summaryrefslogtreecommitdiff
path: root/strings
diff options
context:
space:
mode:
Diffstat (limited to 'strings')
-rwxr-xr-xstrings/CMakeLists.txt5
-rw-r--r--strings/Makefile.am8
-rw-r--r--strings/bcmp.c4
-rw-r--r--strings/bmove512.c2
-rw-r--r--strings/conf_to_src.c5
-rw-r--r--strings/ctype-extra.c70
-rw-r--r--strings/ctype-mb.c11
-rw-r--r--strings/ctype-sjis.c4
-rw-r--r--strings/ctype-uca.c386
-rw-r--r--strings/ctype-ucs2.c25
-rw-r--r--strings/ctype-utf8.c2
-rw-r--r--strings/ctype.c20
-rw-r--r--strings/decimal.c10
-rw-r--r--strings/llstr.c1
-rw-r--r--strings/strmake.c46
-rw-r--r--strings/strmov.c7
-rw-r--r--strings/strmov_overlapp.c26
17 files changed, 436 insertions, 196 deletions
diff --git a/strings/CMakeLists.txt b/strings/CMakeLists.txt
index 3d9de566670..5ba1c0e5747 100755
--- a/strings/CMakeLists.txt
+++ b/strings/CMakeLists.txt
@@ -23,9 +23,10 @@ SET(STRINGS_SOURCES bchange.c bcmp.c bfill.c bmove512.c bmove_upp.c ctype-big5.c
ctype-latin1.c ctype-mb.c ctype-simple.c ctype-sjis.c ctype-tis620.c ctype-uca.c
ctype-ucs2.c ctype-ujis.c ctype-utf8.c ctype-win1250ch.c ctype.c decimal.c int2str.c
is_prefix.c llstr.c longlong2str.c my_strtoll10.c my_vsnprintf.c r_strinstr.c
- str2int.c str_alloc.c strcend.c strend.c strfill.c strmake.c strmov.c strnmov.c
+ str2int.c str_alloc.c strcend.c strend.c strfill.c strmake.c strmov.c strmov_overlapp.c strnmov.c
strtod.c strtol.c strtoll.c strtoul.c strtoull.c strxmov.c strxnmov.c xml.c
- my_strchr.c strcont.c strinstr.c strnlen.c)
+ my_strchr.c strcont.c strinstr.c strnlen.c
+ strappend.c)
IF(NOT SOURCE_SUBLIBS)
ADD_LIBRARY(strings ${STRINGS_SOURCES})
diff --git a/strings/Makefile.am b/strings/Makefile.am
index db9016b7148..94e24045d03 100644
--- a/strings/Makefile.am
+++ b/strings/Makefile.am
@@ -21,19 +21,19 @@ pkglib_LIBRARIES = libmystrings.a
# Exact one of ASSEMBLER_X
if ASSEMBLER_x86
ASRCS = strings-x86.s longlong2str-x86.s my_strtoll10-x86.s
-CSRCS = bfill.c bmove.c bmove512.c bchange.c strxnmov.c int2str.c str2int.c r_strinstr.c strtod.c bcmp.c strtol.c strtoul.c strtoll.c strtoull.c llstr.c strnlen.c ctype.c ctype-simple.c ctype-mb.c ctype-big5.c ctype-cp932.c ctype-czech.c ctype-eucjpms.c ctype-euc_kr.c ctype-gb2312.c ctype-gbk.c ctype-sjis.c ctype-tis620.c ctype-ujis.c ctype-utf8.c ctype-ucs2.c ctype-uca.c ctype-win1250ch.c ctype-bin.c ctype-latin1.c my_vsnprintf.c xml.c decimal.c ctype-extra.c str_alloc.c longlong2str_asm.c my_strchr.c strmov.c
+CSRCS = bfill.c bmove.c bmove512.c bchange.c strxnmov.c int2str.c str2int.c r_strinstr.c strtod.c bcmp.c strtol.c strtoul.c strtoll.c strtoull.c llstr.c strnlen.c ctype.c ctype-simple.c ctype-mb.c ctype-big5.c ctype-cp932.c ctype-czech.c ctype-eucjpms.c ctype-euc_kr.c ctype-gb2312.c ctype-gbk.c ctype-sjis.c ctype-tis620.c ctype-ujis.c ctype-utf8.c ctype-ucs2.c ctype-uca.c ctype-win1250ch.c ctype-bin.c ctype-latin1.c my_vsnprintf.c xml.c decimal.c ctype-extra.c str_alloc.c longlong2str_asm.c my_strchr.c strmov.c strmov_overlapp.c
else
if ASSEMBLER_sparc32
# These file MUST all be on the same line!! Otherwise automake
# generats a very broken makefile
ASRCS = bmove_upp-sparc.s strappend-sparc.s strend-sparc.s strinstr-sparc.s strmake-sparc.s strmov-sparc.s strnmov-sparc.s strstr-sparc.s
-CSRCS = strcont.c strfill.c strcend.c is_prefix.c longlong2str.c bfill.c bmove.c bmove512.c bchange.c strxnmov.c int2str.c str2int.c r_strinstr.c strtod.c bcmp.c strtol.c strtoul.c strtoll.c strtoull.c llstr.c strnlen.c strxmov.c ctype.c ctype-simple.c ctype-mb.c ctype-big5.c ctype-cp932.c ctype-czech.c ctype-eucjpms.c ctype-euc_kr.c ctype-gb2312.c ctype-gbk.c ctype-sjis.c ctype-tis620.c ctype-ujis.c ctype-utf8.c ctype-ucs2.c ctype-uca.c ctype-win1250ch.c ctype-bin.c ctype-latin1.c my_vsnprintf.c xml.c decimal.c ctype-extra.c my_strtoll10.c str_alloc.c my_strchr.c strmov.c
+CSRCS = strcont.c strfill.c strcend.c is_prefix.c longlong2str.c bfill.c bmove.c bmove512.c bchange.c strxnmov.c int2str.c str2int.c r_strinstr.c strtod.c bcmp.c strtol.c strtoul.c strtoll.c strtoull.c llstr.c strnlen.c strxmov.c ctype.c ctype-simple.c ctype-mb.c ctype-big5.c ctype-cp932.c ctype-czech.c ctype-eucjpms.c ctype-euc_kr.c ctype-gb2312.c ctype-gbk.c ctype-sjis.c ctype-tis620.c ctype-ujis.c ctype-utf8.c ctype-ucs2.c ctype-uca.c ctype-win1250ch.c ctype-bin.c ctype-latin1.c my_vsnprintf.c xml.c decimal.c ctype-extra.c my_strtoll10.c str_alloc.c my_strchr.c strmov.c strmov_overlapp.c
else
#no assembler
ASRCS =
# These file MUST all be on the same line!! Otherwise automake
# generats a very broken makefile
-CSRCS = strxmov.c bmove_upp.c strappend.c strcont.c strend.c strfill.c strcend.c is_prefix.c strstr.c strinstr.c strmake.c strnmov.c strmov.c longlong2str.c bfill.c bmove.c bmove512.c bchange.c strxnmov.c int2str.c str2int.c r_strinstr.c strtod.c bcmp.c strtol.c strtoul.c strtoll.c strtoull.c llstr.c strnlen.c ctype.c ctype-simple.c ctype-mb.c ctype-big5.c ctype-cp932.c ctype-czech.c ctype-eucjpms.c ctype-euc_kr.c ctype-gb2312.c ctype-gbk.c ctype-sjis.c ctype-tis620.c ctype-ujis.c ctype-utf8.c ctype-ucs2.c ctype-uca.c ctype-win1250ch.c ctype-bin.c ctype-latin1.c my_vsnprintf.c xml.c decimal.c ctype-extra.c my_strtoll10.c str_alloc.c my_strchr.c
