1 /*
2 Unicode character type helpers.
3
4 Written by Marc-Andre Lemburg (mal@lemburg.com).
5 Modified for Python 2.0 by Fredrik Lundh (fredrik@pythonware.com)
6
7 Copyright (c) Corporation for National Research Initiatives.
8
9 */
10
11 #include "Python.h"
12 #include "unicodeobject.h"
13
14 #define ALPHA_MASK 0x01
15 #define DECIMAL_MASK 0x02
16 #define DIGIT_MASK 0x04
17 #define LOWER_MASK 0x08
18 #define LINEBREAK_MASK 0x10
19 #define SPACE_MASK 0x20
20 #define TITLE_MASK 0x40
21 #define UPPER_MASK 0x80
22 #define NODELTA_MASK 0x100
23 #define NUMERIC_MASK 0x200
24
25 typedef struct {
26 const Py_UNICODE upper;
27 const Py_UNICODE lower;
28 const Py_UNICODE title;
29 const unsigned char decimal;
30 const unsigned char digit;
31 const unsigned short flags;
32 } _PyUnicode_TypeRecord;
33
34 #include "unicodetype_db.h"
35
36 static const _PyUnicode_TypeRecord *
gettyperecord(Py_UNICODE code)37 gettyperecord(Py_UNICODE code)
38 {
39 int index;
40
41 #ifdef Py_UNICODE_WIDE
42 if (code >= 0x110000)
43 index = 0;
44 else
45 #endif
46 {
47 index = index1[(code>>SHIFT)];
48 index = index2[(index<<SHIFT)+(code&((1<<SHIFT)-1))];
49 }
50
51 return &_PyUnicode_TypeRecords[index];
52 }
53
54 /* Returns the titlecase Unicode characters corresponding to ch or just
55 ch if no titlecase mapping is known. */
56
_PyUnicode_ToTitlecase(register Py_UNICODE ch)57 Py_UNICODE _PyUnicode_ToTitlecase(register Py_UNICODE ch)
58 {
59 const _PyUnicode_TypeRecord *ctype = gettyperecord(ch);
60 int delta = ctype->title;
61
62 if (ctype->flags & NODELTA_MASK)
63 return delta;
64
65 if (delta >= 32768)
66 delta -= 65536;
67
68 return ch + delta;
69 }
70
71 /* Returns 1 for Unicode characters having the category 'Lt', 0
72 otherwise. */
73
_PyUnicode_IsTitlecase(Py_UNICODE ch)74 int _PyUnicode_IsTitlecase(Py_UNICODE ch)
75 {
76 const _PyUnicode_TypeRecord *ctype = gettyperecord(ch);
77
78 return (ctype->flags & TITLE_MASK) != 0;
79 }
80
81 /* Returns the integer decimal (0-9) for Unicode characters having
82 this property, -1 otherwise. */
83
_PyUnicode_ToDecimalDigit(Py_UNICODE ch)84 int _PyUnicode_ToDecimalDigit(Py_UNICODE ch)
85 {
86 const _PyUnicode_TypeRecord *ctype = gettyperecord(ch);
87
88 return (ctype->flags & DECIMAL_MASK) ? ctype->decimal : -1;
89 }
90
_PyUnicode_IsDecimalDigit(Py_UNICODE ch)91 int _PyUnicode_IsDecimalDigit(Py_UNICODE ch)
92 {
93 if (_PyUnicode_ToDecimalDigit(ch) < 0)
94 return 0;
95 return 1;
96 }
97
98 /* Returns the integer digit (0-9) for Unicode characters having
99 this property, -1 otherwise. */
100
_PyUnicode_ToDigit(Py_UNICODE ch)101 int _PyUnicode_ToDigit(Py_UNICODE ch)
102 {
103 const _PyUnicode_TypeRecord *ctype = gettyperecord(ch);
104
105 return (ctype->flags & DIGIT_MASK) ? ctype->digit : -1;
106 }
107
_PyUnicode_IsDigit(Py_UNICODE ch)108 int _PyUnicode_IsDigit(Py_UNICODE ch)
109 {
110 if (_PyUnicode_ToDigit(ch) < 0)
111 return 0;
112 return 1;
