/import/m4/locale-ja: Update. * gnulib/import/m4/locale-zh: Update. * gnulib/import/m4/longlong: Update. * gnulib/import/m4/lstat: Update. * gnulib/import/m4/malloc: Update. * gnulib/import/m4/malloca: Update. * gnulib/import/m4/math_h: Update. * gnulib/import/m4/mbrtowc: Update. * gnulib/import/m4/mbsinit: Update. * gnulib/import/m4/mbsrtowcs: Update. * gnulib/import/m4/mbstate_t: Update. * gnulib/import/m4/memchr: Update. * gnulib/import/m4/memmem: Update. * gnulib/import/m4/mmap-anon: Update. * gnulib/import/m4/multiarch: Update. * gnulib/import/m4/nocrash: Update. * gnulib/import/m4/off_t: Update. * gnulib/import/m4/pathmax: Update. * gnulib/import/m4/rawmemchr: Update. * gnulib/import/m4/readlink: Update. * gnulib/import/m4/rename: Update. * gnulib/import/m4/rmdir: Update. * gnulib/import/m4/signal_h: Update. * gnulib/import/m4/ssize_t: Update. * gnulib/import/m4/stat: Update. * gnulib/import/m4/stdbool: Update. * gnulib/import/m4/stddef_h: Update. * gnulib/import/m4/stdint: Update. * gnulib/import/m4/stdio_h: Update. * gnulib/import/m4/stdlib_h: Update. * gnulib/import/m4/strchrnul: Update. * gnulib/import/m4/string_h: Update. * gnulib/import/m4/strstr: Update. * gnulib/import/m4/strtok_r: Update. * gnulib/import/m4/sys_socket_h: Update. * gnulib/import/m4/sys_stat_h: Update. * gnulib/import/m4/sys_time_h: Update. * gnulib/import/m4/sys_types_h: Update. * gnulib/import/m4/time_h: Update. * gnulib/import/m4/unistd_h: Update. * gnulib/import/m4/warn-on-use: Update. * gnulib/import/m4/wchar_h: Update. * gnulib/import/m4/wchar_t: Update. * gnulib/import/m4/wctype_h: Update. * gnulib/import/m4/wint_t: Update. * gnulib/import/malloc: Update. * gnulib/import/malloca: Update. * gnulib/import/malloca.h: Update. * gnulib/import/math: Update. * gnulib/import/math: Update. * gnulib/import/mbrtowc: Update. * gnulib/import/mbsinit: Update. * gnulib/import/mbsrtowcs-impl.h: Update. * gnulib/import/mbsrtowcs-state: Update. * gnulib/import/mbsrtowcs: Update. * gnulib/import/memchr: Update. * gnulib/import/memmem: Update. * gnulib/import/pathmax.h: Update. * gnulib/import/rawmemchr: Update. * gnulib/import/readlink: Update. * gnulib/import/ref-add.sin: Update. * gnulib/import/ref-del.sin: Update. * gnulib/import/rename: Update. * gnulib/import/rmdir: Update. * gnulib/import/same-inode.h: Update. * gnulib/import/signal: Update. * gnulib/import/stat: Update. * gnulib/import/stdbool: Update. * gnulib/import/stddef: Update. * gnulib/import/stdint: Update. * gnulib/import/stdio: Update. * gnulib/import/stdlib: Update. * gnulib/import/str-two-way.h: Update. * gnulib/import/strchrnul: Update. * gnulib/import/streq.h: Update. * gnulib/import/string: Update. * gnulib/import/stripslash: Update. * gnulib/import/strnlen1: Update. * gnulib/import/strnlen1.h: Update. * gnulib/import/strstr: Update. * gnulib/import/strtok_r: Update. * gnulib/import/sys_stat: Update. * gnulib/import/sys_time: Update. * gnulib/import/sys_types: Update. * gnulib/import/time: Update. * gnulib/import/unistd: Update. * gnulib/import/unistd: Update. * gnulib/import/verify.h: Update. * gnulib/import/wchar: Update. * gnulib/import/wctype: Update. * gnulib/import/flexmember.h: New file. * gnulib/import/hard-locale.c: New file. * gnulib/import/hard-locale.h: New file. * gnulib/import/limits.in.h: New file. * gnulib/import/m4/flexmember.m4: New file. * gnulib/import/m4/hard-locale.m4: New file. * gnulib/import/m4/limits-h.m4: New file.