+CSRCS = strxmov.c bmove_upp.c strappend.c strcont.c strend.c strfill.c strcend.c is_prefix.c strstr.c strinstr.c strmake.c strnmov.c strmov.c longlong2str.c bfill.c bmove.c bmove512.c bchange.c strxnmov.c int2str.c str2int.c r_strinstr.c strtod.c bcmp.c strtol.c strtoul.c strtoll.c strtoull.c llstr.c strnlen.c ctype.c ctype-simple.c ctype-mb.c ctype-big5.c ctype-cp932.c ctype-czech.c ctype-eucjpms.c ctype-euc_kr.c ctype-gb2312.c ctype-gbk.c ctype-sjis.c ctype-tis620.c ctype-ujis.c ctype-utf8.c ctype-ucs2.c ctype-uca.c ctype-win1250ch.c ctype-bin.c ctype-latin1.c my_vsnprintf.c xml.c decimal.c ctype-extra.c my_strtoll10.c str_alloc.c my_strchr.c strmov_overlapp.c
endif
endif
@@ -54,7 +54,7 @@ EXTRA_DIST = ctype-big5.c ctype-cp932.c ctype-czech.c ctype-eucjpms.c ctype-euc
strinstr-sparc.s strmake-sparc.s strmov-sparc.s \
strnmov-sparc.s strstr-sparc.s strxmov-sparc.s \
t_ctype.h my_strchr.c CMakeLists.txt \
- CHARSET_INFO.txt
+ CHARSET_INFO.txt strmov_overlapp.c
libmystrings_a_LIBADD=
conf_to_src_SOURCES = conf_to_src.c xml.c ctype.c bcmp.c
diff --git a/strings/bcmp.c b/strings/bcmp.c
index 1b6ed22fc22..39b5f24e055 100644
--- a/strings/bcmp.c
+++ b/strings/bcmp.c
@@ -24,7 +24,7 @@
#include <my_global.h>
#include "m_string.h"
-#ifdef HAVE_purify
+#ifdef HAVE_valgrind
#undef bcmp
#undef HAVE_BCMP
#endif
@@ -50,7 +50,7 @@ uint len; /* 0 <= len <= 65535 */
#else
-#ifndef HAVE_purify
+#ifndef HAVE_valgrind
size_t bcmp(register const uchar *s1,register const uchar *s2,
register size_t len)
#else
diff --git a/strings/bmove512.c b/strings/bmove512.c
index c3f0446ead6..0ae23d1f42d 100644
--- a/strings/bmove512.c
+++ b/strings/bmove512.c
@@ -33,7 +33,7 @@
#ifdef HAVE_LONG_LONG
#define LONG ulonglong
#else
-#define LONG ulonglong
+#define LONG ulong
#endif
void bmove512(uchar *to, const uchar *from, register size_t length)
diff --git a/strings/conf_to_src.c b/strings/conf_to_src.c
index 7e742050aa8..aa40a269848 100644
--- a/strings/conf_to_src.c
+++ b/strings/conf_to_src.c
@@ -184,11 +184,12 @@ void dispcset(FILE *f,CHARSET_INFO *cs)
{
fprintf(f,"{\n");
fprintf(f," %d,%d,%d,\n",cs->number,0,0);
- fprintf(f," MY_CS_COMPILED%s%s%s%s,\n",
+ fprintf(f," MY_CS_COMPILED%s%s%s%s%s,\n",
cs->state & MY_CS_BINSORT ? "|MY_CS_BINSORT" : "",
cs->state & MY_CS_PRIMARY ? "|MY_CS_PRIMARY" : "",
is_case_sensitive(cs) ? "|MY_CS_CSSORT" : "",
- my_charset_is_8bit_pure_ascii(cs) ? "|MY_CS_PUREASCII" : "");
+ my_charset_is_8bit_pure_ascii(cs) ? "|MY_CS_PUREASCII" : "",
+ !my_charset_is_ascii_compatible(cs) ? "|MY_CS_NONASCII": "");
if (cs->name)
{
diff --git a/strings/ctype-extra.c b/strings/ctype-extra.c
index 75244e40435..64cf30e3673 100644
--- a/strings/ctype-extra.c
+++ b/strings/ctype-extra.c
@@ -6804,7 +6804,7 @@ CHARSET_INFO compiled_charsets[] = {
#ifdef HAVE_CHARSET_swe7
{
10,0,0,
- MY_CS_COMPILED|MY_CS_PRIMARY,
+ MY_CS_COMPILED|MY_CS_PRIMARY|MY_CS_NONASCII,
"swe7", /* cset name */
"swe7_swedish_ci", /* coll name */
"", /* comment */
@@ -8454,7 +8454,7 @@ CHARSET_INFO compiled_charsets[] = {
#ifdef HAVE_CHARSET_swe7
{
82,0,0,
- MY_CS_COMPILED|MY_CS_BINSORT,
+ MY_CS_COMPILED|MY_CS_BINSORT|MY_CS_NONASCII,
"swe7", /* cset name */
"swe7_bin", /* coll name */
"", /* comment */
@@ -8550,72 +8550,6 @@ CHARSET_INFO compiled_charsets[] = {
}
,
#endif
-#ifdef HAVE_CHARSET_geostd8
-{
- 92,0,0,
- MY_CS_COMPILED|MY_CS_PRIMARY,
- "geostd8", /* cset name */
- "geostd8_general_ci", /* coll name */
- "", /* comment */
- NULL, /* tailoring */
- ctype_geostd8_general_ci, /* ctype */
- to_lower_geostd8_general_ci, /* lower */
- to_upper_geostd8_general_ci, /* upper */
- sort_order_geostd8_general_ci, /* sort_order */
- NULL, /* contractions */
- NULL, /* sort_order_big*/
- to_uni_geostd8_general_ci, /* to_uni */
- NULL, /* from_uni */
- my_unicase_default, /* caseinfo */
- NULL, /* state map */
- NULL, /* ident map */
- 1, /* strxfrm_multiply*/
- 1, /* caseup_multiply*/
- 1, /* casedn_multiply*/
- 1, /* mbminlen */
- 1, /* mbmaxlen */
- 0, /* min_sort_char */
- 255, /* max_sort_char */
- ' ', /* pad_char */
- 0, /* escape_with_backslash_is_dangerous */
- &my_charset_8bit_handler,
- &my_collation_8bit_simple_ci_handler,
-}
-,
-#endif
-#ifdef HAVE_CHARSET_geostd8
-{
- 93,0,0,
- MY_CS_COMPILED|MY_CS_BINSORT,
- "geostd8", /* cset name */
- "geostd8_bin", /* coll name */
- "", /* comment */
- NULL, /* tailoring */
- ctype_geostd8_bin, /* ctype */
- to_lower_geostd8_bin, /* lower */
- to_upper_geostd8_bin, /* upper */
- NULL, /* sort_order */
- NULL, /* contractions */
- NULL, /* sort_order_big*/
- to_uni_geostd8_bin, /* to_uni */
- NULL, /* from_uni */
- my_unicase_default, /* caseinfo */
- NULL, /* state map */
- NULL, /* ident map */
- 1, /* strxfrm_multiply*/
- 1, /* caseup_multiply*/
- 1, /* casedn_multiply*/
- 1, /* mbminlen */
- 1, /* mbmaxlen */
- 0, /* min_sort_char */
- 255, /* max_sort_char */
- ' ', /* pad_char */
- 0, /* escape_with_backslash_is_dangerous */
- &my_charset_8bit_handler,
- &my_collation_8bit_bin_handler,
-}
-,
-#endif
#ifdef HAVE_CHARSET_latin1
{
94,0,0,
diff --git a/strings/ctype-mb.c b/strings/ctype-mb.c
index 903811e2ab9..51f93954bd5 100644
--- a/strings/ctype-mb.c
+++ b/strings/ctype-mb.c
@@ -567,8 +567,7 @@ my_bool my_like_range_mb(CHARSET_INFO *cs,
char *min_end= min_str + res_length;
char *max_end= max_str + res_length;
size_t maxcharlen= res_length / cs->mbmaxlen;
- const char *contraction_flags= cs->contractions ?
