1
by brian
clean slate |
1 |
/* Copyright (C) 2003 MySQL AB
|
2 |
||
3 |
This program is free software; you can redistribute it and/or modify
|
|
4 |
it under the terms of the GNU General Public License as published by
|
|
5 |
the Free Software Foundation; version 2 of the License.
|
|
6 |
||
7 |
This program is distributed in the hope that it will be useful,
|
|
8 |
but WITHOUT ANY WARRANTY; without even the implied warranty of
|
|
9 |
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
|
|
10 |
GNU General Public License for more details.
|
|
11 |
||
12 |
You should have received a copy of the GNU General Public License
|
|
13 |
along with this program; if not, write to the Free Software
|
|
14 |
Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA */
|
|
15 |
||
16 |
/* Some useful string utility functions used by the MySQL server */
|
|
243.1.17
by Jay Pipes
FINAL PHASE removal of mysql_priv.h (Bye, bye my friend.) |
17 |
#include <drizzled/server_includes.h> |
1
by brian
clean slate |
18 |
|
19 |
/*
|
|
20 |
Return bitmap for strings used in a set
|
|
21 |
||
22 |
SYNOPSIS
|
|
23 |
find_set()
|
|
24 |
lib Strings in set
|
|
25 |
str Strings of set-strings separated by ','
|
|
26 |
err_pos If error, set to point to start of wrong set string
|
|
27 |
err_len If error, set to the length of wrong set string
|
|
28 |
set_warning Set to 1 if some string in set couldn't be used
|
|
29 |
||
30 |
NOTE
|
|
31 |
We delete all end space from str before comparison
|
|
32 |
||
33 |
RETURN
|
|
34 |
bitmap of all sets found in x.
|
|
35 |
set_warning is set to 1 if there was any sets that couldn't be set
|
|
36 |
*/
|
|
37 |
||
38 |
static const char field_separator=','; |
|
39 |
||
482
by Brian Aker
Remove uint. |
40 |
uint64_t find_set(TYPELIB *lib, const char *str, uint32_t length, |
264.2.6
by Andrey Hristov
Constify the usage of CHARSET_INFO almost to the last place in the code. |
41 |
const CHARSET_INFO * const cs, |
482
by Brian Aker
Remove uint. |
42 |
char **err_pos, uint32_t *err_len, bool *set_warning) |
1
by brian
clean slate |
43 |
{
|
383.1.12
by Brian Aker
Much closer toward UTF8 being around all the time... |
44 |
const CHARSET_INFO * const strip= cs ? cs : &my_charset_utf8_general_ci; |
1
by brian
clean slate |
45 |
const char *end= str + strip->cset->lengthsp(strip, str, length); |
151
by Brian Aker
Ulonglong to uint64_t |
46 |
uint64_t found= 0; |
1
by brian
clean slate |
47 |
*err_pos= 0; // No error yet |
48 |
if (str != end) |
|
49 |
{
|
|
660.1.3
by Eric Herman
removed trailing whitespace with simple script: |
50 |
const char *start= str; |
1
by brian
clean slate |
51 |
for (;;) |
52 |
{
|
|
53 |
const char *pos= start; |
|
482
by Brian Aker
Remove uint. |
54 |
uint32_t var_len; |
1
by brian
clean slate |
55 |
int mblen= 1; |
56 |
||
1054.2.12
by Monty Taylor
First pass at removing strconvert. |
57 |
for (; pos != end && *pos != field_separator; pos++) |
58 |
{}
|
|
895
by Brian Aker
Completion (?) of uint conversion. |
59 |
var_len= (uint32_t) (pos - start); |
482
by Brian Aker
Remove uint. |
60 |
uint32_t find= cs ? find_type2(lib, start, var_len, cs) : |
1
by brian
clean slate |
61 |
find_type(lib, start, var_len, (bool) 0); |
62 |
if (!find) |
|
63 |
{
|
|
64 |
*err_pos= (char*) start; |
|
65 |
*err_len= var_len; |
|
66 |
*set_warning= 1; |
|
67 |
}
|
|
68 |
else
|
|
152
by Brian Aker
longlong replacement |
69 |
found|= ((int64_t) 1 << (find - 1)); |
1
by brian
clean slate |
70 |
if (pos >= end) |
71 |
break; |
|
72 |
start= pos + mblen; |
|
73 |
}
|
|
74 |
}
|
|
75 |
return found; |
|
76 |
}
|
|
77 |
||
78 |
||
79 |
/*
|
|
80 |
Function to find a string in a TYPELIB
|
|
81 |
