| 41 |
/* This module compiles code for supporting the use of Unicode character |
/* This module compiles code for supporting the use of Unicode character |
| 42 |
properties. We use the (embryonic at the time of writing) UCP library, by |
properties. We use the (embryonic at the time of writing) UCP library, by |
| 43 |
including some of its files, copies of which have been put in the PCRE |
including some of its files, copies of which have been put in the PCRE |
| 44 |
distribution. There is a macro in pcre_internal.h that changes the name |
distribution. The actual search function is reproduced here, with its name |
| 45 |
ucp_findchar into _pcre_ucp_findchar. */ |
changed. */ |
| 46 |
|
|
| 47 |
|
|
| 48 |
#include "pcre_internal.h" |
#include "pcre_internal.h" |
| 49 |
|
|
| 50 |
#include "ucp_findchar.c" |
#include "ucp.h" /* Category definitions */ |
| 51 |
|
#include "ucpinternal.h" /* Internal table details */ |
| 52 |
|
#include "ucptable.c" /* The table itself */ |
| 53 |
|
|
| 54 |
|
|
| 55 |
|
|
| 56 |
|
/************************************************* |
| 57 |
|
* Search table and return data * |
| 58 |
|
*************************************************/ |
| 59 |
|
|
| 60 |
|
/* Two values are returned: the category is ucp_C, ucp_L, etc. The detailed |
| 61 |
|
character type is ucp_Lu, ucp_Nd, etc. |
| 62 |
|
|
| 63 |
|
Arguments: |
| 64 |
|
c the character value |
| 65 |
|
type_ptr the detailed character type is returned here |
| 66 |
|
case_ptr for letters, the opposite case is returned here, if there |
| 67 |
|
is one, else zero |
| 68 |
|
|
| 69 |
|
Returns: the character type category or -1 if not found |
| 70 |
|
*/ |
| 71 |
|
|
| 72 |
|
PCRE_EXPORT int |
| 73 |
|
_pcre_ucp_findchar(const int c, int *type_ptr, int *case_ptr) |
| 74 |
|
{ |
| 75 |
|
cnode *node = ucp_table; |
| 76 |
|
register int cc = c; |
| 77 |
|
int case_offset; |
| 78 |
|
|
| 79 |
|
for (;;) |
| 80 |
|
{ |
| 81 |
|
register int d = node->f1 | ((node->f0 & f0_chhmask) << 16); |
| 82 |
|
if (cc == d) break; |
| 83 |
|
if (cc < d) |
| 84 |
|
{ |
| 85 |
|
if ((node->f0 & f0_leftexists) == 0) return -1; |
| 86 |
|
node ++; |
| 87 |
|
} |
| 88 |
|
else |
| 89 |
|
{ |
| 90 |
|
register int roffset = (node->f2 & f2_rightmask) >> f2_rightshift; |
| 91 |
|
if (roffset == 0) return -1; |
| 92 |
|
node += 1 << (roffset - 1); |
| 93 |
|
} |
| 94 |
|
} |
| 95 |
|
|
| 96 |
|
switch ((*type_ptr = ((node->f0 & f0_typemask) >> f0_typeshift))) |
| 97 |
|
{ |
| 98 |
|
case ucp_Cc: |
| 99 |
|
case ucp_Cf: |
| 100 |
|
case ucp_Cn: |
| 101 |
|
case ucp_Co: |
| 102 |
|
case ucp_Cs: |
| 103 |
|
return ucp_C; |
| 104 |
|
break; |
| 105 |
|
|
| 106 |
|
case ucp_Ll: |
| 107 |
|
case ucp_Lu: |
| 108 |
|
case_offset = node->f2 & f2_casemask; |
| 109 |
|
if ((case_offset & 0x0100) != 0) case_offset |= 0xfffff000; |
| 110 |
|
*case_ptr = (case_offset == 0)? 0 : cc + case_offset; |
| 111 |
|
return ucp_L; |
| 112 |
|
|
| 113 |
|
case ucp_Lm: |
| 114 |
|
case ucp_Lo: |
| 115 |
|
case ucp_Lt: |
| 116 |
|
*case_ptr = 0; |
| 117 |
|
return ucp_L; |
| 118 |
|
break; |
| 119 |
|
|
| 120 |
|
case ucp_Mc: |
| 121 |
|
case ucp_Me: |
| 122 |
|
case ucp_Mn: |
| 123 |
|
return ucp_M; |
| 124 |
|
break; |
| 125 |
|
|
| 126 |
|
case ucp_Nd: |
| 127 |
|
case ucp_Nl: |
| 128 |
|
case ucp_No: |
| 129 |
|
return ucp_N; |
| 130 |
|
break; |
| 131 |
|
|
| 132 |
|
case ucp_Pc: |
| 133 |
|
case ucp_Pd: |
| 134 |
|
case ucp_Pe: |
| 135 |
|
case ucp_Pf: |
| 136 |
|
case ucp_Pi: |
| 137 |
|
case ucp_Ps: |
| 138 |
|
case ucp_Po: |
| 139 |
|
return ucp_P; |
| 140 |
|
break; |
| 141 |
|
|
| 142 |
|
case ucp_Sc: |
| 143 |
|
case ucp_Sk: |
| 144 |
|
case ucp_Sm: |
| 145 |
|
case ucp_So: |
| 146 |
|
return ucp_S; |
| 147 |
|
break; |
| 148 |
|
|
| 149 |
|
case ucp_Zl: |
| 150 |
|
case ucp_Zp: |
| 151 |
|
case ucp_Zs: |
| 152 |
|
return ucp_Z; |
| 153 |
|
break; |
| 154 |
|
|
| 155 |
|
default: /* "Should never happen" */ |
| 156 |
|
return -1; |
| 157 |
|
break; |
| 158 |
|
} |
| 159 |
|
} |
| 160 |
|
|
| 161 |
/* End of pcre_ucp_findchar.c */ |
/* End of pcre_ucp_findchar.c */ |