- ((const char*) cs->contractions) + 0x40*0x40 : NULL;
+ my_bool have_contractions= my_uca_have_contractions(cs);
for (; ptr != end && min_str != min_end && maxcharlen ; maxcharlen--)
{
@@ -636,8 +635,8 @@ fill_max_and_min:
'ab\min\min\min\min' and 'ab\max\max\max\max'.
*/
- if (contraction_flags && ptr + 1 < end &&
- contraction_flags[(uchar) *ptr])
+ if (have_contractions && ptr + 1 < end &&
+ my_uca_can_be_contraction_head(cs, (uchar) *ptr))
{
/* Ptr[0] is a contraction head. */
@@ -659,8 +658,8 @@ fill_max_and_min:
is not a contraction, then we put only ptr[0],
and continue with ptr[1] on the next loop.
*/
- if (contraction_flags[(uchar) ptr[1]] &&
- cs->contractions[(*ptr-0x40)*0x40 + ptr[1] - 0x40])
+ if (my_uca_can_be_contraction_tail(cs, (uchar) ptr[1]) &&
+ my_uca_contraction2_weight(cs, (uchar) ptr[0], (uchar) ptr[1]))
{
/* Contraction found */
if (maxcharlen == 1 || min_str + 1 >= min_end)
diff --git a/strings/ctype-sjis.c b/strings/ctype-sjis.c
index ac426e0d7b5..cc7581ffe6e 100644
--- a/strings/ctype-sjis.c
+++ b/strings/ctype-sjis.c
@@ -4672,7 +4672,7 @@ static MY_CHARSET_HANDLER my_charset_handler=
CHARSET_INFO my_charset_sjis_japanese_ci=
{
13,0,0, /* number */
- MY_CS_COMPILED|MY_CS_PRIMARY|MY_CS_STRNXFRM, /* state */
+ MY_CS_COMPILED|MY_CS_PRIMARY|MY_CS_STRNXFRM|MY_CS_NONASCII, /* state */
"sjis", /* cs name */
"sjis_japanese_ci", /* name */
"", /* comment */
@@ -4704,7 +4704,7 @@ CHARSET_INFO my_charset_sjis_japanese_ci=
CHARSET_INFO my_charset_sjis_bin=
{
88,0,0, /* number */
- MY_CS_COMPILED|MY_CS_BINSORT, /* state */
+ MY_CS_COMPILED|MY_CS_BINSORT|MY_CS_NONASCII, /* state */
"sjis", /* cs name */
"sjis_bin", /* name */
"", /* comment */
diff --git a/strings/ctype-uca.c b/strings/ctype-uca.c
index 6ae0cc3a293..b2a60265a0a 100644
--- a/strings/ctype-uca.c
+++ b/strings/ctype-uca.c
@@ -36,6 +36,12 @@
#include "m_string.h"
#include "m_ctype.h"
+
+#define MY_UCA_CNT_FLAG_SIZE 4096
+#define MY_UCA_CNT_FLAG_MASK 4095
+#define MY_UCA_CNT_HEAD 1
+#define MY_UCA_CNT_TAIL 2
+
#ifdef HAVE_UCA_COLLATIONS
#define MY_UCA_NPAGES 256
@@ -6713,6 +6719,16 @@ static const char hungarian[]=
"&U < \\u00FC <<< \\u00DC << \\u0171 <<< \\u0170";
+static const char croatian[]=
+
+"&C < \\u010D <<< \\u010C < \\u0107 <<< \\u0106 "
+"&D < d\\u017E <<< \\u01C6 <<< D\\u017E <<< \\u01C5 <<< D\\u017D <<< \\u01C4 "
+" < \\u0111 <<< \\u0110 "
+"&L < lj <<< \\u01C9 <<< Lj <<< \\u01C8 <<< LJ <<< \\u01C7 "
+"&N < nj <<< \\u01CC <<< Nj <<< \\u01CB <<< NJ <<< \\u01CA "
+"&S < \\u0161 <<< \\u0160 "
+"&Z < \\u017E <<< \\u017D";
+
/*
Unicode Collation Algorithm:
Collation element (weight) scanner,
@@ -6726,7 +6742,7 @@ typedef struct my_uca_scanner_st
const uchar *send; /* End of the input string */
uchar *uca_length;
uint16 **uca_weight;
- uint16 *contractions;
+ MY_CONTRACTIONS *contractions;
uint16 implicit[2];
int page;
int code;
@@ -6746,6 +6762,75 @@ typedef struct my_uca_scanner_handler_st
static uint16 nochar[]= {0,0};
+/********** Helper functions to handle contraction ************/
+
+
+/**
+ Mark a character as a contraction part
+
+ @cs Pointer to CHARSET_INFO data
+ @wc Unicode code point
+ @flag flag: "is contraction head", "is contraction tail"
+*/
+
+static void
+my_uca_add_contraction_flag(CHARSET_INFO *cs, my_wc_t wc, int flag)
+{
+ cs->contractions->flags[wc & MY_UCA_CNT_FLAG_MASK]|= flag;
+}
+
+
+/**
+ Add a new contraction into contraction list
+
+ @cs Pointer to CHARSET_INFO data
+ @wc Unicode code points of the characters
+ @len Number of characters
+
+ @return New contraction
+ @retval Pointer to a newly added contraction
+*/
+
+static MY_CONTRACTION *
+my_uca_add_contraction(CHARSET_INFO *cs,
+ my_wc_t *wc, int len __attribute__((unused)))
+{
+ MY_CONTRACTIONS *list= cs->contractions;
+ MY_CONTRACTION *next= &list->item[list->nitems];
+ DBUG_ASSERT(len == 2); /* We currently support only contraction2 */
+ next->ch[0]= wc[0];
+ next->ch[1]= wc[1];
+ list->nitems++;
+ return next;
+}
+
+
+/**
+ Allocate and initialize memory for contraction list and flags
+
+ @cs Pointer to CHARSET_INFO data
+ @alloc Memory allocation function (typically points to my_alloc_once)
+ @n Number of contractions
+
+ @return Error code
+ @retval 0 - memory allocated successfully
+ @retval 1 - not enough memory
+*/
+
+static my_bool
+my_uca_alloc_contractions(CHARSET_INFO *cs, void *(*alloc)(size_t), size_t n)
+{
+ uint size= n * sizeof(MY_CONTRACTION);
+ if (!(cs->contractions= (*alloc)(sizeof(MY_CONTRACTIONS))))
