summaryrefslogtreecommitdiff
path: root/strings
diff options
context:
space:
mode:
authorunknown <msvensson@neptunus.(none)>2006-11-23 18:38:27 +0100
committerunknown <msvensson@neptunus.(none)>2006-11-23 18:38:27 +0100
commit32f349ed757a265b03b30e3635775ead21bdb54e (patch)
tree4de70d98bf438e372e1b73c5e179846ca768e6c8 /strings
parent5c9d6a68e314cb8492417eb50bb2e8ea8e5c7200 (diff)
parent07d18b0f6761208f1f24e040bfea723274240e3f (diff)
downloadmariadb-git-32f349ed757a265b03b30e3635775ead21bdb54e.tar.gz
Merge neptunus.(none):/home/msvensson/mysql/mysql-5.1
into neptunus.(none):/home/msvensson/mysql/mysql-5.1-new-maint BitKeeper/etc/collapsed: auto-union configure.in: Auto merged mysql-test/Makefile.am: Auto merged mysql-test/extra/rpl_tests/rpl_deadlock.test: Auto merged mysql-test/mysql-test-run-shell.sh: Auto merged mysql-test/r/func_time.result: Auto merged mysql-test/r/ps.result: Auto merged mysql-test/r/windows.result: Auto merged mysql-test/t/disabled.def: Auto merged mysql-test/t/func_time.test: Auto merged mysql-test/t/ps.test: Auto merged mysql-test/t/windows.test: Auto merged sql/item.cc: Auto merged sql/item_timefunc.cc: Auto merged sql/mysqld.cc: Auto merged sql-common/my_time.c: Auto merged sql/sql_parse.cc: Auto merged BitKeeper/deleted/.del-ps_not_windows.result: Delete: mysql-test/r/ps_not_windows.result BitKeeper/deleted/.del-ps_not_windows.test: Delete: mysql-test/t/ps_not_windows.test mysql-test/mysql-test-run.pl: Manual merge mysql-test/r/rpl_deadlock_innodb.result: Manual merge
Diffstat (limited to 'strings')
-rw-r--r--strings/ctype-bin.c3
-rw-r--r--strings/ctype-mb.c24
-rw-r--r--strings/ctype-simple.c18
-rw-r--r--strings/ctype-ucs2.c8
-rw-r--r--strings/ctype-utf8.c136
5 files changed, 163 insertions, 26 deletions
diff --git a/strings/ctype-bin.c b/strings/ctype-bin.c
index 5758960ef6c..289b76c9e6e 100644
--- a/strings/ctype-bin.c
+++ b/strings/ctype-bin.c
@@ -211,9 +211,10 @@ static int my_strnncollsp_8bit_bin(CHARSET_INFO * cs __attribute__((unused)),
/* This function is used for all conversion functions */
-static void my_case_str_bin(CHARSET_INFO *cs __attribute__((unused)),
+static uint my_case_str_bin(CHARSET_INFO *cs __attribute__((unused)),
char *str __attribute__((unused)))
{
+ return 0;
}
static uint my_case_bin(CHARSET_INFO *cs __attribute__((unused)),
diff --git a/strings/ctype-mb.c b/strings/ctype-mb.c
index c945164ac9c..c3848c64219 100644
--- a/strings/ctype-mb.c
+++ b/strings/ctype-mb.c
@@ -21,40 +21,44 @@
#ifdef USE_MB
-void my_caseup_str_mb(CHARSET_INFO * cs, char *str)
+uint my_caseup_str_mb(CHARSET_INFO * cs, char *str)
{
register uint32 l;
- register uchar *map=cs->to_upper;
+ register uchar *map= cs->to_upper;
+ char *str_orig= str;
while (*str)
{
/* Pointing after the '\0' is safe here. */
- if ((l=my_ismbchar(cs, str, str + cs->mbmaxlen)))
- str+=l;
+ if ((l= my_ismbchar(cs, str, str + cs->mbmaxlen)))
+ str+= l;
else
{
- *str=(char) map[(uchar)*str];
+ *str= (char) map[(uchar)*str];
str++;
}
}
+ return str - str_orig;
}
-void my_casedn_str_mb(CHARSET_INFO * cs, char *str)
+uint my_casedn_str_mb(CHARSET_INFO * cs, char *str)
{
register uint32 l;
- register uchar *map=cs->to_lower;
+ register uchar *map= cs->to_lower;
+ char *str_orig= str;
while (*str)
{
/* Pointing after the '\0' is safe here. */
- if ((l=my_ismbchar(cs, str, str + cs->mbmaxlen)))
- str+=l;
+ if ((l= my_ismbchar(cs, str, str + cs->mbmaxlen)))
