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
13 #define ALPHA_MASK 0x01
14 #define DECIMAL_MASK 0x02
15 #define DIGIT_MASK 0x04
16 #define LOWER_MASK 0x08
17 #define LINEBREAK_MASK 0x10
18 #define SPACE_MASK 0x20
19 #define TITLE_MASK 0x40
20 #define UPPER_MASK 0x80
21 #define XID_START_MASK 0x100
22 #define XID_CONTINUE_MASK 0x200
23 #define PRINTABLE_MASK 0x400
24 #define NUMERIC_MASK 0x800
25 #define CASE_IGNORABLE_MASK 0x1000
26 #define CASED_MASK 0x2000
27 #define EXTENDED_CASE_MASK 0x4000
28
29 typedef struct {
30 /*
31 These are either deltas to the character or offsets in
32 _PyUnicode_ExtendedCase.
33 */
34 const int upper;
35 const int lower;
36 const int title;
37 /* Note if more flag space is needed, decimal and digit could be unified. */
38 const unsigned char decimal;
39 const unsigned char digit;
40 const unsigned short flags;
41 } _PyUnicode_TypeRecord;
42
43 #include "unicodetype_db.h"
44
45 static const _PyUnicode_TypeRecord *
gettyperecord(Py_UCS4 code)46 gettyperecord(Py_UCS4 code)
47 {
48 int index;
49
50 if (code >= 0x110000)
51 index = 0;
52 else
53 {
54 index = index1[(code>>SHIFT)];
55 index = index2[(index<<SHIFT)+(code&((1<<SHIFT)-1))];
56 }
57
58 return &_PyUnicode_TypeRecords[index];
59 }
60
61 /* Returns the titlecase Unicode characters corresponding to ch or just
62 ch if no titlecase mapping is known. */
63
_PyUnicode_ToTitlecase(Py_UCS4 ch)64 Py_UCS4 _PyUnicode_ToTitlecase(Py_UCS4 ch)
65 {
66 const _PyUnicode_TypeRecord *ctype = gettyperecord(ch);
67
68 if (ctype->flags & EXTENDED_CASE_MASK)
69 return _PyUnicode_ExtendedCase[ctype->title & 0xFFFF];
70 return ch + ctype->title;
71 }
72
73 /* Returns 1 for Unicode characters having the category 'Lt', 0
74 otherwise. */
75
_PyUnicode_IsTitlecase(Py_UCS4 ch)76 int _PyUnicode_IsTitlecase(Py_UCS4 ch)
77 {
78 const _PyUnicode_TypeRecord *ctype = gettyperecord(ch);
79
80 return (ctype->flags & TITLE_MASK) != 0;
81 }
82
83 /* Returns 1 for Unicode characters having the XID_Start property, 0
84 otherwise. */
85
_PyUnicode_IsXidStart(Py_UCS4 ch)86 int _PyUnicode_IsXidStart(Py_UCS4 ch)
87 {
88 const _PyUnicode_TypeRecord *ctype = gettyperecord(ch);
89
90 return (ctype->flags & XID_START_MASK) != 0;
91 }
92
93 /* Returns 1 for Unicode characters having the XID_Continue property,
94 0 otherwise. */
95
_PyUnicode_IsXidContinue(Py_UCS4 ch)96 int _PyUnicode_IsXidContinue(Py_UCS4 ch)
97 {
98 const _PyUnicode_TypeRecord *ctype = gettyperecord(ch);
99
100 return (ctype->flags & XID_CONTINUE_MASK) != 0;
101 }
102
103 /* Returns the integer decimal (0-9) for Unicode characters having
104 this property, -1 otherwise. */
105
_PyUnicode_ToDecimalDigit(Py_UCS4 ch)106 int _PyUnicode_ToDecimalDigit(Py_UCS4 ch)
107 {
108 const _PyUnicode_TypeRecord *ctype = gettyperecord(ch);
109
110 return (ctype->flags & DECIMAL_MASK) ? ctype->decimal : -1;
111 }
112