+ return 1;
+ bzero(cs->contractions, sizeof(MY_CONTRACTIONS));
+ if (!(cs->contractions->item= (*alloc)(size)) ||
+ !(cs->contractions->flags= (char*) (*alloc)(MY_UCA_CNT_FLAG_SIZE)))
+ return 1;
+ bzero((void*) cs->contractions->item, size);
+ bzero((void*) cs->contractions->flags, MY_UCA_CNT_FLAG_SIZE);
+ return 0;
+}
#ifdef HAVE_CHARSET_ucs2
/*
@@ -6766,7 +6851,7 @@ static uint16 nochar[]= {0,0};
*/
static void my_uca_scanner_init_ucs2(my_uca_scanner *scanner,
- CHARSET_INFO *cs __attribute__((unused)),
+ CHARSET_INFO *cs,
const uchar *str, size_t length)
{
scanner->wbeg= nochar;
@@ -6777,6 +6862,7 @@ static void my_uca_scanner_init_ucs2(my_uca_scanner *scanner,
scanner->uca_length= cs->sort_order;
scanner->uca_weight= cs->sort_order_big;
scanner->contractions= cs->contractions;
+ scanner->cs= cs;
return;
}
@@ -6865,18 +6951,23 @@ static int my_uca_scanner_next_ucs2(my_uca_scanner *scanner)
if (scanner->contractions && (scanner->sbeg <= scanner->send))
{
- int cweight;
+ my_wc_t wc1= ((scanner->page << 8) | scanner->code);
- if (!scanner->page && !scanner->sbeg[0] &&
- (scanner->sbeg[1] > 0x40) && (scanner->sbeg[1] < 0x80) &&
- (scanner->code > 0x40) && (scanner->code < 0x80) &&
- (cweight= scanner->contractions[(scanner->code-0x40)*0x40+scanner->sbeg[1]-0x40]))
+ if (my_uca_can_be_contraction_head(scanner->cs, wc1))
+ {
+ uint16 *cweight;
+ my_wc_t wc2= (((my_wc_t) scanner->sbeg[0]) << 8) | scanner->sbeg[1];
+ if (my_uca_can_be_contraction_tail(scanner->cs, wc2) &&
+ (cweight= my_uca_contraction2_weight(scanner->cs,
+ scanner->code,
+ scanner->sbeg[1])))
{
scanner->implicit[0]= 0;
scanner->wbeg= scanner->implicit;
scanner->sbeg+=2;
- return cweight;
+ return *cweight;
}
+ }
}
if (!ucaw[scanner->page])
@@ -6959,23 +7050,22 @@ static int my_uca_scanner_next_any(my_uca_scanner *scanner)
scanner->code= wc & 0xFF;
scanner->sbeg+= mb_len;
- if (scanner->contractions && !scanner->page &&
- (scanner->code > 0x40) && (scanner->code < 0x80))
+ if (my_uca_have_contractions(scanner->cs) &&
+ my_uca_can_be_contraction_head(scanner->cs, wc))
{
- uint page1, code1, cweight;
+ my_wc_t wc2;
+ uint16 *cweight;
- if (((mb_len= scanner->cs->cset->mb_wc(scanner->cs, &wc,
+ if (((mb_len= scanner->cs->cset->mb_wc(scanner->cs, &wc2,
scanner->sbeg,
scanner->send)) >=0) &&
- (!(page1= (wc >> 8))) &&
- ((code1= (wc & 0xFF)) > 0x40) &&
- (code1 < 0x80) &&
- (cweight= scanner->contractions[(scanner->code-0x40)*0x40 + code1-0x40]))
+ my_uca_can_be_contraction_tail(scanner->cs, wc2) &&
+ (cweight= my_uca_contraction2_weight(scanner->cs, wc, wc2)))
{
scanner->implicit[0]= 0;
scanner->wbeg= scanner->implicit;
scanner->sbeg+= mb_len;
- return cweight;
+ return *cweight;
}
}
@@ -7012,6 +7102,33 @@ static my_uca_scanner_handler my_any_uca_scanner_handler=
my_uca_scanner_next_any
};
+
+
+/**
+ Helper function:
+ Find address of weights of the given character.
+
+ @weights UCA weight array
+ @lengths UCA length array
+ @ch character Unicode code point
+
+ @return Weight array
+ @retval pointer to weight array for the given character,
+ or NULL if this page does not have implicit weights.
+*/
+
+static inline uint16 *
+my_char_weight_addr(CHARSET_INFO *cs, uint wc)
+{
+ uint page= (wc >> 8);
+ uint ofst= wc & 0xFF;
+ return cs->sort_order_big[page] ?
+ cs->sort_order_big[page] + ofst * cs->sort_order[page] :
+ NULL;
+}
+
+
+
/*
Compares two strings according to the collation
@@ -7683,8 +7800,8 @@ ex:
typedef struct my_coll_rule_item_st
{
- uint base; /* Base character */
- uint curr[2]; /* Current character */
+ my_wc_t base; /* Base character */
+ my_wc_t curr[2]; /* Current character */
int diff[3]; /* Primary, Secondary and Tertiary difference */
} MY_COLL_RULE;
@@ -7834,6 +7951,7 @@ static int my_coll_rule_parse(MY_COLL_RULE *rule, size_t mitems,
static my_bool create_tailoring(CHARSET_INFO *cs, void *(*alloc)(size_t))
{
MY_COLL_RULE rule[MY_MAX_COLL_RULE];
+ MY_COLL_RULE *r, *rfirst, *rlast;
char errstr[128];
uchar *newlengths;
uint16 **newweights;
@@ -7858,6 +7976,9 @@ static my_bool create_tailoring(CHARSET_INFO *cs, void *(*alloc)(size_t))
return 1;
}
+ rfirst= rule;
+ rlast= rule + rc;
+
if (!cs->caseinfo)
cs->caseinfo= my_unicase_default;
@@ -7941,44 +8062,21 @@ static my_bool create_tailoring(CHARSET_INFO *cs, void *(*alloc)(size_t))
/* Now process contractions */
if (ncontractions)
{
- /*
- 8K for weights for basic latin letter pairs,
- plus 256 bytes for "is contraction part" flags.