(Same format as mysys/typelib.c)
|
|
82 |
||
83 |
SYNOPSIS
|
|
84 |
find_type()
|
|
85 |
lib TYPELIB (struct of pointer to values + count)
|
|
86 |
find String to find
|
|
87 |
length Length of string to find
|
|
88 |
part_match Allow part matching of value
|
|
89 |
||
90 |
RETURN
|
|
91 |
0 error
|
|
92 |
> 0 position in TYPELIB->type_names +1
|
|
93 |
*/
|
|
94 |
||
482
by Brian Aker
Remove uint. |
95 |
uint32_t find_type(const TYPELIB *lib, const char *find, uint32_t length, |
1
by brian
clean slate |
96 |
bool part_match) |
97 |
{
|
|
482
by Brian Aker
Remove uint. |
98 |
uint32_t found_count=0, found_pos=0; |
1
by brian
clean slate |
99 |
const char *end= find+length; |
100 |
const char *i; |
|
101 |
const char *j; |
|
482
by Brian Aker
Remove uint. |
102 |
for (uint32_t pos=0 ; (j=lib->type_names[pos++]) ; ) |
1
by brian
clean slate |
103 |
{
|
660.1.3
by Eric Herman
removed trailing whitespace with simple script: |
104 |
for (i=find ; i != end && |
105 |
my_toupper(system_charset_info,*i) == |
|
1
by brian
clean slate |
106 |
my_toupper(system_charset_info,*j) ; i++, j++) ; |
107 |
if (i == end) |
|
108 |
{
|
|
109 |
if (! *j) |
|
110 |
return(pos); |
|
111 |
found_count++; |
|
112 |
found_pos= pos; |
|
113 |
}
|
|
114 |
}
|
|
115 |
return(found_count == 1 && part_match ? found_pos : 0); |
|
116 |
}
|
|
117 |
||
118 |
||
119 |
/*
|
|
120 |
Find a string in a list of strings according to collation
|
|
121 |
||
122 |
SYNOPSIS
|
|
123 |
find_type2()
|
|
124 |
lib TYPELIB (struct of pointer to values + count)
|
|
125 |
x String to find
|
|
126 |
length String length
|
|
127 |
cs Character set + collation to use for comparison
|
|
128 |
||
129 |
NOTES
|
|
130 |
||
131 |
RETURN
|
|
132 |
0 No matching value
|
|
133 |
>0 Offset+1 in typelib for matched string
|
|
134 |
*/
|
|
135 |
||
482
by Brian Aker
Remove uint. |
136 |
uint32_t find_type2(const TYPELIB *typelib, const char *x, uint32_t length, |
264.2.6
by Andrey Hristov
Constify the usage of CHARSET_INFO almost to the last place in the code. |
137 |
const CHARSET_INFO * const cs) |
1
by brian
clean slate |
138 |
{
|
139 |
int pos; |
|
140 |
const char *j; |
|
141 |
||
142 |
if (!typelib->count) |
|
143 |
{
|
|
51.1.69
by Jay Pipes
Removed/replaced DBUG symbols |
144 |
return(0); |
1
by brian
clean slate |
145 |
}
|
146 |
||
147 |
for (pos=0 ; (j=typelib->type_names[pos]) ; pos++) |
|
148 |
{
|
|
481
by Brian Aker
Remove all of uchar. |
149 |
if (!my_strnncoll(cs, (const unsigned char*) x, length, |
150 |
(const unsigned char*) j, typelib->type_lengths[pos])) |
|
51.1.69
by Jay Pipes
Removed/replaced DBUG symbols |
151 |
return(pos+1); |
1
by brian
clean slate |
152 |
}
|
51.1.69
by Jay Pipes
Removed/replaced DBUG symbols |
153 |
return(0); |
1
by brian
clean slate |
154 |
} /* find_type */ |
155 |
||
156 |
||
157 |
/*
|
|
158 |
Un-hex all elements in a typelib
|
|
159 |
||
160 |
SYNOPSIS
|
|
161 |
unhex_type2()
|
|
162 |
interval TYPELIB (struct of pointer to values + lengths + count)
|
|
163 |
||
164 |
NOTES
|
|
165 |
||
166 |
RETURN
|
|
167 |
N/A
|
|
168 |
*/
|
|
169 |
||
170 |
void unhex_type2(TYPELIB *interval) |
|
171 |
{
|
|
482
by Brian Aker
Remove uint. |
172 |
for (uint32_t pos= 0; pos < interval->count; pos++) |
1
by brian
clean slate |
173 |
{
|
174 |
char *from, *to; |
|
175 |
for (from= to= (char*) interval->type_names[pos]; *from; ) |
|
176 |
{
|
|
177 |
/*
|
|
178 |
Note, hexchar_to_int(*from++) doesn't work
|
|
179 |
one some compilers, e.g. IRIX. Looks like a compiler
|
|
180 |
bug in inline functions in combination with arguments
|
|
181 |
that have a side effect. So, let's use from[0] and from[1]
|
|
182 |
and increment 'from' by two later.