453 lines
17 KiB
C
453 lines
17 KiB
C
/* Byte-wise substring search, using the Two-Way algorithm.
|
|
Copyright (C) 2008-2016 Free Software Foundation, Inc.
|
|
This file is part of the GNU C Library.
|
|
Written by Eric Blake <ebb9@byu.net>, 2008.
|
|
|
|
This program is free software; you can redistribute it and/or modify
|
|
it under the terms of the GNU General Public License as published by
|
|
the Free Software Foundation; either version 3, or (at your option)
|
|
any later version.
|
|
|
|
This program 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 General Public License for more details.
|
|
|
|
You should have received a copy of the GNU General Public License along
|
|
with this program; if not, see <http://www.gnu.org/licenses/>. */
|
|
|
|
/* Before including this file, you need to include <config.h> and
|
|
<string.h>, and define:
|
|
RESULT_TYPE A macro that expands to the return type.
|
|
AVAILABLE(h, h_l, j, n_l)
|
|
A macro that returns nonzero if there are
|
|
at least N_L bytes left starting at H[J].
|
|
H is 'unsigned char *', H_L, J, and N_L
|
|
are 'size_t'; H_L is an lvalue. For
|
|
NUL-terminated searches, H_L can be
|
|
modified each iteration to avoid having
|
|
to compute the end of H up front.
|
|
|
|
For case-insensitivity, you may optionally define:
|
|
CMP_FUNC(p1, p2, l) A macro that returns 0 iff the first L
|
|
characters of P1 and P2 are equal.
|
|
CANON_ELEMENT(c) A macro that canonicalizes an element right after
|
|
it has been fetched from one of the two strings.
|
|
The argument is an 'unsigned char'; the result
|
|
must be an 'unsigned char' as well.
|
|
|
|
This file undefines the macros documented above, and defines
|
|
LONG_NEEDLE_THRESHOLD.
|
|
*/
|
|
|
|
#include <limits.h>
|
|
#include <stdint.h>
|
|
|
|
/* We use the Two-Way string matching algorithm (also known as
|
|
Chrochemore-Perrin), which guarantees linear complexity with
|
|
constant space. Additionally, for long needles, we also use a bad
|
|
character shift table similar to the Boyer-Moore algorithm to
|
|
achieve improved (potentially sub-linear) performance.
|
|
|
|
See http://www-igm.univ-mlv.fr/~lecroq/string/node26.html#SECTION00260,
|
|
http://en.wikipedia.org/wiki/Boyer-Moore_string_search_algorithm,
|
|
http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.34.6641&rep=rep1&type=pdf
|
|
*/
|
|
|
|
/* Point at which computing a bad-byte shift table is likely to be
|
|
worthwhile. Small needles should not compute a table, since it
|
|
adds (1 << CHAR_BIT) + NEEDLE_LEN computations of preparation for a
|
|
speedup no greater than a factor of NEEDLE_LEN. The larger the
|
|
needle, the better the potential performance gain. On the other
|
|
hand, on non-POSIX systems with CHAR_BIT larger than eight, the
|
|
memory required for the table is prohibitive. */
|
|
#if CHAR_BIT < 10
|
|
# define LONG_NEEDLE_THRESHOLD 32U
|
|
#else
|
|
# define LONG_NEEDLE_THRESHOLD SIZE_MAX
|
|
#endif
|
|
|
|
#ifndef MAX
|
|
# define MAX(a, b) ((a < b) ? (b) : (a))
|
|
#endif
|
|
|
|
#ifndef CANON_ELEMENT
|
|
# define CANON_ELEMENT(c) c
|
|
#endif
|
|
#ifndef CMP_FUNC
|
|
# define CMP_FUNC memcmp
|
|
#endif
|
|
|
|
/* Perform a critical factorization of NEEDLE, of length NEEDLE_LEN.
