mirror of
https://sourceware.org/git/glibc.git
synced 2024-11-21 12:30:06 +00:00
* string/Makefile (distribute): Add str-two-way.h.
2008-03-29 Eric Blake <ebb9@byu.net> Rewrite string searches to O(n) rather than O(n^2). * string/str-two-way.h: New file. For linear fixed-allocation string searching. * string/memmem.c: New implementation. * string/strstr.c: New implementation. * string/strcasestr.c: New implementation. * sysdeps/posix/getaddrinfo.c (getaddrinfo): Call _res_hconf_init
This commit is contained in:
parent
b194db7985
commit
0caca71ac9
15
ChangeLog
15
ChangeLog
@ -1,3 +1,16 @@
|
||||
2008-05-14 Ulrich Drepper <drepper@redhat.com>
|
||||
|
||||
* string/Makefile (distribute): Add str-two-way.h.
|
||||
|
||||
2008-03-29 Eric Blake <ebb9@byu.net>
|
||||
|
||||
Rewrite string searches to O(n) rather than O(n^2).
|
||||
* string/str-two-way.h: New file. For linear fixed-allocation
|
||||
string searching.
|
||||
* string/memmem.c: New implementation.
|
||||
* string/strstr.c: New implementation.
|
||||
* string/strcasestr.c: New implementation.
|
||||
|
||||
2008-04-11 Paolo Bonzini <bonzini@gnu.org>
|
||||
|
||||
* posix/regcomp.c (optimize_utf8): Add a note on why we test
|
||||
@ -47,7 +60,7 @@
|
||||
(match_prefix): Don't treat IPv4 loopback address special when
|
||||
converting to v4 mapped addressed.
|
||||
|
||||
* sysdeps/posix/getaddrinfo.c (getaddrinfo): Add _res_hconf_init
|
||||
* sysdeps/posix/getaddrinfo.c (getaddrinfo): Call _res_hconf_init
|
||||
if necessary.
|
||||
* posix/tst-rfc3484.c: Add dummy definition of _res_hconf_init.
|
||||
* posix/tst-rfc3484-2.c: Likewise.
|
||||
|
7
NEWS
7
NEWS
@ -15,6 +15,13 @@ Version 2.9
|
||||
|
||||
* getaddrinfo now handles DCCP and UDPlite.
|
||||
Implemented by Ulrich Drepper.
|
||||
|
||||
* New fixed-size conversion macros: htobe16, htole16, be16toh, le16toh,
|
||||
htobe32, htole32, be32toh, le32toh, htobe64, htole64, be64toh, le64toh.
|
||||
Implemented by Ulrich Drepper.
|
||||
|
||||
* New implementation of memmem, strstr, and strcasestr which is O(n).
|
||||
Implemented by Eric Blake.
|
||||
|
||||
Version 2.8
|
||||
|
||||
|
@ -1038,7 +1038,9 @@ optimize_utf8 (re_dfa_t *dfa)
|
||||
case BUF_LAST:
|
||||
break;
|
||||
default:
|
||||
/* Word anchors etc. cannot be handled. */
|
||||
/* Word anchors etc. cannot be handled. It's okay to test
|
||||
opr.ctx_type since constraints (for all DFA nodes) are
|
||||
created by ORing one or more opr.ctx_type values. */
|
||||
return;
|
||||
}
|
||||
break;
|
||||
@ -1318,6 +1320,8 @@ calc_first (void *extra, bin_tree_t *node)
|
||||
node->node_idx = re_dfa_add_node (dfa, node->token);
|
||||
if (BE (node->node_idx == -1, 0))
|
||||
return REG_ESPACE;
|
||||
if (node->token.type == ANCHOR)
|
||||
dfa->nodes[node->node_idx].constraint = node->token.opr.ctx_type;
|
||||
}
|
||||
return REG_NOERROR;
|
||||
}
|
||||
@ -1446,22 +1450,17 @@ duplicate_node_closure (re_dfa_t *dfa, int top_org_node, int top_clone_node,
|
||||
destination. */
|
||||
org_dest = dfa->edests[org_node].elems[0];
|
||||
re_node_set_empty (dfa->edests + clone_node);
|
||||
if (dfa->nodes[org_node].type == ANCHOR)