_PyUnicode_IsDecimalDigit(Py_UCS4 ch)113 int _PyUnicode_IsDecimalDigit(Py_UCS4 ch)
114 {
115 if (_PyUnicode_ToDecimalDigit(ch) < 0)
116 return 0;
117 return 1;
118 }
119
120 /* Returns the integer digit (0-9) for Unicode characters having
121 this property, -1 otherwise. */
122
_PyUnicode_ToDigit(Py_UCS4 ch)123 int _PyUnicode_ToDigit(Py_UCS4 ch)
124 {
125 const _PyUnicode_TypeRecord *ctype = gettyperecord(ch);
126
127 return (ctype->flags & DIGIT_MASK) ? ctype->digit : -1;
128 }
129
_PyUnicode_IsDigit(Py_UCS4 ch)130 int _PyUnicode_IsDigit(Py_UCS4 ch)
131 {
132 if (_PyUnicode_ToDigit(ch) < 0)
133 return 0;
134 return 1;
135 }
136
137 /* Returns the numeric value as double for Unicode characters having
138 this property, -1.0 otherwise. */
139
_PyUnicode_IsNumeric(Py_UCS4 ch)140 int _PyUnicode_IsNumeric(Py_UCS4 ch)
141 {
142 const _PyUnicode_TypeRecord *ctype = gettyperecord(ch);
143
144 return (ctype->flags & NUMERIC_MASK) != 0;
145 }
146
147 /* Returns 1 for Unicode characters to be hex-escaped when repr()ed,
148 0 otherwise.
149 All characters except those characters defined in the Unicode character
150 database as following categories are considered printable.
151 * Cc (Other, Control)
152 * Cf (Other, Format)
153 * Cs (Other, Surrogate)
154 * Co (Other, Private Use)
155 * Cn (Other, Not Assigned)
156 * Zl Separator, Line ('\u2028', LINE SEPARATOR)
157 * Zp Separator, Paragraph ('\u2029', PARAGRAPH SEPARATOR)
158 * Zs (Separator, Space) other than ASCII space('\x20').
159 */
_PyUnicode_IsPrintable(Py_UCS4 ch)160 int _PyUnicode_IsPrintable(Py_UCS4 ch)
161 {
162 const _PyUnicode_TypeRecord *ctype = gettyperecord(ch);
163
164 return (ctype->flags & PRINTABLE_MASK) != 0;
165 }
166
167 /* Returns 1 for Unicode characters having the category 'Ll', 0
168 otherwise. */
169
_PyUnicode_IsLowercase(Py_UCS4 ch)170 int _PyUnicode_IsLowercase(Py_UCS4 ch)
171 {
172 const _PyUnicode_TypeRecord *ctype = gettyperecord(ch);
173
174 return (ctype->flags & LOWER_MASK) != 0;
175 }
176
177 /* Returns 1 for Unicode characters having the category 'Lu', 0
178 otherwise. */
179
_PyUnicode_IsUppercase(Py_UCS4 ch)180 int _PyUnicode_IsUppercase(Py_UCS4 ch)
181 {
182 const _PyUnicode_TypeRecord *ctype = gettyperecord(ch);
183
184 return (ctype->flags & UPPER_MASK) != 0;
185 }
186
187 /* Returns the uppercase Unicode characters corresponding to ch or just
188 ch if no uppercase mapping is known. */
189
_PyUnicode_ToUppercase(Py_UCS4 ch)190 Py_UCS4 _PyUnicode_ToUppercase(Py_UCS4 ch)
191 {
192 const _PyUnicode_TypeRecord *ctype = gettyperecord(ch);
193
194 if (ctype->flags & EXTENDED_CASE_MASK)
195 return _PyUnicode_ExtendedCase[ctype->upper & 0xFFFF];
196 return ch + ctype->upper;
197 }
198
199 /* Returns the lowercase Unicode characters corresponding to ch or just
200 ch if no lowercase mapping is known. */
201