|
|
Return the index of the first byte in the right half, and set
|
|
*PERIOD to the global period of the right half.
|
|
|
|
The global period of a string is the smallest index (possibly its
|
|
length) at which all remaining bytes in the string are repetitions
|
|
of the prefix (the last repetition may be a subset of the prefix).
|
|
|
|
When NEEDLE is factored into two halves, a local period is the
|
|
length of the smallest word that shares a suffix with the left half
|
|
and shares a prefix with the right half. All factorizations of a
|
|
non-empty NEEDLE have a local period of at least 1 and no greater
|
|
than NEEDLE_LEN.
|
|
|
|
A critical factorization has the property that the local period
|
|
equals the global period. All strings have at least one critical
|
|
factorization with the left half smaller than the global period.
|
|
And while some strings have more than one critical factorization,
|
|
it is provable that with an ordered alphabet, at least one of the
|
|
critical factorizations corresponds to a maximal suffix.
|
|
|
|
Given an ordered alphabet, a critical factorization can be computed
|
|
in linear time, with 2 * NEEDLE_LEN comparisons, by computing the
|
|
shorter of two ordered maximal suffixes. The ordered maximal
|
|
suffixes are determined by lexicographic comparison while tracking
|
|
periodicity. */
|
|
static size_t
|
|
critical_factorization (const unsigned char *needle, size_t needle_len,
|
|
size_t *period)
|
|
{
|
|
/* Index of last byte of left half, or SIZE_MAX. */
|
|
size_t max_suffix, max_suffix_rev;
|
|
size_t j; /* Index into NEEDLE for current candidate suffix. */
|
|
size_t k; /* Offset into current period. */
|
|
size_t p; /* Intermediate period. */
|
|
unsigned char a, b; /* Current comparison bytes. */
|
|
|
|
/* Special case NEEDLE_LEN of 1 or 2 (all callers already filtered
|
|
out 0-length needles. */
|
|
if (needle_len < 3)
|
|
{
|
|
*period = 1;
|
|
return needle_len - 1;
|
|
}
|
|
|
|
/* Invariants:
|
|
0 <= j < NEEDLE_LEN - 1
|
|
-1 <= max_suffix{,_rev} < j (treating SIZE_MAX as if it were signed)
|
|
min(max_suffix, max_suffix_rev) < global period of NEEDLE
|
|
1 <= p <= global period of NEEDLE
|
|
p == global period of the substring NEEDLE[max_suffix{,_rev}+1...j]
|
|
1 <= k <= p
|
|
*/
|
|
|
|
/* Perform lexicographic search. */
|
|
max_suffix = SIZE_MAX;
|
|
j = 0;
|
|
k = p = 1;
|
|
while (j + k < needle_len)
|
|
{
|
|
a = CANON_ELEMENT (needle[j + k]);
|
|
b = CANON_ELEMENT (needle[max_suffix + k]);
|
|
if (a < b)
|
|
{
|
|
/* Suffix is smaller, period is entire prefix so far. */
|
|
j += k;
|
|
k = 1;
|
|
p = j - max_suffix;
|
|
}
|
|
else if (a == b)
|
|
{
|
|
/* Advance through repetition of the current period. */
|
|
if (k != p)
|
|
++k;
|
|
else
|
|
{
|
|
j += p;
|
|
k = 1;
|
|
}
|
|
}
|
|
else /* b < a */
|
|
{
|
|
/* Suffix is larger, start over from current location. */
|
|
max_suffix = j++;
|
|
k = p = 1;
|
|
}
|
|
}
|
|
*period = p;
|
|
|
|
/* Perform reverse lexicographic search. */
|
|
max_suffix_rev = SIZE_MAX;
|
|
j = 0;
|
|
k = p = 1;
|
|
while (j + k < needle_len)
|
|
{
|
|
a = CANON_ELEMENT (needle[j + k]);
|
|
b = CANON_ELEMENT (needle[max_suffix_rev + k]);
|
|
if (b < a)
|
|
{
|
|
/* Suffix is smaller, period is entire prefix so far. */
|
|
j += k;
|
|
k = 1;
|
|
p = j - max_suffix_rev;
|
|
}
|
|
else if (a == b)
|
|
{
|
|
/* Advance through repetition of the current period. */
|
|
if (k != p)
|
|
++k;
|
|
else
|
|
{
|
|
j += p;
|
|
k = 1;
|
|
}
|
|
}
|
|
else /* a < b */
|
|
{
|
|
/* Suffix is larger, start over from current location. */
|
|
max_suffix_rev = j++;
|
|
k = p = 1;
|
|
}
|
|
}
|
|
|
|
/* Choose the shorter suffix. Return the index of the first byte of
|
|
the right half, rather than the last byte of the left half.