|
||||
/* If the node is root_node itself, it means the epsilon clsoure
|
||||
has a loop. Then tie it to the destination of the root_node. */
|
||||
if (org_node == root_node && clone_node != org_node)
|
||||
{
|
||||
/* In case of the node has another constraint, append it. */
|
||||
if (org_node == root_node && clone_node != org_node)
|
||||
{
|
||||
/* ...but if the node is root_node itself, it means the
|
||||
epsilon closure have a loop, then tie it to the
|
||||
destination of the root_node. */
|
||||
ret = re_node_set_insert (dfa->edests + clone_node,
|
||||
org_dest);
|
||||
if (BE (ret < 0, 0))
|
||||
return REG_ESPACE;
|
||||
break;
|
||||
}
|
||||
constraint |= dfa->nodes[org_node].opr.ctx_type;
|
||||
ret = re_node_set_insert (dfa->edests + clone_node, org_dest);
|
||||
if (BE (ret < 0, 0))
|
||||
return REG_ESPACE;
|
||||
break;
|
||||
}
|
||||
/* In case of the node has another constraint, add it. */
|
||||
constraint |= dfa->nodes[org_node].constraint;
|
||||
clone_dest = duplicate_node (dfa, org_dest, constraint);
|
||||
if (BE (clone_dest == -1, 0))
|
||||
return REG_ESPACE;
|
||||
@ -1479,7 +1478,7 @@ duplicate_node_closure (re_dfa_t *dfa, int top_org_node, int top_clone_node,
|
||||
clone_dest = search_duplicated_node (dfa, org_dest, constraint);
|
||||
if (clone_dest == -1)
|
||||
{
|
||||
/* There are no such a duplicated node, create a new one. */
|
||||
/* There is no such duplicated node, create a new one. */
|
||||
reg_errcode_t err;
|
||||
clone_dest = duplicate_node (dfa, org_dest, constraint);
|
||||
if (BE (clone_dest == -1, 0))
|
||||
@ -1494,7 +1493,7 @@ duplicate_node_closure (re_dfa_t *dfa, int top_org_node, int top_clone_node,
|
||||
}
|
||||
else
|
||||
{
|
||||
/* There are a duplicated node which satisfy the constraint,
|
||||
/* There is a duplicated node which satisfies the constraint,
|
||||
use it to avoid infinite loop. */
|
||||
ret = re_node_set_insert (dfa->edests + clone_node, clone_dest);
|
||||
if (BE (ret < 0, 0))
|
||||
@ -1543,8 +1542,7 @@ duplicate_node (re_dfa_t *dfa, int org_idx, unsigned int constraint)
|
||||
if (BE (dup_idx != -1, 1))
|
||||
{
|
||||
dfa->nodes[dup_idx].constraint = constraint;
|
||||
if (dfa->nodes[org_idx].type == ANCHOR)
|
||||
dfa->nodes[dup_idx].constraint |= dfa->nodes[org_idx].opr.ctx_type;
|
||||
dfa->nodes[dup_idx].constraint |= dfa->nodes[org_idx].constraint;
|
||||
dfa->nodes[dup_idx].duplicated = 1;
|
||||
|
||||
/* Store the index of the original node. */
|
||||
@ -1624,7 +1622,6 @@ static reg_errcode_t
|
||||
calc_eclosure_iter (re_node_set *new_set, re_dfa_t *dfa, int node, int root)
|
||||
{
|
||||
reg_errcode_t err;
|
||||
unsigned int constraint;
|
||||
int i, incomplete;
|
||||
re_node_set eclosure;
|
||||
incomplete = 0;
|
||||
@ -1636,15 +1633,14 @@ calc_eclosure_iter (re_node_set *new_set, re_dfa_t *dfa, int node, int root)
|
||||
We reference this value to avoid infinite loop. */
|
||||
dfa->eclosures[node].nelem = -1;
|
||||
|
||||
constraint = ((dfa->nodes[node].type == ANCHOR)
|
||||
? dfa->nodes[node].opr.ctx_type : 0);
|
||||
/* If the current node has constraints, duplicate all nodes.