- */
- uint size= 0x40*0x40*sizeof(uint16) + 256;
- char *contraction_flags;
- if (!(cs->contractions= (uint16*) (*alloc)(size)))
- return 1;
- bzero((void*)cs->contractions, size);
- contraction_flags= ((char*) cs->contractions) + 0x40*0x40;
- for (i=0; i < rc; i++)
+ if (my_uca_alloc_contractions(cs, alloc, ncontractions))
+ return 1;
+ for (r= rfirst; r < rlast; r++)
{
- if (rule[i].curr[1])
+ uint16 *to;
+ if (r->curr[1]) /* Contraction */
{
- uint pageb= (rule[i].base >> 8) & 0xFF;
- uint chb= rule[i].base & 0xFF;
- uint16 *offsb= defweights[pageb] + chb*deflengths[pageb];
- uint offsc;
-
- if (offsb[1] ||
- rule[i].curr[0] < 0x40 || rule[i].curr[0] > 0x7f ||
- rule[i].curr[1] < 0x40 || rule[i].curr[1] > 0x7f)
- {
- /*
- TODO: add error reporting;
- We support only basic latin letters contractions at this point.
- Also, We don't support contractions with weight longer than one.
- Otherwise, we'd need much more memory.
- */
- return 1;
- }
- offsc= (rule[i].curr[0]-0x40)*0x40+(rule[i].curr[1]-0x40);
-
- /* Copy base weight applying primary difference */
- cs->contractions[offsc]= offsb[0] + rule[i].diff[0];
- /* Mark both letters as "is contraction part */
- contraction_flags[rule[i].curr[0]]= 1;
- contraction_flags[rule[i].curr[1]]= 1;
+ /* Mark both letters as "is contraction part" */
+ my_uca_add_contraction_flag(cs, r->curr[0], MY_UCA_CNT_HEAD);
+ my_uca_add_contraction_flag(cs, r->curr[1], MY_UCA_CNT_TAIL);
+ to= my_uca_add_contraction(cs, r->curr, 2)->weight;
+ /* Copy weight from the reset character */
+ to[0]= my_char_weight_addr(cs, r->base)[0];
+ /* Apply primary difference */
+ to[0]+= r->diff[0];
}
}
}
@@ -8090,7 +8188,7 @@ MY_COLLATION_HANDLER my_collation_ucs2_uca_handler =
CHARSET_INFO my_charset_ucs2_unicode_ci=
{
128,0,0, /* number */
- MY_CS_COMPILED|MY_CS_STRNXFRM|MY_CS_UNICODE,
+ MY_CS_COMPILED|MY_CS_STRNXFRM|MY_CS_UNICODE|MY_CS_NONASCII,
"ucs2", /* cs name */
"ucs2_unicode_ci", /* name */
"", /* comment */
@@ -8122,7 +8220,7 @@ CHARSET_INFO my_charset_ucs2_unicode_ci=
CHARSET_INFO my_charset_ucs2_icelandic_uca_ci=
{
129,0,0, /* number */
- MY_CS_COMPILED|MY_CS_STRNXFRM|MY_CS_UNICODE,
+ MY_CS_COMPILED|MY_CS_STRNXFRM|MY_CS_UNICODE|MY_CS_NONASCII,
"ucs2", /* cs name */
"ucs2_icelandic_ci",/* name */
"", /* comment */
@@ -8154,7 +8252,7 @@ CHARSET_INFO my_charset_ucs2_icelandic_uca_ci=
CHARSET_INFO my_charset_ucs2_latvian_uca_ci=
{
130,0,0, /* number */
- MY_CS_COMPILED|MY_CS_STRNXFRM|MY_CS_UNICODE,
+ MY_CS_COMPILED|MY_CS_STRNXFRM|MY_CS_UNICODE|MY_CS_NONASCII,
"ucs2", /* cs name */
"ucs2_latvian_ci", /* name */
"", /* comment */
@@ -8186,7 +8284,7 @@ CHARSET_INFO my_charset_ucs2_latvian_uca_ci=
CHARSET_INFO my_charset_ucs2_romanian_uca_ci=
{
131,0,0, /* number */
- MY_CS_COMPILED|MY_CS_STRNXFRM|MY_CS_UNICODE,
+ MY_CS_COMPILED|MY_CS_STRNXFRM|MY_CS_UNICODE|MY_CS_NONASCII,
"ucs2", /* cs name */
"ucs2_romanian_ci", /* name */
"", /* comment */
@@ -8218,7 +8316,7 @@ CHARSET_INFO my_charset_ucs2_romanian_uca_ci=
CHARSET_INFO my_charset_ucs2_slovenian_uca_ci=
{
132,0,0, /* number */
- MY_CS_COMPILED|MY_CS_STRNXFRM|MY_CS_UNICODE,
+ MY_CS_COMPILED|MY_CS_STRNXFRM|MY_CS_UNICODE|MY_CS_NONASCII,
"ucs2", /* cs name */
"ucs2_slovenian_ci",/* name */
"", /* comment */
@@ -8250,7 +8348,7 @@ CHARSET_INFO my_charset_ucs2_slovenian_uca_ci=
CHARSET_INFO my_charset_ucs2_polish_uca_ci=
{
133,0,0, /* number */
- MY_CS_COMPILED|MY_CS_STRNXFRM|MY_CS_UNICODE,
+ MY_CS_COMPILED|MY_CS_STRNXFRM|MY_CS_UNICODE|MY_CS_NONASCII,
"ucs2", /* cs name */
"ucs2_polish_ci", /* name */
"", /* comment */
@@ -8282,7 +8380,7 @@ CHARSET_INFO my_charset_ucs2_polish_uca_ci=
CHARSET_INFO my_charset_ucs2_estonian_uca_ci=
{
134,0,0, /* number */
- MY_CS_COMPILED|MY_CS_STRNXFRM|MY_CS_UNICODE,
+ MY_CS_COMPILED|MY_CS_STRNXFRM|MY_CS_UNICODE|MY_CS_NONASCII,
"ucs2", /* cs name */
"ucs2_estonian_ci", /* name */
"", /* comment */
@@ -8314,7 +8412,7 @@ CHARSET_INFO my_charset_ucs2_estonian_uca_ci=
CHARSET_INFO my_charset_ucs2_spanish_uca_ci=
{
135,0,0, /* number */
- MY_CS_COMPILED|MY_CS_STRNXFRM|MY_CS_UNICODE,
+ MY_CS_COMPILED|MY_CS_STRNXFRM|MY_CS_UNICODE|MY_CS_NONASCII,
"ucs2", /* cs name */
"ucs2_spanish_ci", /* name */
"", /* comment */
@@ -8346,7 +8444,7 @@ CHARSET_INFO my_charset_ucs2_spanish_uca_ci=
CHARSET_INFO my_charset_ucs2_swedish_uca_ci=
{
136,0,0, /* number */
- MY_CS_COMPILED|MY_CS_STRNXFRM|MY_CS_UNICODE,
+ MY_CS_COMPILED|MY_CS_STRNXFRM|MY_CS_UNICODE|MY_CS_NONASCII,
"ucs2", /* cs name */
"ucs2_swedish_ci", /* name */
"", /* comment */
@@ -8378,7 +8476,7 @@ CHARSET_INFO my_charset_ucs2_swedish_uca_ci=
CHARSET_INFO my_charset_ucs2_turkish_uca_ci=
{
137,0,0, /* number */
- MY_CS_COMPILED|MY_CS_STRNXFRM|MY_CS_UNICODE,
+ MY_CS_COMPILED|MY_CS_STRNXFRM|MY_CS_UNICODE|MY_CS_NONASCII,
"ucs2", /* cs name */
"ucs2_turkish_ci", /* name */
"", /* comment */
@@ -8410,7 +8508,7 @@ CHARSET_INFO my_charset_ucs2_turkish_uca_ci=