+ str+= l;
else
{
- *str=(char) map[(uchar)*str];
+ *str= (char) map[(uchar)*str];
str++;
}
}
+ return str - str_orig;
}
uint my_caseup_mb(CHARSET_INFO * cs, char *src, uint srclen,
diff --git a/strings/ctype-simple.c b/strings/ctype-simple.c
index 9b45d5a03b7..9d10ba82114 100644
--- a/strings/ctype-simple.c
+++ b/strings/ctype-simple.c
@@ -188,20 +188,26 @@ int my_strnncollsp_simple(CHARSET_INFO * cs, const uchar *a, uint a_length,
}
-void my_caseup_str_8bit(CHARSET_INFO * cs,char *str)
+uint my_caseup_str_8bit(CHARSET_INFO * cs,char *str)
{
- register uchar *map=cs->to_upper;
- while ((*str = (char) map[(uchar) *str]) != 0)
+ register uchar *map= cs->to_upper;
+ char *str_orig= str;
+ while ((*str= (char) map[(uchar) *str]) != 0)
str++;
+ return str - str_orig;
}
-void my_casedn_str_8bit(CHARSET_INFO * cs,char *str)
+
+uint my_casedn_str_8bit(CHARSET_INFO * cs,char *str)
{
- register uchar *map=cs->to_lower;
- while ((*str = (char) map[(uchar)*str]) != 0)
+ register uchar *map= cs->to_lower;
+ char *str_orig= str;
+ while ((*str= (char) map[(uchar) *str]) != 0)
str++;
+ return str - str_orig;
}
+
uint my_caseup_8bit(CHARSET_INFO * cs, char *src, uint srclen,
char *dst __attribute__((unused)),
uint dstlen __attribute__((unused)))
diff --git a/strings/ctype-ucs2.c b/strings/ctype-ucs2.c
index df43eff3d73..5089db6bf48 100644
--- a/strings/ctype-ucs2.c
+++ b/strings/ctype-ucs2.c
@@ -159,13 +159,13 @@ static void my_hash_sort_ucs2(CHARSET_INFO *cs, const uchar *s, uint slen,
}
-static void my_caseup_str_ucs2(CHARSET_INFO * cs __attribute__((unused)),
+static uint my_caseup_str_ucs2(CHARSET_INFO * cs __attribute__((unused)),
char * s __attribute__((unused)))
{
+ return 0;
}
-
static uint my_casedn_ucs2(CHARSET_INFO *cs, char *src, uint srclen,
char *dst __attribute__((unused)),
uint dstlen __attribute__((unused)))
@@ -188,9 +188,11 @@ static uint my_casedn_ucs2(CHARSET_INFO *cs, char *src, uint srclen,
return srclen;
}
-static void my_casedn_str_ucs2(CHARSET_INFO *cs __attribute__((unused)),
+
+static uint my_casedn_str_ucs2(CHARSET_INFO *cs __attribute__((unused)),
char * s __attribute__((unused)))
{
+ return 0;
}
diff --git a/strings/ctype-utf8.c b/strings/ctype-utf8.c
index 6c3ceaf868b..8a4ed48bef5 100644
--- a/strings/ctype-utf8.c
+++ b/strings/ctype-utf8.c
@@ -2047,6 +2047,52 @@ static int my_utf8_uni(CHARSET_INFO *cs __attribute__((unused)),
return MY_CS_ILSEQ;
}
+
+/*
+ The same as above, but without range check
+ for example, for a null-terminated string
+*/
+static int my_utf8_uni_no_range(CHARSET_INFO *cs __attribute__((unused)),
+ my_wc_t * pwc, const uchar *s)
+{
+ unsigned char c;
+
+ c= s[0];
+ if (c < 0x80)
+ {
+ *pwc = c;
+ return 1;
+ }
+
+ if (c < 0xc2)
+ return MY_CS_ILSEQ;
+
+ if (c < 0xe0)
+ {
+ if (!((s[1] ^ 0x80) < 0x40))
+ return MY_CS_ILSEQ;
+
+ *pwc = ((my_wc_t) (c & 0x1f) << 6) | (my_wc_t) (s[1] ^ 0x80);
+ return 2;
+ }
+
+ if (c < 0xf0)
+ {
+ if (!((s[1] ^ 0x80) < 0x40 &&
+ (s[2] ^ 0x80) < 0x40 &&
+ (c >= 0xe1 || s[1] >= 0xa0)))
+ return MY_CS_ILSEQ;
+
+ *pwc= ((my_wc_t) (c & 0x0f) << 12) |
+ ((my_wc_t) (s[1] ^ 0x80) << 6) |
+ (my_wc_t) (s[2] ^ 0x80);
+
+ return 3;
+ }
+ return MY_CS_ILSEQ;
+}
+
+
static int my_uni_utf8 (CHARSET_INFO *cs __attribute__((unused)) ,
my_wc_t wc, uchar *r, uchar *e)
{
@@ -2093,6 +2139,34 @@ static int my_uni_utf8 (CHARSET_INFO *cs __attribute__((unused)) ,
}
+/*
+ The same as above, but without range check.