|
||||
Since they must inherit the constraints. */
|
||||
if (constraint
|
||||
/* If the current node has constraints, duplicate all nodes
|
||||
since they must inherit the constraints. */
|
||||
if (dfa->nodes[node].constraint
|
||||
&& dfa->edests[node].nelem
|
||||
&& !dfa->nodes[dfa->edests[node].elems[0]].duplicated)
|
||||
{
|
||||
err = duplicate_node_closure (dfa, node, node, node, constraint);
|
||||
err = duplicate_node_closure (dfa, node, node, node,
|
||||
dfa->nodes[node].constraint);
|
||||
if (BE (err != REG_NOERROR, 0))
|
||||
return err;
|
||||
}
|
||||
|
@ -1665,11 +1665,9 @@ create_cd_newstate (const re_dfa_t *dfa, const re_node_set *nodes,
|
||||
|
||||
for (i = 0 ; i < nodes->nelem ; i++)
|
||||
{
|
||||
unsigned int constraint = 0;
|
||||
re_token_t *node = dfa->nodes + nodes->elems[i];
|
||||
re_token_type_t type = node->type;
|
||||
if (node->constraint)
|
||||
constraint = node->constraint;
|
||||
unsigned int constraint = node->constraint;
|
||||
|
||||
if (type == CHARACTER && !constraint)
|
||||
continue;
|
||||
@ -1682,8 +1680,6 @@ create_cd_newstate (const re_dfa_t *dfa, const re_node_set *nodes,
|
||||
newstate->halt = 1;
|
||||
else if (type == OP_BACK_REF)
|
||||
newstate->has_backref = 1;
|
||||
else if (type == ANCHOR)
|
||||
constraint = node->opr.ctx_type;
|
||||
|
||||
if (constraint)
|
||||
{
|
||||
|
@ -55,7 +55,8 @@ tests := tester inl-tester noinl-tester testcopy test-ffs \
|
||||
tst-strtok tst-strxfrm bug-strcoll1 tst-strfry \
|
||||
bug-strtok1 $(addprefix test-,$(strop-tests)) \
|
||||
bug-envz1 tst-strxfrm2 tst-endian
|
||||
distribute := memcopy.h pagecopy.h tst-svc.expect test-string.h
|
||||
distribute := memcopy.h pagecopy.h tst-svc.expect test-string.h \
|
||||
str-two-way.h
|
||||
|
||||
|
||||
include ../Rules
|
||||
|
@ -1,4 +1,4 @@
|
||||
/* Copyright (C) 1991,92,93,94,96,97,98,2000,2004 Free Software Foundation, Inc.
|
||||
/* Copyright (C) 1991,92,93,94,96,97,98,2000,2004,2008 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
|
||||
@ -16,26 +16,36 @@
|
||||
Software Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA
|
||||
02111-1307 USA. */
|
||||
|
||||
#include <stddef.h>
|
||||
/* This particular implementation was written by Eric Blake, 2008. */
|
||||
|
||||
#ifndef _LIBC
|
||||
# include <config.h>
|
||||
#endif
|
||||
|
||||
/* Specification of memmem. */
|
||||
#include <string.h>
|
||||
|
||||
#ifndef _LIBC
|
||||
# define __builtin_expect(expr, val) (expr)
|
||||
#endif
|
||||
|
||||
#define RETURN_TYPE void *
|
||||
#define AVAILABLE(h, h_l, j, n_l) ((j) <= (h_l) - (n_l))
|
||||
#include "str-two-way.h"
|
||||
|
||||
#undef memmem
|
||||
|
||||
/* Return the first occurrence of NEEDLE in HAYSTACK. */
|
||||
/* Return the first occurrence of NEEDLE in HAYSTACK. Return HAYSTACK
|
||||
if NEEDLE_LEN is 0, otherwise NULL if NEEDLE is not found in
|
||||
HAYSTACK. */
|
||||
void *
|
||||
memmem (haystack, haystack_len, needle, needle_len)
|
||||
const void *haystack;
|
||||
size_t haystack_len;
|
||||
const void *needle;
|
||||
size_t needle_len;
|
||||
memmem (const void *haystack_start, size_t haystack_len,
|
||||
const void *needle_start, size_t needle_len)
|
||||
{
|
||||
const char *begin;
|
||||
const char *const last_possible
|
||||
= (const char *) haystack + haystack_len - needle_len;
|
||||
/* Abstract memory is considered to be an array of 'unsigned char' values,
|
||||
not an array of 'char' values. See ISO C 99 section 6.2.6.1. */
|
||||
const unsigned char *haystack = (const unsigned char *) haystack_start;
