textsearch: ts_kmp: support case insensitive searching in Knuth-Morris-Pratt algorithm

Add support for case insensitive search to Knuth-Morris-Pratt algorithm.

Signed-off-by: Joonwoo Park <joonwpark81@gmail.com>
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
This commit is contained in:
Joonwoo Park 2008-07-08 02:38:09 -07:00 committed by David S. Miller
parent 3b76d08190
commit 2523c3fc2b

View file

@ -33,6 +33,7 @@
#include <linux/module.h> #include <linux/module.h>
#include <linux/types.h> #include <linux/types.h>
#include <linux/string.h> #include <linux/string.h>
#include <linux/ctype.h>
#include <linux/textsearch.h> #include <linux/textsearch.h>
struct ts_kmp struct ts_kmp
@ -47,6 +48,7 @@ static unsigned int kmp_find(struct ts_config *conf, struct ts_state *state)
struct ts_kmp *kmp = ts_config_priv(conf); struct ts_kmp *kmp = ts_config_priv(conf);
unsigned int i, q = 0, text_len, consumed = state->offset; unsigned int i, q = 0, text_len, consumed = state->offset;
const u8 *text; const u8 *text;
const int icase = conf->flags & TS_IGNORECASE;
for (;;) { for (;;) {
text_len = conf->get_next_block(consumed, &text, conf, state); text_len = conf->get_next_block(consumed, &text, conf, state);
@ -55,9 +57,11 @@ static unsigned int kmp_find(struct ts_config *conf, struct ts_state *state)
break; break;
for (i = 0; i < text_len; i++) { for (i = 0; i < text_len; i++) {
while (q > 0 && kmp->pattern[q] != text[i]) while (q > 0 && kmp->pattern[q]
!= (icase ? toupper(text[i]) : text[i]))
q = kmp->prefix_tbl[q - 1]; q = kmp->prefix_tbl[q - 1];
if (kmp->pattern[q] == text[i]) if (kmp->pattern[q]
== (icase ? toupper(text[i]) : text[i]))
q++; q++;
if (unlikely(q == kmp->pattern_len)) { if (unlikely(q == kmp->pattern_len)) {
state->offset = consumed + i + 1; state->offset = consumed + i + 1;
@ -72,24 +76,28 @@ static unsigned int kmp_find(struct ts_config *conf, struct ts_state *state)
} }
static inline void compute_prefix_tbl(const u8 *pattern, unsigned int len, static inline void compute_prefix_tbl(const u8 *pattern, unsigned int len,
unsigned int *prefix_tbl) unsigned int *prefix_tbl, int flags)
{ {
unsigned int k, q; unsigned int k, q;
const u8 icase = flags & TS_IGNORECASE;
for (k = 0, q = 1; q < len; q++) { for (k = 0, q = 1; q < len; q++) {
while (k > 0 && pattern[k] != pattern[q]) while (k > 0 && (icase ? toupper(pattern[k]) : pattern[k])
!= (icase ? toupper(pattern[q]) : pattern[q]))
k = prefix_tbl[k-1]; k = prefix_tbl[k-1];
if (pattern[k] == pattern[q]) if ((icase ? toupper(pattern[k]) : pattern[k])
== (icase ? toupper(pattern[q]) : pattern[q]))
k++; k++;
prefix_tbl[q] = k; prefix_tbl[q] = k;
} }
} }
static struct ts_config *kmp_init(const void *pattern, unsigned int len, static struct ts_config *kmp_init(const void *pattern, unsigned int len,
gfp_t gfp_mask) gfp_t gfp_mask, int flags)
{ {
struct ts_config *conf; struct ts_config *conf;
struct ts_kmp *kmp; struct ts_kmp *kmp;
int i;
unsigned int prefix_tbl_len = len * sizeof(unsigned int); unsigned int prefix_tbl_len = len * sizeof(unsigned int);
size_t priv_size = sizeof(*kmp) + len + prefix_tbl_len; size_t priv_size = sizeof(*kmp) + len + prefix_tbl_len;
@ -97,11 +105,16 @@ static struct ts_config *kmp_init(const void *pattern, unsigned int len,
if (IS_ERR(conf)) if (IS_ERR(conf))
return conf; return conf;
conf->flags = flags;
kmp = ts_config_priv(conf); kmp = ts_config_priv(conf);
kmp->pattern_len = len; kmp->pattern_len = len;
compute_prefix_tbl(pattern, len, kmp->prefix_tbl); compute_prefix_tbl(pattern, len, kmp->prefix_tbl, flags);
kmp->pattern = (u8 *) kmp->prefix_tbl + prefix_tbl_len; kmp->pattern = (u8 *) kmp->prefix_tbl + prefix_tbl_len;
memcpy(kmp->pattern, pattern, len); if (flags & TS_IGNORECASE)
for (i = 0; i < len; i++)
kmp->pattern[i] = toupper(((u8 *)pattern)[i]);
else
memcpy(kmp->pattern, pattern, len);
return conf; return conf;
} }