|
|
|
|
For some examples, 'banana' has two critical factorizations, both
|
|
exposed by the two lexicographic extreme suffixes of 'anana' and
|
|
'nana', where both suffixes have a period of 2. On the other
|
|
hand, with 'aab' and 'bba', both strings have a single critical
|
|
factorization of the last byte, with the suffix having a period
|
|
of 1. While the maximal lexicographic suffix of 'aab' is 'b',
|
|
the maximal lexicographic suffix of 'bba' is 'ba', which is not a
|
|
critical factorization. Conversely, the maximal reverse
|
|
lexicographic suffix of 'a' works for 'bba', but not 'ab' for
|
|
'aab'. The shorter suffix of the two will always be a critical
|
|
factorization. */
|
|
if (max_suffix_rev + 1 < max_suffix + 1)
|
|
return max_suffix + 1;
|
|
*period = p;
|
|
return max_suffix_rev + 1;
|
|
}
|
|
|
|
/* Return the first location of non-empty NEEDLE within HAYSTACK, or
|
|
NULL. HAYSTACK_LEN is the minimum known length of HAYSTACK. This
|
|
method is optimized for NEEDLE_LEN < LONG_NEEDLE_THRESHOLD.
|
|
Performance is guaranteed to be linear, with an initialization cost
|
|
of 2 * NEEDLE_LEN comparisons.
|
|
|
|
If AVAILABLE does not modify HAYSTACK_LEN (as in memmem), then at
|
|
most 2 * HAYSTACK_LEN - NEEDLE_LEN comparisons occur in searching.
|
|
If AVAILABLE modifies HAYSTACK_LEN (as in strstr), then at most 3 *
|
|
HAYSTACK_LEN - NEEDLE_LEN comparisons occur in searching. */
|
|
static RETURN_TYPE
|
|
two_way_short_needle (const unsigned char *haystack, size_t haystack_len,
|
|
const unsigned char *needle, size_t needle_len)
|
|
{
|
|
size_t i; /* Index into current byte of NEEDLE. */
|
|
size_t j; /* Index into current window of HAYSTACK. */
|
|
size_t period; /* The period of the right half of needle. */
|
|
size_t suffix; /* The index of the right half of needle. */
|
|
|
|
/* Factor the needle into two halves, such that the left half is
|
|
smaller than the global period, and the right half is
|
|
periodic (with a period as large as NEEDLE_LEN - suffix). */
|
|
suffix = critical_factorization (needle, needle_len, &period);
|
|
|
|
/* Perform the search. Each iteration compares the right half
|
|
first. */
|
|
if (CMP_FUNC (needle, needle + period, suffix) == 0)
|
|
{
|
|
/* Entire needle is periodic; a mismatch in the left half can
|
|
only advance by the period, so use memory to avoid rescanning
|
|
known occurrences of the period in the right half. */
|
|
size_t memory = 0;
|
|
j = 0;
|
|
while (AVAILABLE (haystack, haystack_len, j, needle_len))
|
|
{
|
|
/* Scan for matches in right half. */
|
|
i = MAX (suffix, memory);
|
|
while (i < needle_len && (CANON_ELEMENT (needle[i])
|
|
== CANON_ELEMENT (haystack[i + j])))
|
|
++i;
|
|
if (needle_len <= i)
|
|
{
|
|
/* Scan for matches in left half. */
|
|
i = suffix - 1;
|
|