_PyUnicode_ToLowercase(Py_UCS4 ch)202 Py_UCS4 _PyUnicode_ToLowercase(Py_UCS4 ch)
203 {
204 const _PyUnicode_TypeRecord *ctype = gettyperecord(ch);
205
206 if (ctype->flags & EXTENDED_CASE_MASK)
207 return _PyUnicode_ExtendedCase[ctype->lower & 0xFFFF];
208 return ch + ctype->lower;
209 }
210
_PyUnicode_ToLowerFull(Py_UCS4 ch,Py_UCS4 * res)211 int _PyUnicode_ToLowerFull(Py_UCS4 ch, Py_UCS4 *res)
212 {
213 const _PyUnicode_TypeRecord *ctype = gettyperecord(ch);
214
215 if (ctype->flags & EXTENDED_CASE_MASK) {
216 int index = ctype->lower & 0xFFFF;
217 int n = ctype->lower >> 24;
218 int i;
219 for (i = 0; i < n; i++)
220 res[i] = _PyUnicode_ExtendedCase[index + i];
221 return n;
222 }
223 res[0] = ch + ctype->lower;
224 return 1;
225 }
226
_PyUnicode_ToTitleFull(Py_UCS4 ch,Py_UCS4 * res)227 int _PyUnicode_ToTitleFull(Py_UCS4 ch, Py_UCS4 *res)
228 {
229 const _PyUnicode_TypeRecord *ctype = gettyperecord(ch);
230
231 if (ctype->flags & EXTENDED_CASE_MASK) {
232 int index = ctype->title & 0xFFFF;
233 int n = ctype->title >> 24;
234 int i;
235 for (i = 0; i < n; i++)
236 res[i] = _PyUnicode_ExtendedCase[index + i];
237 return n;
238 }
239 res[0] = ch + ctype->title;
240 return 1;
241 }
242
_PyUnicode_ToUpperFull(Py_UCS4 ch,Py_UCS4 * res)243 int _PyUnicode_ToUpperFull(Py_UCS4 ch, Py_UCS4 *res)
244 {
245 const _PyUnicode_TypeRecord *ctype = gettyperecord(ch);
246
247 if (ctype->flags & EXTENDED_CASE_MASK) {
248 int index = ctype->upper & 0xFFFF;
249 int n = ctype->upper >> 24;
250 int i;
251 for (i = 0; i < n; i++)
252 res[i] = _PyUnicode_ExtendedCase[index + i];
253 return n;
254 }
255 res[0] = ch + ctype->upper;
256 return 1;
257 }
258
_PyUnicode_ToFoldedFull(Py_UCS4 ch,Py_UCS4 * res)259 int _PyUnicode_ToFoldedFull(Py_UCS4 ch, Py_UCS4 *res)
260 {
261 const _PyUnicode_TypeRecord *ctype = gettyperecord(ch);
262
263 if (ctype->flags & EXTENDED_CASE_MASK && (ctype->lower >> 20) & 7) {
264 int index = (ctype->lower & 0xFFFF) + (ctype->lower >> 24);
265 int n = (ctype->lower >> 20) & 7;
266 int i;
267 for (i = 0; i < n; i++)
268 res[i] = _PyUnicode_ExtendedCase[index + i];
269 return n;
270 }
271 return _PyUnicode_ToLowerFull(ch, res);
272 }
273
_PyUnicode_IsCased(Py_UCS4 ch)274 int _PyUnicode_IsCased(Py_UCS4 ch)
275 {
276 const _PyUnicode_TypeRecord *ctype = gettyperecord(ch);
277
278 return (ctype->flags & CASED_MASK) != 0;
279 }
280
_PyUnicode_IsCaseIgnorable(Py_UCS4 ch)281 int _PyUnicode_IsCaseIgnorable(Py_UCS4 ch)
282 {
283 const _PyUnicode_TypeRecord *ctype = gettyperecord(ch);
284
285 return (ctype->flags & CASE_IGNORABLE_MASK) != 0;
286 }
287
288 /* Returns 1 for Unicode characters having the category 'Ll', 'Lu', 'Lt',
289 'Lo' or 'Lm', 0 otherwise. */
290
_PyUnicode_IsAlpha(Py_UCS4 ch)291 int _PyUnicode_IsAlpha(Py_UCS4 ch)
292 {
293 const _PyUnicode_TypeRecord *ctype = gettyperecord(ch);
294
295 return (ctype->flags & ALPHA_MASK) != 0;
296 }
297
298