|
||||
const unsigned char *needle = (const unsigned char *) needle_start;
|
||||
|
||||
if (needle_len == 0)
|
||||
/* The first occurrence of the empty string is deemed to occur at
|
||||
@ -47,12 +57,22 @@ memmem (haystack, haystack_len, needle, needle_len)
|
||||
if (__builtin_expect (haystack_len < needle_len, 0))
|
||||
return NULL;
|
||||
|
||||
for (begin = (const char *) haystack; begin <= last_possible; ++begin)
|
||||
if (begin[0] == ((const char *) needle)[0] &&
|
||||
!memcmp ((const void *) &begin[1],
|
||||
(const void *) ((const char *) needle + 1),
|
||||
needle_len - 1))
|
||||
return (void *) begin;
|
||||
|
||||
return NULL;
|
||||
/* Use optimizations in memchr when possible, to reduce the search
|
||||
size of haystack using a linear algorithm with a smaller
|
||||
coefficient. However, avoid memchr for long needles, since we
|
||||
can often achieve sublinear performance. */
|
||||
if (needle_len < LONG_NEEDLE_THRESHOLD)
|
||||
{
|
||||
haystack = memchr (haystack, *needle, haystack_len);
|
||||
if (!haystack || __builtin_expect (needle_len == 1, 0))
|
||||
return (void *) haystack;
|
||||
haystack_len -= haystack - (const unsigned char *) haystack_start;
|
||||
if (haystack_len < needle_len)
|
||||
return NULL;
|
||||
return two_way_short_needle (haystack, haystack_len, needle, needle_len);
|
||||
}
|
||||
else
|
||||
return two_way_long_needle (haystack, haystack_len, needle, needle_len);
|
||||
}
|
||||
|
||||
#undef LONG_NEEDLE_THRESHOLD
|
||||
|
430
string/str-two-way.h
Normal file
430
string/str-two-way.h
Normal file
@ -0,0 +1,430 @@
|
||||
/* Byte-wise substring search, using the Two-Way algorithm.
|
||||
Copyright (C) 2008 Free Software Foundation, Inc.
|
||||
This file is part of the GNU C Library.
|
||||
Written by Eric Blake <ebb9@byu.net>, 2008.
|
||||
|
||||
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, write to the Free
|
||||
Software Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA
|
||||
02111-1307 USA. */
|
||||
|
||||
/* Before including this file, you need to include <string.h> (and
|
||||
<config.h> before that, if not part of libc), 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, 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
|
||||
and http://en.wikipedia.org/wiki/Boyer-Moore_string_search_algorithm
|
||||
*/
|
||||
|
||||
/* 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.
|
||||
|
||||
Given an ordered alphabet, a critical factorization can be computed
|
||||
in linear time, with 2 * NEEDLE_LEN comparisons, by computing the
|
||||
larger of two ordered maximal suffixes. The ordered maximal
|
||||
suffixes are determined by lexicographic comparison of
|
||||
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. */
|
||||
|
||||
/* 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 longer suffix. Return the first byte of the right
|
||||
half, rather than the last byte of the left half. */
|
||||
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 can only advance by the
|
||||
period, so use memory to avoid rescanning known occurrences
|
||||
of the period. */
|
||||
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 can only advance by the
|
||||
period, so use memory to avoid rescanning known occurrences
|
||||
of the period. */
|
||||
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
|
@ -1,5 +1,5 @@
|
||||
/* Return the offset of one string within another.
|
||||
Copyright (C) 1994, 1996-2000, 2004 Free Software Foundation, Inc.