113 }
114
115 /* Returns the numeric value as double for Unicode characters having
116 this property, -1.0 otherwise. */
117
_PyUnicode_IsNumeric(Py_UNICODE ch)118 int _PyUnicode_IsNumeric(Py_UNICODE ch)
119 {
120 const _PyUnicode_TypeRecord *ctype = gettyperecord(ch);
121
122 return (ctype->flags & NUMERIC_MASK) != 0;
123 }
124
125 #ifndef WANT_WCTYPE_FUNCTIONS
126
127 /* Returns 1 for Unicode characters having the category 'Ll', 0
128 otherwise. */
129
_PyUnicode_IsLowercase(Py_UNICODE ch)130 int _PyUnicode_IsLowercase(Py_UNICODE ch)
131 {
132 const _PyUnicode_TypeRecord *ctype = gettyperecord(ch);
133
134 return (ctype->flags & LOWER_MASK) != 0;
135 }
136
137 /* Returns 1 for Unicode characters having the category 'Lu', 0
138 otherwise. */
139
_PyUnicode_IsUppercase(Py_UNICODE ch)140 int _PyUnicode_IsUppercase(Py_UNICODE ch)
141 {
142 const _PyUnicode_TypeRecord *ctype = gettyperecord(ch);
143
144 return (ctype->flags & UPPER_MASK) != 0;
145 }
146
147 /* Returns the uppercase Unicode characters corresponding to ch or just
148 ch if no uppercase mapping is known. */
149
_PyUnicode_ToUppercase(Py_UNICODE ch)150 Py_UNICODE _PyUnicode_ToUppercase(Py_UNICODE ch)
151 {
152 const _PyUnicode_TypeRecord *ctype = gettyperecord(ch);
153 int delta = ctype->upper;
154 if (ctype->flags & NODELTA_MASK)
155 return delta;
156 if (delta >= 32768)
157 delta -= 65536;
158 return ch + delta;
159 }
160
161 /* Returns the lowercase Unicode characters corresponding to ch or just
162 ch if no lowercase mapping is known. */
163
_PyUnicode_ToLowercase(Py_UNICODE ch)164 Py_UNICODE _PyUnicode_ToLowercase(Py_UNICODE ch)
165 {
166 const _PyUnicode_TypeRecord *ctype = gettyperecord(ch);
167 int delta = ctype->lower;
168 if (ctype->flags & NODELTA_MASK)
169 return delta;
170 if (delta >= 32768)
171 delta -= 65536;
172 return ch + delta;
173 }
174
175 /* Returns 1 for Unicode characters having the category 'Ll', 'Lu', 'Lt',
176 'Lo' or 'Lm', 0 otherwise. */
177
_PyUnicode_IsAlpha(Py_UNICODE ch)178 int _PyUnicode_IsAlpha(Py_UNICODE ch)
179 {
180 const _PyUnicode_TypeRecord *ctype = gettyperecord(ch);
181
182 return (ctype->flags & ALPHA_MASK) != 0;
183 }
184
185 #else
186
187 /* Export the interfaces using the wchar_t type for portability
188 reasons: */
189
_PyUnicode_IsLowercase(Py_UNICODE ch)190 int _PyUnicode_IsLowercase(Py_UNICODE ch)
191 {
192 return iswlower(ch);
193 }
194
_PyUnicode_IsUppercase(Py_UNICODE ch)195 int _PyUnicode_IsUppercase(Py_UNICODE ch)
196 {
197 return iswupper(ch);
198 }
199
_PyUnicode_ToLowercase(Py_UNICODE ch)200 Py_UNICODE _PyUnicode_ToLowercase(Py_UNICODE ch)
201 {
202 return towlower(ch);
203 }
204
_PyUnicode_ToUppercase(Py_UNICODE ch)205 Py_UNICODE _PyUnicode_ToUppercase(Py_UNICODE ch)
206 {
207 return towupper(ch);
208 }
209
_PyUnicode_IsAlpha(Py_UNICODE ch)210 int _PyUnicode_IsAlpha(Py_UNICODE ch)
211 {
212 return iswalpha(ch);
213 }
214
215 #endif
216