|
|
183 |
*/
|
|
184 |
||
185 |
*to++= (char) (hexchar_to_int(from[0]) << 4) + |
|
186 |
hexchar_to_int(from[1]); |
|
187 |
from+= 2; |
|
188 |
}
|
|
189 |
interval->type_lengths[pos] /= 2; |
|
190 |
}
|
|
191 |
}
|
|
192 |
||
193 |
||
194 |
/*
|
|
195 |
Check if the first word in a string is one of the ones in TYPELIB
|
|
196 |
||
197 |
SYNOPSIS
|
|
198 |
check_word()
|
|
199 |
lib TYPELIB
|
|
200 |
val String to check
|
|
201 |
end End of input
|
|
202 |
end_of_word Store value of last used byte here if we found word
|
|
203 |
||
204 |
RETURN
|
|
205 |
0 No matching value
|
|
206 |
> 1 lib->type_names[#-1] matched
|
|
207 |
end_of_word will point to separator character/end in 'val'
|
|
208 |
*/
|
|
209 |
||
482
by Brian Aker
Remove uint. |
210 |
uint32_t check_word(TYPELIB *lib, const char *val, const char *end, |
1
by brian
clean slate |
211 |
const char **end_of_word) |
212 |
{
|
|
213 |
int res; |
|
214 |
const char *ptr; |
|
215 |
||
216 |
/* Fiend end of word */
|
|
383.1.12
by Brian Aker
Much closer toward UTF8 being around all the time... |
217 |
for (ptr= val ; ptr < end && my_isalpha(&my_charset_utf8_general_ci, *ptr) ; ptr++) |
1
by brian
clean slate |
218 |
;
|
895
by Brian Aker
Completion (?) of uint conversion. |
219 |
if ((res=find_type(lib, val, (uint32_t) (ptr - val), 1)) > 0) |
1
by brian
clean slate |
220 |
*end_of_word= ptr; |
221 |
return res; |
|
222 |
}
|
|
223 |
||
224 |
||
225 |
/*
|
|
226 |
Searches for a LEX_STRING in an LEX_STRING array.
|
|
227 |
||
228 |
SYNOPSIS
|
|
229 |
find_string_in_array()
|
|
230 |
heap The array
|
|
231 |
needle The string to search for
|
|
232 |
||
233 |
NOTE
|
|
234 |
The last LEX_STRING in the array should have str member set to NULL
|
|
235 |
||
236 |
RETURN VALUES
|
|
237 |
-1 Not found
|
|
238 |
>=0 Ordinal position
|
|
239 |
*/
|
|
240 |
||
241 |
int find_string_in_array(LEX_STRING * const haystack, LEX_STRING * const needle, |
|
264.2.6
by Andrey Hristov
Constify the usage of CHARSET_INFO almost to the last place in the code. |
242 |
const CHARSET_INFO * const cs) |
1
by brian
clean slate |
243 |
{
|
244 |
const LEX_STRING *pos; |
|
245 |
for (pos= haystack; pos->str; pos++) |
|
481
by Brian Aker
Remove all of uchar. |
246 |
if (!cs->coll->strnncollsp(cs, (unsigned char *) pos->str, pos->length, |
247 |
(unsigned char *) needle->str, needle->length, 0)) |
|
1
by brian
clean slate |
248 |
{
|
249 |
return (pos - haystack); |
|
250 |
}
|
|
251 |
return -1; |
|
252 |
}
|