+*/
+static int my_uni_utf8_no_range(CHARSET_INFO *cs __attribute__((unused)),
+ my_wc_t wc, uchar *r)
+{
+ int count;
+
+ if (wc < 0x80)
+ count= 1;
+ else if (wc < 0x800)
+ count= 2;
+ else if (wc < 0x10000)
+ count= 3;
+ else
+ return MY_CS_ILUNI;
+
+ switch (count)
+ {
+ /* Fall through all cases!!! */
+ case 3: r[2]= (uchar) (0x80 | (wc & 0x3f)); wc= wc >> 6; wc |= 0x800;
+ case 2: r[1]= (uchar) (0x80 | (wc & 0x3f)); wc= wc >> 6; wc |= 0xc0;
+ case 1: r[0]= (uchar) wc;
+ }
+ return count;
+}
+
+
static uint my_caseup_utf8(CHARSET_INFO *cs, char *src, uint srclen,
char *dst, uint dstlen)
{
@@ -2143,10 +2217,26 @@ static void my_hash_sort_utf8(CHARSET_INFO *cs, const uchar *s, uint slen,
}
-static void my_caseup_str_utf8(CHARSET_INFO * cs, char * s)
+static uint my_caseup_str_utf8(CHARSET_INFO *cs, char *src)
{
- uint len= (uint) strlen(s);
- my_caseup_utf8(cs, s, len, s, len);
+ my_wc_t wc;
+ int srcres, dstres;
+ char *dst= src, *dst0= src;
+ MY_UNICASE_INFO **uni_plane= cs->caseinfo;
+ DBUG_ASSERT(cs->caseup_multiply == 1);
+
+ while (*src &&
+ (srcres= my_utf8_uni_no_range(cs, &wc, (uchar *) src)) > 0)
+ {
+ int plane= (wc>>8) & 0xFF;
+ wc= uni_plane[plane] ? uni_plane[plane][wc & 0xFF].toupper : wc;
+ if ((dstres= my_uni_utf8_no_range(cs, wc, (uchar*) dst)) <= 0)
+ break;
+ src+= srcres;
+ dst+= dstres;
+ }
+ *dst= '\0';
+ return (uint) (dst - dst0);
}
@@ -2172,10 +2262,43 @@ static uint my_casedn_utf8(CHARSET_INFO *cs, char *src, uint srclen,
return (uint) (dst - dst0);
}
-static void my_casedn_str_utf8(CHARSET_INFO *cs, char * s)
+
+static uint my_casedn_str_utf8(CHARSET_INFO *cs, char *src)
{
- uint len= (uint) strlen(s);
- my_casedn_utf8(cs, s, len, s, len);
+ my_wc_t wc;
+ int srcres, dstres;
+ char *dst= src, *dst0= src;
+ MY_UNICASE_INFO **uni_plane= cs->caseinfo;
+ DBUG_ASSERT(cs->casedn_multiply == 1);
+
+ while (*src &&
+ (srcres= my_utf8_uni_no_range(cs, &wc, (uchar *) src)) > 0)
+ {
+ int plane= (wc>>8) & 0xFF;
+ wc= uni_plane[plane] ? uni_plane[plane][wc & 0xFF].tolower : wc;
+ if ((dstres= my_uni_utf8_no_range(cs, wc, (uchar*) dst)) <= 0)
+ break;
+ src+= srcres;
+ dst+= dstres;
+ }
+
+ /*
+ In rare cases lower string can be shorter than
+ the original string, for example:
+
+ "U+0130 LATIN CAPITAL LETTER I WITH DOT ABOVE"
+ (which is 0xC4B0 in utf8, i.e. two bytes)
+
+ is converted into
+
+ "U+0069 LATIN SMALL LETTER I"
+ (which is 0x69 in utf8, i.e. one byte)
+
+ So, we need to put '\0' terminator after converting.
+ */
+
+ *dst= '\0';
+ return (uint) (dst - dst0);
}
@@ -4051,6 +4174,7 @@ static MY_CHARSET_HANDLER my_charset_filename_handler=
my_strntoull_8bit,
my_strntod_8bit,
my_strtoll10_8bit,
+ my_strntoull10rnd_8bit,
my_scan_8bit
};