while (memory < i + 1 && (CANON_ELEMENT (needle[i])
|
|
== CANON_ELEMENT (haystack[i + j])))
|
|
--i;
|
|
if (i + 1 < memory + 1)
|
|
return (RETURN_TYPE) (haystack + j);
|
|
/* No match, so remember how many repetitions of period
|
|
on the right half were scanned. */
|
|
j += period;
|
|
memory = needle_len - period;
|
|
}
|
|
else
|
|
{
|
|
j += i - suffix + 1;
|
|
memory = 0;
|
|
}
|
|
}
|
|
}
|
|
else
|
|
{
|
|
/* The two halves of needle are distinct; no extra memory is
|
|
required, and any mismatch results in a maximal shift. */
|
|
period = MAX (suffix, needle_len - suffix) + 1;
|
|
j = 0;
|
|
while (AVAILABLE (haystack, haystack_len, j, needle_len))
|
|
{
|
|
/* Scan for matches in right half. */
|
|
i = suffix;
|
|
while (i < needle_len && (CANON_ELEMENT (needle[i])
|
|
== CANON_ELEMENT (haystack[i + j])))
|
|
++i;
|
|
if (needle_len <= i)
|
|
{
|
|
/* Scan for matches in left half. */
|
|
i = suffix - 1;
|
|
while (i != SIZE_MAX && (CANON_ELEMENT (needle[i])
|
|
== CANON_ELEMENT (haystack[i + j])))
|
|
--i;
|
|
if (i == SIZE_MAX)
|
|
return (RETURN_TYPE) (haystack + j);
|
|
j += period;
|
|
}
|
|
else
|
|
j += i - suffix + 1;
|
|
}
|
|
}
|
|
return NULL;
|
|
}
|
|
|
|
/* Return the first location of non-empty NEEDLE within HAYSTACK, or
|
|
NULL. HAYSTACK_LEN is the minimum known length of HAYSTACK. This
|
|
method is optimized for LONG_NEEDLE_THRESHOLD <= NEEDLE_LEN.
|
|
Performance is guaranteed to be linear, with an initialization cost
|
|
of 3 * NEEDLE_LEN + (1 << CHAR_BIT) operations.
|
|
|
|
If AVAILABLE does not modify HAYSTACK_LEN (as in memmem), then at
|
|
most 2 * HAYSTACK_LEN - NEEDLE_LEN comparisons occur in searching,
|
|
and sublinear performance O(HAYSTACK_LEN / NEEDLE_LEN) is possible.
|
|
If AVAILABLE modifies HAYSTACK_LEN (as in strstr), then at most 3 *
|
|
HAYSTACK_LEN - NEEDLE_LEN comparisons occur in searching, and
|
|
sublinear performance is not possible. */
|
|
static RETURN_TYPE
|
|
two_way_long_needle (const unsigned char *haystack, size_t haystack_len,
|
|
const unsigned char *needle, size_t needle_len)
|
|
{
|
|
size_t i; /* Index into current byte of NEEDLE. */
|
|
size_t j; /* Index into current window of HAYSTACK. */
|
|
size_t period; /* The period of the right half of needle. */
|
|
size_t suffix; /* The index of the right half of needle. */
|
|
size_t shift_table[1U << CHAR_BIT]; /* See below. */
|
|
|
|
/* Factor the needle into two halves, such that the left half is
|
|
smaller than the global period, and the right half is
|
|
periodic (with a period as large as NEEDLE_LEN - suffix). */
|
|
suffix = critical_factorization (needle, needle_len, &period);
|
|
|
|
/* Populate shift_table. For each possible byte value c,
|
|
shift_table[c] is the distance from the last occurrence of c to
|
|
the end of NEEDLE, or NEEDLE_LEN if c is absent from the NEEDLE.