CHARSET_INFO my_charset_ucs2_czech_uca_ci=
{
138,0,0, /* number */
- MY_CS_COMPILED|MY_CS_STRNXFRM|MY_CS_UNICODE,
+ MY_CS_COMPILED|MY_CS_STRNXFRM|MY_CS_UNICODE|MY_CS_NONASCII,
"ucs2", /* cs name */
"ucs2_czech_ci", /* name */
"", /* comment */
@@ -8443,7 +8541,7 @@ CHARSET_INFO my_charset_ucs2_czech_uca_ci=
CHARSET_INFO my_charset_ucs2_danish_uca_ci=
{
139,0,0, /* number */
- MY_CS_COMPILED|MY_CS_STRNXFRM|MY_CS_UNICODE,
+ MY_CS_COMPILED|MY_CS_STRNXFRM|MY_CS_UNICODE|MY_CS_NONASCII,
"ucs2", /* cs name */
"ucs2_danish_ci", /* name */
"", /* comment */
@@ -8475,7 +8573,7 @@ CHARSET_INFO my_charset_ucs2_danish_uca_ci=
CHARSET_INFO my_charset_ucs2_lithuanian_uca_ci=
{
140,0,0, /* number */
- MY_CS_COMPILED|MY_CS_STRNXFRM|MY_CS_UNICODE,
+ MY_CS_COMPILED|MY_CS_STRNXFRM|MY_CS_UNICODE|MY_CS_NONASCII,
"ucs2", /* cs name */
"ucs2_lithuanian_ci",/* name */
"", /* comment */
@@ -8507,7 +8605,7 @@ CHARSET_INFO my_charset_ucs2_lithuanian_uca_ci=
CHARSET_INFO my_charset_ucs2_slovak_uca_ci=
{
141,0,0, /* number */
- MY_CS_COMPILED|MY_CS_STRNXFRM|MY_CS_UNICODE,
+ MY_CS_COMPILED|MY_CS_STRNXFRM|MY_CS_UNICODE|MY_CS_NONASCII,
"ucs2", /* cs name */
"ucs2_slovak_ci", /* name */
"", /* comment */
@@ -8539,7 +8637,7 @@ CHARSET_INFO my_charset_ucs2_slovak_uca_ci=
CHARSET_INFO my_charset_ucs2_spanish2_uca_ci=
{
142,0,0, /* number */
- MY_CS_COMPILED|MY_CS_STRNXFRM|MY_CS_UNICODE,
+ MY_CS_COMPILED|MY_CS_STRNXFRM|MY_CS_UNICODE|MY_CS_NONASCII,
"ucs2", /* cs name */
"ucs2_spanish2_ci", /* name */
"", /* comment */
@@ -8572,7 +8670,7 @@ CHARSET_INFO my_charset_ucs2_spanish2_uca_ci=
CHARSET_INFO my_charset_ucs2_roman_uca_ci=
{
143,0,0, /* number */
- MY_CS_COMPILED|MY_CS_STRNXFRM|MY_CS_UNICODE,
+ MY_CS_COMPILED|MY_CS_STRNXFRM|MY_CS_UNICODE|MY_CS_NONASCII,
"ucs2", /* cs name */
"ucs2_roman_ci", /* name */
"", /* comment */
@@ -8605,7 +8703,7 @@ CHARSET_INFO my_charset_ucs2_roman_uca_ci=
CHARSET_INFO my_charset_ucs2_persian_uca_ci=
{
144,0,0, /* number */
- MY_CS_COMPILED|MY_CS_STRNXFRM|MY_CS_UNICODE,
+ MY_CS_COMPILED|MY_CS_STRNXFRM|MY_CS_UNICODE|MY_CS_NONASCII,
"ucs2", /* cs name */
"ucs2_persian_ci", /* name */
"", /* comment */
@@ -8638,7 +8736,7 @@ CHARSET_INFO my_charset_ucs2_persian_uca_ci=
CHARSET_INFO my_charset_ucs2_esperanto_uca_ci=
{
145,0,0, /* number */
- MY_CS_COMPILED|MY_CS_STRNXFRM|MY_CS_UNICODE,
+ MY_CS_COMPILED|MY_CS_STRNXFRM|MY_CS_UNICODE|MY_CS_NONASCII,
"ucs2", /* cs name */
"ucs2_esperanto_ci",/* name */
"", /* comment */
@@ -8671,7 +8769,7 @@ CHARSET_INFO my_charset_ucs2_esperanto_uca_ci=
CHARSET_INFO my_charset_ucs2_hungarian_uca_ci=
{
146,0,0, /* number */
- MY_CS_COMPILED|MY_CS_STRNXFRM|MY_CS_UNICODE,
+ MY_CS_COMPILED|MY_CS_STRNXFRM|MY_CS_UNICODE|MY_CS_NONASCII,
"ucs2", /* cs name */
"ucs2_hungarian_ci",/* name */
"", /* comment */
@@ -8701,6 +8799,39 @@ CHARSET_INFO my_charset_ucs2_hungarian_uca_ci=
};
+CHARSET_INFO my_charset_ucs2_croatian_uca_ci=
+{
+ 149,0,0, /* number */
+ MY_CS_COMPILED|MY_CS_STRNXFRM|MY_CS_UNICODE,
+ "ucs2", /* cs name */
+ "ucs2_croatian_ci", /* name */
+ "", /* comment */
+ croatian, /* tailoring */
+ NULL, /* ctype */
+ NULL, /* to_lower */
+ NULL, /* to_upper */
+ NULL, /* sort_order */
+ NULL, /* contractions */
+ NULL, /* sort_order_big*/
+ NULL, /* tab_to_uni */
+ NULL, /* tab_from_uni */
+ my_unicase_default, /* caseinfo */
+ NULL, /* state_map */
+ NULL, /* ident_map */
+ 8, /* strxfrm_multiply */
+ 1, /* caseup_multiply */
+ 1, /* casedn_multiply */
+ 2, /* mbminlen */
+ 2, /* mbmaxlen */
+ 9, /* min_sort_char */
+ 0xFFFF, /* max_sort_char */
+ ' ', /* pad char */
+ 0, /* escape_with_backslash_is_dangerous */
+ &my_charset_ucs2_handler,
+ &my_collation_ucs2_uca_handler
+};
+
+
#endif
@@ -9358,6 +9489,113 @@ CHARSET_INFO my_charset_utf8_hungarian_uca_ci=
&my_collation_any_uca_handler
};
+CHARSET_INFO my_charset_utf8_croatian_uca_ci=
+{
+ 213,0,0, /* number */
+ MY_CS_COMPILED|MY_CS_STRNXFRM|MY_CS_UNICODE,
+ "utf8", /* cs name */
+ "utf8_croatian_ci", /* name */
+ "", /* comment */
+ croatian, /* tailoring */
+ ctype_utf8, /* ctype */
+ NULL, /* to_lower */
+ NULL, /* to_upper */
+ NULL, /* sort_order */
+ NULL, /* contractions */
+ NULL, /* sort_order_big*/
+ NULL, /* tab_to_uni */
+ NULL, /* tab_from_uni */
+ my_unicase_default, /* caseinfo */
+ NULL, /* state_map */
+ NULL, /* ident_map */
+ 8, /* strxfrm_multiply */
+ 1, /* caseup_multiply */
+ 1, /* casedn_multiply */
+ 1, /* mbminlen */
+ 3, /* mbmaxlen */
+ 9, /* min_sort_char */
+ 0xFFFF, /* max_sort_char */
+ ' ', /* pad char */
+ 0, /* escape_with_backslash_is_dangerous */
+ &my_charset_utf8_handler,
+ &my_collation_any_uca_handler
+};
+
#endif /* HAVE_CHARSET_utf8 */
#endif /* HAVE_UCA_COLLATIONS */
+
+/**
+ Check if UCA data has contractions (public version)
+
+ @cs Pointer to CHARSET_INFO data
+ @retval 0 - no contraction, 1 - have contractions.
