1 /* -*- Mode: C++; tab-width: 4; indent-tabs-mode: nil; c-basic-offset: 4 -*- */
3 * This file is part of the LibreOffice project.
5 * This Source Code Form is subject to the terms of the Mozilla Public
6 * License, v. 2.0. If a copy of the MPL was not distributed with this
7 * file, You can obtain one at http://mozilla.org/MPL/2.0/.
9 * This file incorporates work covered by the following license notice:
11 * Licensed to the Apache Software Foundation (ASF) under one or more
12 * contributor license agreements. See the NOTICE file distributed
13 * with this work for additional information regarding copyright
14 * ownership. The ASF licenses this file to you under the Apache
15 * License, Version 2.0 (the "License"); you may not use this file
16 * except in compliance with the License. You may obtain a copy of
17 * the License at http://www.apache.org/licenses/LICENSE-2.0 .
20 #ifndef INCLUDED_RTL_CHARACTER_HXX
21 #define INCLUDED_RTL_CHARACTER_HXX
23 #include "sal/config.h"
28 #include "sal/types.h"
33 /** Check for Unicode code point.
35 @param code An integer.
37 @return True if code is a Unicode code point.
39 @since LibreOffice 5.2
41 inline bool isUnicodeCodePoint(sal_uInt32 code
)
43 return code
<= 0x10FFFF;
46 /** Check for ASCII character.
48 @param code A Unicode code point.
50 @return True if code is an ASCII character (0x00--0x7F).
52 @since LibreOffice 4.1
54 inline bool isAscii(sal_uInt32 code
)
56 assert(isUnicodeCodePoint(code
));
60 /** Check for Unicode variation sequence selectors
62 @param code A Unicode code point.
64 @return True if code is an Unicode variation sequence selector.
66 @since LibreOffice 6.3
68 inline bool isIVSSelector(sal_uInt32 nCode
)
70 return (nCode
>= 0xFE00 && nCode
<= 0xFE0F) // Variation Selectors block
71 || (nCode
>= 0xE0100 && nCode
<= 0xE01EF);// Variation Selectors Supplement block
74 /** Check for base characters of a CJK ideographic variation sequence (IVS)
76 @param code A Unicode code point.
78 @return True if code is an Unicode base character part of CJK IVS
80 @since LibreOffice 6.3
82 inline bool isCJKIVSCharacter(sal_uInt32 nCode
)
84 return (nCode
>= 0x4E00 && nCode
<= 0x9FFF) // CJK Unified Ideographs
85 || (nCode
>= 0x3400 && nCode
<= 0x4DBF) // CJK Unified Ideographs Extension A
86 || (nCode
>= 0x20000 && nCode
<= 0x2A6DF); // CJK Unified Ideographs Extension B
89 #if defined LIBO_INTERNAL_ONLY
90 bool isAscii(char) = delete;
91 bool isAscii(signed char) = delete;
92 template<typename T
> inline bool isAscii(T code
)
93 { return isAscii(sal_uInt32(code
)); }
96 /** Check for ASCII lower case character.
98 @param code A Unicode code point.
100 @return True if code is an ASCII lower case alphabetic character (ASCII
103 @since LibreOffice 4.1
105 inline bool isAsciiLowerCase(sal_uInt32 code
)
107 assert(isUnicodeCodePoint(code
));
108 return code
>= 'a' && code
<= 'z';
111 #if defined LIBO_INTERNAL_ONLY
112 bool isAsciiLowerCase(char) = delete;
113 bool isAsciiLowerCase(signed char) = delete;
114 template<typename T
> inline bool isAsciiLowerCase(T code
)
115 { return isAsciiLowerCase(sal_uInt32(code
)); }
118 /** Check for ASCII upper case character.
120 @param code A Unicode code point.
122 @return True if code is an ASCII upper case alphabetic character (ASCII
125 @since LibreOffice 4.1
127 inline bool isAsciiUpperCase(sal_uInt32 code
)
129 assert(isUnicodeCodePoint(code
));
130 return code
>= 'A' && code
<= 'Z';
133 #if defined LIBO_INTERNAL_ONLY
134 bool isAsciiUpperCase(char) = delete;
135 bool isAsciiUpperCase(signed char) = delete;
136 template<typename T
> inline bool isAsciiUpperCase(T code
)
137 { return isAsciiUpperCase(sal_uInt32(code
)); }
140 /** Check for ASCII alphabetic character.