|
||||
Copyright (C) 1994, 1996-2000, 2004, 2008 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
|
||||
@ -30,113 +30,71 @@
|
||||
# include <config.h>
|
||||
#endif
|
||||
|
||||
/* Specification. */
|
||||
#include <string.h>
|
||||
|
||||
#include <ctype.h>
|
||||
#include <stdbool.h>
|
||||
#include <strings.h>
|
||||
|
||||
#if defined _LIBC || defined HAVE_STRING_H
|
||||
# include <string.h>
|
||||
#endif
|
||||
#define TOLOWER(Ch) (isupper (Ch) ? tolower (Ch) : (Ch))
|
||||
|
||||
#ifdef _LIBC
|
||||
# include <locale/localeinfo.h>
|
||||
# define TOLOWER(c) __tolower_l ((unsigned char) c, loc)
|
||||
#else
|
||||
# define TOLOWER(c) _tolower (c)
|
||||
#endif
|
||||
|
||||
typedef unsigned chartype;
|
||||
/* Two-Way algorithm. */
|
||||
#define RETURN_TYPE char *
|
||||
#define AVAILABLE(h, h_l, j, n_l) \
|
||||
(!memchr ((h) + (h_l), '\0', (j) + (n_l) - (h_l)) \
|
||||
&& ((h_l) = (j) + (n_l)))
|
||||
#define CANON_ELEMENT(c) TOLOWER (c)
|
||||
#define CMP_FUNC(p1, p2, l) \
|
||||
strncasecmp ((const char *) (p1), (const char *) (p2), l)
|
||||
#include "str-two-way.h"
|
||||
|
||||
#undef strcasestr
|
||||
#undef __strcasestr
|
||||
|
||||
/* Find the first occurrence of NEEDLE in HAYSTACK, using
|
||||
case-insensitive comparison. This function gives unspecified
|
||||
results in multibyte locales. */
|
||||
char *
|
||||
__strcasestr (phaystack, pneedle)
|
||||
const char *phaystack;
|
||||
const char *pneedle;
|
||||
__strcasestr (const char *haystack_start, const char *needle_start)
|
||||
{
|
||||
register const unsigned char *haystack, *needle;
|
||||
register chartype b, c;
|
||||
#ifdef _LIBC
|
||||
__locale_t loc = _NL_CURRENT_LOCALE;
|
||||
#endif
|
||||
const char *haystack = haystack_start;
|
||||
const char *needle = needle_start;
|
||||
size_t needle_len; /* Length of NEEDLE. */
|
||||
size_t haystack_len; /* Known minimum length of HAYSTACK. */
|
||||
bool ok = true; /* True if NEEDLE is prefix of HAYSTACK. */
|
||||
|
||||
haystack = (const unsigned char *) phaystack;
|
||||
needle = (const unsigned char *) pneedle;
|
||||
|
||||
b = TOLOWER (*needle);
|
||||
if (b != '\0')
|
||||
/* Determine length of NEEDLE, and in the process, make sure
|
||||
HAYSTACK is at least as long (no point processing all of a long
|
||||
NEEDLE if HAYSTACK is too short). */
|
||||
while (*haystack && *needle)
|
||||
{
|
||||
haystack--; /* possible ANSI violation */
|
||||
do
|
||||
{
|
||||
c = *++haystack;
|
||||
if (c == '\0')
|
||||
goto ret0;
|
||||
}
|
||||
while (TOLOWER (c) != (int) b);
|
||||
|
||||
c = TOLOWER (*++needle);
|
||||
if (c == '\0')
|
||||
goto foundneedle;
|
||||
++needle;
|
||||
goto jin;
|
||||
|
||||
for (;;)
|
||||
{
|
||||
register chartype a;
|
||||
register const unsigned char *rhaystack, *rneedle;
|
||||
|
||||
do
|
||||
{
|
||||
a = *++haystack;
|
||||
if (a == '\0')
|
||||
goto ret0;
|
||||
if (TOLOWER (a) == (int) b)
|
||||
break;
|
||||
a = *++haystack;
|
||||
if (a == '\0')
|
||||
goto ret0;
|
||||
shloop:
|
||||
;
|
||||
}
|
||||
while (TOLOWER (a) != (int) b);
|
||||
|
||||
jin: a = *++haystack;
|
||||
if (a == '\0')
|
||||
goto ret0;
|
||||
|
||||
if (TOLOWER (a) != (int) c)
|
||||
goto shloop;
|
||||
|
||||
rhaystack = haystack-- + 1;
|
||||
rneedle = needle;
|
||||
a = TOLOWER (*rneedle);
|
||||
|
||||
if (TOLOWER (*rhaystack) == (int) a)
|
||||
do
|
||||
{
|
||||