|
|
shift_table[NEEDLE[NEEDLE_LEN - 1]] contains the only 0. */
|
|
for (i = 0; i < 1U << CHAR_BIT; i++)
|
|
shift_table[i] = needle_len;
|
|
for (i = 0; i < needle_len; i++)
|
|
shift_table[CANON_ELEMENT (needle[i])] = needle_len - i - 1;
|
|
|
|
/* Perform the search. Each iteration compares the right half
|
|
first. */
|
|
if (CMP_FUNC (needle, needle + period, suffix) == 0)
|
|
{
|
|
/* Entire needle is periodic; a mismatch in the left half can
|
|
only advance by the period, so use memory to avoid rescanning
|
|
known occurrences of the period in the right half. */
|
|
size_t memory = 0;
|
|
size_t shift;
|
|
j = 0;
|
|
while (AVAILABLE (haystack, haystack_len, j, needle_len))
|
|
{
|
|
/* Check the last byte first; if it does not match, then
|
|
shift to the next possible match location. */
|
|
shift = shift_table[CANON_ELEMENT (haystack[j + needle_len - 1])];
|
|
if (0 < shift)
|
|
{
|
|
if (memory && shift < period)
|
|
{
|
|
/* Since needle is periodic, but the last period has
|
|
a byte out of place, there can be no match until
|
|
after the mismatch. */
|
|
shift = needle_len - period;
|
|
}
|
|
memory = 0;
|
|
j += shift;
|
|
continue;
|
|
}
|
|
/* Scan for matches in right half. The last byte has
|
|
already been matched, by virtue of the shift table. */
|
|
i = MAX (suffix, memory);
|
|
while (i < needle_len - 1 && (CANON_ELEMENT (needle[i])
|
|
== CANON_ELEMENT (haystack[i + j])))
|
|
++i;
|
|
if (needle_len - 1 <= i)
|
|
{
|
|
/* Scan for matches in left half. */
|
|
i = suffix - 1;
|
|
while (memory < i + 1 && (CANON_ELEMENT (needle[i])
|
|
== CANON_ELEMENT (haystack[i + j])))
|
|
--i;
|
|
if (i + 1 < memory + 1)
|
|
return (RETURN_TYPE) (haystack + j);
|
|
/* No match, so remember how many repetitions of period
|
|
on the right half were scanned. */
|
|
j += period;
|
|
memory = needle_len - period;
|
|
}
|
|
else
|
|
{
|
|
j += i - suffix + 1;
|
|
memory = 0;
|
|
}
|
|
}
|
|
}
|
|
else
|
|
{
|
|
/* The two halves of needle are distinct; no extra memory is
|
|
required, and any mismatch results in a maximal shift. */
|
|
size_t shift;
|
|
period = MAX (suffix, needle_len - suffix) + 1;
|
|
j = 0;
|
|
while (AVAILABLE (haystack, haystack_len, j, needle_len))
|
|
{
|
|
/* Check the last byte first; if it does not match, then
|
|
shift to the next possible match location. */
|
|
shift = shift_table[CANON_ELEMENT (haystack[j + needle_len - 1])];
|
|
if (0 < shift)
|
|
{
|
|
j += shift;
|
|
continue;
|
|
}
|
|
/* Scan for matches in right half. The last byte has
|
|
already been matched, by virtue of the shift table. */
|
|
i = suffix;
|
|
while (i < needle_len - 1 && (CANON_ELEMENT (needle[i])
|
|
== CANON_ELEMENT (haystack[i + j])))
|
|
++i;
|
|
if (needle_len - 1 <= i)
|
|
{
|
|
/* Scan for matches in left half. */
|
|
i = suffix - 1;
|
|
while (i != SIZE_MAX && (CANON_ELEMENT (needle[i])
|
|
== CANON_ELEMENT (haystack[i + j])))
|
|
--i;
|
|
if (i == SIZE_MAX)
|
|
return (RETURN_TYPE) (haystack + j);
|
|
j += period;
|
|
}
|
|
else
|
|
j += i - suffix + 1;
|
|
}
|
|
}
|
|
return NULL;
|
|
}
|
|
|
|
#undef AVAILABLE
|
|
#undef CANON_ELEMENT
|
|
#undef CMP_FUNC
|
|
#undef MAX
|
|
#undef RETURN_TYPE
|