+*/
+
+my_bool
+my_uca_have_contractions(CHARSET_INFO *cs)
+{
+ return cs->contractions != NULL;
+}
+
+/**
+ Check if a character can be contraction head
+
+ @cs Pointer to CHARSET_INFO data
+ @wc Code point
+
+ @retval 0 - cannot be contraction head
+ @retval 1 - can be contraction head
+*/
+
+my_bool
+my_uca_can_be_contraction_head(CHARSET_INFO *cs, my_wc_t wc)
+{
+ return cs->contractions->flags[wc & MY_UCA_CNT_FLAG_MASK] & MY_UCA_CNT_HEAD;
+}
+
+
+/**
+ Check if a character can be contraction tail
+
+ @cs Pointer to CHARSET_INFO data
+ @wc Code point
+
+ @retval 0 - cannot be contraction tail
+ @retval 1 - can be contraction tail
+*/
+
+my_bool
+my_uca_can_be_contraction_tail(CHARSET_INFO *cs, my_wc_t wc)
+{
+ return cs->contractions->flags[wc & MY_UCA_CNT_FLAG_MASK] & MY_UCA_CNT_TAIL;
+}
+
+
+/**
+ Find a contraction and return its weight array
+
+ @cs Pointer to CHARSET data
+ @wc1 First character
+ @wc2 Second character
+
+ @return Weight array
+ @retval NULL - no contraction found
+ @retval ptr - contraction weight array
+*/
+
+uint16 *
+my_uca_contraction2_weight(CHARSET_INFO *cs, my_wc_t wc1, my_wc_t wc2)
+{
+ MY_CONTRACTIONS *list= cs->contractions;
+ MY_CONTRACTION *c, *last;
+ for (c= list->item, last= &list->item[list->nitems]; c < last; c++)
+ {
+ if (c->ch[0] == wc1 && c->ch[1] == wc2)
+ {
+ return c->weight;
+ }
+ }
+ return NULL;
+}
+
diff --git a/strings/ctype-ucs2.c b/strings/ctype-ucs2.c
index cead55f8a0a..bf9c6f9626a 100644
--- a/strings/ctype-ucs2.c
+++ b/strings/ctype-ucs2.c
@@ -1498,6 +1498,14 @@ void my_hash_sort_ucs2_bin(CHARSET_INFO *cs __attribute__((unused)),
}
}
+
+static inline my_wc_t
+ucs2_to_wc(const uchar *ptr)
+{
+ return (((uint) ptr[0]) << 8) + ptr[1];
+}
+
+
/*
** Calculate min_str and max_str that ranges a LIKE string.
** Arguments:
@@ -1526,12 +1534,12 @@ my_bool my_like_range_ucs2(CHARSET_INFO *cs,
char *min_org=min_str;
char *min_end=min_str+res_length;
size_t charlen= res_length / cs->mbmaxlen;
- const char *contraction_flags= cs->contractions ?
- ((const char*) cs->contractions) + 0x40*0x40 : NULL;
+ my_bool have_contractions= my_uca_have_contractions(cs);
for ( ; ptr + 1 < end && min_str + 1 < min_end && charlen > 0
; ptr+=2, charlen--)
{
+ my_wc_t wc;
if (ptr[0] == '\0' && ptr[1] == escape && ptr + 1 < end)
{
ptr+=2; /* Skip escape */
@@ -1567,9 +1575,10 @@ fill_max_and_min:
return 0;
}
- if (contraction_flags && ptr + 3 < end &&
- ptr[0] == '\0' && contraction_flags[(uchar) ptr[1]])
+ if (have_contractions && ptr + 3 < end &&
+ my_uca_can_be_contraction_head(cs, (wc= ucs2_to_wc((uchar*) ptr))))
{
+ my_wc_t wc2;
/* Contraction head found */
if (ptr[2] == '\0' && (ptr[3] == w_one || ptr[3] == w_many))
{
@@ -1581,8 +1590,8 @@ fill_max_and_min:
Check if the second letter can be contraction part,
and if two letters really produce a contraction.
*/
- if (ptr[2] == '\0' && contraction_flags[(uchar) ptr[3]] &&
- cs->contractions[(ptr[1]-0x40)*0x40 + ptr[3] - 0x40])
+ if (my_uca_can_be_contraction_tail(cs, (wc2= ucs2_to_wc((uchar*) ptr + 2))) &&
+ my_uca_contraction2_weight(cs, wc , wc2))
{
/* Contraction found */
if (charlen == 1 || min_str + 2 >= min_end)
@@ -1702,7 +1711,7 @@ MY_CHARSET_HANDLER my_charset_ucs2_handler=
CHARSET_INFO my_charset_ucs2_general_ci=
{
35,0,0, /* number */
- MY_CS_COMPILED|MY_CS_PRIMARY|MY_CS_STRNXFRM|MY_CS_UNICODE,
+ MY_CS_COMPILED|MY_CS_PRIMARY|MY_CS_STRNXFRM|MY_CS_UNICODE|MY_CS_NONASCII,
"ucs2", /* cs name */
"ucs2_general_ci", /* name */
"", /* comment */
@@ -1734,7 +1743,7 @@ CHARSET_INFO my_charset_ucs2_general_ci=
CHARSET_INFO my_charset_ucs2_bin=
{
90,0,0, /* number */
- MY_CS_COMPILED|MY_CS_BINSORT|MY_CS_UNICODE,
+ MY_CS_COMPILED|MY_CS_BINSORT|MY_CS_UNICODE|MY_CS_NONASCII,
"ucs2", /* cs name */
"ucs2_bin", /* name */
"", /* comment */
diff --git a/strings/ctype-utf8.c b/strings/ctype-utf8.c
index ae942b59caa..91f633e45ce 100644
--- a/strings/ctype-utf8.c
+++ b/strings/ctype-utf8.c
@@ -4200,7 +4200,7 @@ static MY_CHARSET_HANDLER my_charset_filename_handler=
CHARSET_INFO my_charset_filename=
{
17,0,0, /* number */
- MY_CS_COMPILED|MY_CS_PRIMARY|MY_CS_STRNXFRM|MY_CS_UNICODE|MY_CS_HIDDEN,
+ MY_CS_COMPILED|MY_CS_PRIMARY|MY_CS_STRNXFRM|MY_CS_UNICODE|MY_CS_HIDDEN|MY_CS_NONASCII,
"filename", /* cs name */
"filename", /* name */
"", /* comment */
diff --git a/strings/ctype.c b/strings/ctype.c
index 17ad1256e74..75d76aceea3 100644
--- a/strings/ctype.c
+++ b/strings/ctype.c
@@ -405,3 +405,23 @@ my_charset_is_8bit_pure_ascii(CHARSET_INFO *cs)
}
return 1;
}
+
+
+/*
+ Shared function between conf_to_src and mysys.