if (a == '\0')
|
||||
goto foundneedle;
|
||||
++rhaystack;
|
||||
a = TOLOWER (*++needle);
|
||||
if (TOLOWER (*rhaystack) != (int) a)
|
||||
break;
|
||||
if (a == '\0')
|
||||
goto foundneedle;
|
||||
++rhaystack;
|
||||
a = TOLOWER (*++needle);
|
||||
}
|
||||
while (TOLOWER (*rhaystack) == (int) a);
|
||||
|
||||
needle = rneedle; /* took the register-poor approach */
|
||||
|
||||
if (a == '\0')
|
||||
break;
|
||||
}
|
||||
ok &= (TOLOWER ((unsigned char) *haystack)
|
||||
== TOLOWER ((unsigned char) *needle));
|
||||
haystack++;
|
||||
needle++;
|
||||
}
|
||||
foundneedle:
|
||||
return (char*) haystack;
|
||||
ret0:
|
||||
return 0;
|
||||
if (*needle)
|
||||
return NULL;
|
||||
if (ok)
|
||||
return (char *) haystack_start;
|
||||
needle_len = needle - needle_start;
|
||||
haystack = haystack_start + 1;
|
||||
haystack_len = needle_len - 1;
|
||||
|
||||
/* Perform the search. Abstract memory is considered to be an array
|
||||
of 'unsigned char' values, not an array of 'char' values. See
|
||||
ISO C 99 section 6.2.6.1. */
|
||||
if (needle_len < LONG_NEEDLE_THRESHOLD)
|
||||
return two_way_short_needle ((const unsigned char *) haystack,
|
||||
haystack_len,
|
||||
(const unsigned char *) needle_start,
|
||||
needle_len);
|
||||
return two_way_long_needle ((const unsigned char *) haystack, haystack_len,
|
||||
(const unsigned char *) needle_start,
|
||||
needle_len);
|
||||
}
|
||||
|
||||
#undef LONG_NEEDLE_THRESHOLD
|
||||
|
||||
weak_alias (__strcasestr, strcasestr)
|
||||
|
142
string/strstr.c
142
string/strstr.c
@ -1,5 +1,5 @@
|
||||
/* Return the offset of one string within another.
|
||||
Copyright (C) 1994,1996,1997,2000,2001,2003 Free Software Foundation, Inc.
|
||||
Copyright (C) 1994,1996,1997,2000,2001,2003,2008 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
|
||||
@ -17,107 +17,71 @@
|
||||
Software Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA
|
||||
02111-1307 USA. */
|
||||
|
||||
/*
|
||||
* My personal strstr() implementation that beats most other algorithms.
|
||||
* Until someone tells me otherwise, I assume that this is the
|
||||
* fastest implementation of strstr() in C.
|
||||
* I deliberately chose not to comment it. You should have at least
|
||||
* as much fun trying to understand it, as I had to write it :-).
|
||||
*
|
||||
* Stephen R. van den Berg, berg@pool.informatik.rwth-aachen.de */
|
||||
/* This particular implementation was written by Eric Blake, 2008. */
|
||||
|
||||
#if HAVE_CONFIG_H
|
||||
#ifndef _LIBC
|
||||
# include <config.h>
|
||||
#endif
|
||||
|
||||
#if defined _LIBC || defined HAVE_STRING_H
|
||||
# include <string.h>
|
||||
/* Specification of strstr. */
|
||||
#include <string.h>
|
||||
|
||||
#include <stdbool.h>
|
||||
|
||||
#ifndef _LIBC
|
||||
# define __builtin_expect(expr, val) (expr)
|
||||
#endif
|
||||
|
||||
typedef unsigned chartype;
|
||||
#define RETURN_TYPE char *
|
||||
#define AVAILABLE(h, h_l, j, n_l) \
|
||||
(!memchr ((h) + (h_l), '\0', (j) + (n_l) - (h_l)) \
|
||||
&& ((h_l) = (j) + (n_l)))
|
||||
#include "str-two-way.h"
|
||||
|
||||
#undef strstr
|
||||
|
||||
/* Return the first occurrence of NEEDLE in HAYSTACK. Return HAYSTACK
|
||||
if NEEDLE is empty, otherwise NULL if NEEDLE is not found in
|
||||
HAYSTACK. */
|
||||
char *
|
||||
strstr (phaystack, pneedle)
|
||||
const char *phaystack;
|
||||
const char *pneedle;
|
||||
strstr (const char *haystack_start, const char *needle_start)
|
||||
{
|
||||
const unsigned char *haystack, *needle;
|
||||
chartype b;
|
||||
const unsigned char *rneedle;
|
||||
const char *haystack = haystack_start;
|
||||
const char *needle = needle_start;
|
||||
size_t needle_len; /* Length of NEEDLE. */
|
||||
size_t haystack_len; /* Known minimum length of HAYSTACK. */
|
||||
bool ok = true; /* True if NEEDLE is prefix of HAYSTACK. */
|
||||
|
||||
haystack = (const unsigned char *) phaystack;
|
||||
/* Determine length of NEEDLE, and in the process, make sure
|
||||
HAYSTACK is at least as long (no point processing all of a long
|
||||
NEEDLE if HAYSTACK is too short). */
|
||||
while (*haystack && *needle)
|
||||
ok &= *haystack++ == *needle++;
|
||||
if (*needle)
|
||||
return NULL;
|
||||
if (ok)
|
||||
return (char *) haystack_start;
|
||||
|
||||
if ((b = *(needle = (const unsigned char *) pneedle)))
|
||||
{
|
||||
chartype c;
|
||||
haystack--; /* possible ANSI violation */
|
||||
/* Reduce the size of haystack using strchr, since it has a smaller
|
||||
linear coefficient than the Two-Way algorithm. */
|
||||
needle_len = needle - needle_start;
|
||||
haystack = strchr (haystack_start + 1, *needle_start);
|
||||
if (!haystack || __builtin_expect (needle_len == 1, 0))
|
||||
return (char *) haystack;
|
||||
needle -= needle_len;
|
||||
haystack_len = (haystack > haystack_start + needle_len ? 1
|
||||
: needle_len + haystack_start - haystack);
|
||||
|
||||
{
|
||||
chartype a;
|
||||
do
|
||||
if (!(a = *++haystack))
|
||||
goto ret0;
|
||||
while (a != b);
|
||||
}
|
||||
|
||||
if (!(c = *++needle))
|
||||
goto foundneedle;
|
||||
++needle;
|
||||
goto jin;
|
||||
|
||||
for (;;)
|
||||
{
|
||||
{
|
||||
chartype a;
|
||||
if (0)
|
||||
jin:{
|
||||
if ((a = *++haystack) == c)
|
||||
goto crest;
|
||||
}
|
||||
else
|
||||
a = *++haystack;
|
||||
do
|
||||
{
|
||||
for (; a != b; a = *++haystack)
|
||||
{
|
||||
if (!a)
|
||||
goto ret0;
|
||||
if ((a = *++haystack) == b)
|
||||
break;
|
||||
if (!a)
|
||||
goto ret0;
|
||||
}
|
||||
}
|
||||
while ((a = *++haystack) != c);
|
||||
}
|
||||
crest:
|
||||
{
|
||||
chartype a;
|
||||
{
|
||||
const unsigned char *rhaystack;
|
||||
if (*(rhaystack = haystack-- + 1) == (a = *(rneedle = needle)))
|
||||
do
|
||||
{
|
||||
if (!a)
|
||||
goto foundneedle;
|
||||
if (*++rhaystack != (a = *++needle))
|
||||
break;
|
||||
if (!a)
|
||||
goto foundneedle;
|
||||
}
|
||||
while (*++rhaystack == (a = *++needle));
|
||||
needle = rneedle; /* took the register-poor aproach */
|
||||
}
|
||||
if (!a)
|
||||
break;
|
||||
}
|
||||
}
|
||||
}
|
||||
foundneedle:
|
||||
return (char *) haystack;
|
||||
ret0:
|
||||
return 0;
|
||||
/* Perform the search. Abstract memory is considered to be an array
|
||||
of 'unsigned char' values, not an array of 'char' values. See
|
||||
ISO C 99 section 6.2.6.1. */
|
||||
if (needle_len < LONG_NEEDLE_THRESHOLD)
|
||||
return two_way_short_needle ((const unsigned char *) haystack,
|
||||
haystack_len,
|
||||
(const unsigned char *) needle, needle_len);
|
||||
return two_way_long_needle ((const unsigned char *) haystack, haystack_len,
|
||||
(const unsigned char *) needle, needle_len);
|
||||
}
|
||||
libc_hidden_builtin_def (strstr)
|
||||
|
||||
#undef LONG_NEEDLE_THRESHOLD
|
||||
|
Loading…
Reference in New Issue
Block a user