+ Check if a 8bit character set is compatible with
+ ascii on the range 0x00..0x7F.
+*/
+my_bool
+my_charset_is_ascii_compatible(CHARSET_INFO *cs)
+{
+ uint i;
+ if (!cs->tab_to_uni)
+ return 1;
+ for (i= 0; i < 128; i++)
+ {
+ if (cs->tab_to_uni[i] != i)
+ return 0;
+ }
+ return 1;
+}
diff --git a/strings/decimal.c b/strings/decimal.c
index 282e7cae8ab..184f78a20b7 100644
--- a/strings/decimal.c
+++ b/strings/decimal.c
@@ -30,7 +30,7 @@
integer that determines the number of significant digits in a
particular radix R, where R is either 2 or 10. S is a non-negative
integer. Every value of an exact numeric type of scale S is of the
- form n*10^{-S}, where n is an integer such that ­-R^P <= n <= R^P.
+ form n*10^{-S}, where n is an integer such that �-R^P <= n <= R^P.
[...]
@@ -144,7 +144,7 @@ static double scaler1[]= {
1.0, 10.0, 1e2, 1e3, 1e4, 1e5, 1e6, 1e7, 1e8, 1e9
};
-#ifdef HAVE_purify
+#ifdef HAVE_valgrind
#define sanity(d) DBUG_ASSERT((d)->len > 0)
#else
#define sanity(d) DBUG_ASSERT((d)->len >0 && ((d)->buf[0] | \
@@ -306,7 +306,7 @@ int decimal_actual_fraction(decimal_t *from)
{
for (i= DIG_PER_DEC1 - ((frac - 1) % DIG_PER_DEC1);
*buf0 % powers10[i++] == 0;
- frac--) ;
+ frac--) {}
}
return frac;
}
@@ -500,7 +500,7 @@ static void digits_bounds(decimal_t *from, int *start_result, int *end_result)
stop= (int) ((buf_end - from->buf + 1) * DIG_PER_DEC1);
i= 1;
}
- for (; *buf_end % powers10[i++] == 0; stop--) ;
+ for (; *buf_end % powers10[i++] == 0; stop--) {}
*end_result= stop; /* index of position after last decimal digit (from 0) */
}
@@ -1011,7 +1011,7 @@ static int ull2dec(ulonglong from, decimal_t *to)
sanity(to);
- for (intg1=1; from >= DIG_BASE; intg1++, from/=DIG_BASE) ;
+ for (intg1=1; from >= DIG_BASE; intg1++, from/=DIG_BASE) {}
if (unlikely(intg1 > to->len))
{
intg1=to->len;
diff --git a/strings/llstr.c b/strings/llstr.c
index 643cf36a311..678f8b05f39 100644
--- a/strings/llstr.c
+++ b/strings/llstr.c
@@ -38,3 +38,4 @@ char *ullstr(longlong value,char *buff)
longlong10_to_str(value,buff,10);
return buff;
}
+
diff --git a/strings/strmake.c b/strings/strmake.c
index 2d5fa5e36aa..56bd3a8f084 100644
--- a/strings/strmake.c
+++ b/strings/strmake.c
@@ -29,26 +29,38 @@
char *strmake(register char *dst, register const char *src, size_t length)
{
-#ifdef EXTRA_DEBUG
- /*
- 'length' is the maximum length of the string; the buffer needs
- to be one character larger to accomodate the terminating '\0'.
- This is easy to get wrong, so we make sure we write to the
- entire length of the buffer to identify incorrect buffer-sizes.
- We only initialise the "unused" part of the buffer here, a) for
- efficiency, and b) because dst==src is allowed, so initialising
- the entire buffer would overwrite the source-string. Also, we
- write a character rather than '\0' as this makes spotting these
- problems in the results easier.
- */
- uint n= 0;
- while (n < length && src[n++]);
- memset(dst + n, (int) 'Z', length - n + 1);
-#endif
-
while (length--)
+ {
if (! (*dst++ = *src++))
+ {
+#ifdef EXTRA_DEBUG
+ /*
+ 'length' is the maximum length of the string; the buffer needs
+ to be one character larger to accommodate the terminating
+ '\0'. This is easy to get wrong, so we make sure we write to
+ the entire length of the buffer to identify incorrect
+ buffer-sizes. We only initialism the "unused" part of the
+ buffer here, a) for efficiency, and b) because dst==src is
+ allowed, so initializing the entire buffer would overwrite the
+ source-string. Also, we write a character rather than '\0' as
+ this makes spotting these problems in the results easier.
+
+ If we are using purify/valgrind, we only set one character at
+ end to be able to detect also wrong accesses after the end of
+ dst.
+ */
+ if (length)
+ {
+#ifdef HAVE_valgrind
+ dst[length-1]= 'Z';
+#else
+ bfill(dst, length-1, (int) 'Z');
+#endif /* HAVE_valgrind */
+ }
+#endif /* EXTRA_DEBUG */
return dst-1;
+ }
+ }
*dst=0;
return dst;
}
diff --git a/strings/strmov.c b/strings/strmov.c
index eedf22a4ef1..5112ad8d90e 100644
--- a/strings/strmov.c
+++ b/strings/strmov.c
@@ -24,10 +24,7 @@
#include <my_global.h>
#include "m_string.h"
-#ifdef strmov
-#undef strmov
-#define strmov strmov_overlapp
-#endif
+#ifndef strmov
#if !defined(MC68000) && !defined(DS90)
@@ -51,3 +48,5 @@ char *strmov(dst, src)
}
#endif
+
+#endif /* strmov */
diff --git a/strings/strmov_overlapp.c b/strings/strmov_overlapp.c
new file mode 100644
index 00000000000..4cc3e294620
--- /dev/null
+++ b/strings/strmov_overlapp.c
@@ -0,0 +1,26 @@
+/* Copyright (C) 2000 MySQL AB
+
+ 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; version 2 of the License.
+
+ 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, write to the Free Software
+ Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA */
+
+#include <my_global.h>
+#include "m_string.h"
+
+/* A trivial implementation */
+char *strmov_overlapp(char *dst, const char *src)
+{
+ size_t len= strlen(src);
+ memmove(dst, src, len+1);
+ return dst+len;
+}
+
View Lorry Controller Status