142 @param code A Unicode code point.
144 @return True if code is an ASCII alphabetic character (ASCII 'A'--'Z' or
147 @since LibreOffice 4.1
149 inline bool isAsciiAlpha(sal_uInt32 code
)
151 assert(isUnicodeCodePoint(code
));
152 return isAsciiLowerCase(code
) || isAsciiUpperCase(code
);
155 #if defined LIBO_INTERNAL_ONLY
156 bool isAsciiAlpha(char) = delete;
157 bool isAsciiAlpha(signed char) = delete;
158 template<typename T
> inline bool isAsciiAlpha(T code
)
159 { return isAsciiAlpha(sal_uInt32(code
)); }
162 /** Check for ASCII digit character.
164 @param code A Unicode code point.
166 @return True if code is an ASCII (decimal) digit character (ASCII
169 @since LibreOffice 4.1
171 inline bool isAsciiDigit(sal_uInt32 code
)
173 assert(isUnicodeCodePoint(code
));
174 return code
>= '0' && code
<= '9';
177 #if defined LIBO_INTERNAL_ONLY
178 bool isAsciiDigit(char) = delete;
179 bool isAsciiDigit(signed char) = delete;
180 template<typename T
> inline bool isAsciiDigit(T code
)
181 { return isAsciiDigit(sal_uInt32(code
)); }
184 /** Check for ASCII alphanumeric character.
186 @param code A Unicode code point.
188 @return True if code is an ASCII alphanumeric character (ASCII '0'--'9',
189 'A'--'Z', or 'a'--'z').
191 @since LibreOffice 4.1
193 inline bool isAsciiAlphanumeric(sal_uInt32 code
)
195 assert(isUnicodeCodePoint(code
));
196 return isAsciiDigit(code
) || isAsciiAlpha(code
);
199 #if defined LIBO_INTERNAL_ONLY
200 bool isAsciiAlphanumeric(char) = delete;
201 bool isAsciiAlphanumeric(signed char) = delete;
202 template<typename T
> inline bool isAsciiAlphanumeric(T code
)
203 { return isAsciiAlphanumeric(sal_uInt32(code
)); }
206 /** Check for ASCII canonic hexadecimal digit character.
208 @param code A Unicode code point.
210 @return True if code is an ASCII canonic (i.e., upper case) hexadecimal
211 digit character (ASCII '0'--'9' or 'A'--'F').
213 @since LibreOffice 4.1
215 inline bool isAsciiCanonicHexDigit(sal_uInt32 code
)
217 assert(isUnicodeCodePoint(code
));
218 return isAsciiDigit(code
) || (code
>= 'A' && code
<= 'F');
221 #if defined LIBO_INTERNAL_ONLY
222 bool isAsciiCanonicHexDigit(char) = delete;
223 bool isAsciiCanonicHexDigit(signed char) = delete;
224 template<typename T
> inline bool isAsciiCanonicHexDigit(T code
)
225 { return isAsciiCanonicHexDigit(sal_uInt32(code
)); }
228 /** Check for ASCII hexadecimal digit character.
230 @param code A Unicode code point.
232 @return True if code is an ASCII hexadecimal digit character (ASCII
233 '0'--'9', 'A'--'F', or 'a'--'f').
235 @since LibreOffice 4.1
237 inline bool isAsciiHexDigit(sal_uInt32 code
)
239 assert(isUnicodeCodePoint(code
));
240 return isAsciiCanonicHexDigit(code
) || (code
>= 'a' && code
<= 'f');
243 #if defined LIBO_INTERNAL_ONLY
244 bool isAsciiHexDigit(char) = delete;
245 bool isAsciiHexDigit(signed char) = delete;
246 template<typename T
> inline bool isAsciiHexDigit(T code
)
247 { return isAsciiHexDigit(sal_uInt32(code
)); }
250 /** Check for ASCII octal digit character.
252 @param code A Unicode code point.
254 @return True if code is an ASCII octal digit character (ASCII '0'--'7').
256 @since LibreOffice 5.0
258 inline bool isAsciiOctalDigit(sal_uInt32 code
)
260 assert(isUnicodeCodePoint(code
));
261 return code
>= '0' && code
<= '7';
264 #if defined LIBO_INTERNAL_ONLY
265 bool isAsciiOctalDigit(char) = delete;
266 bool isAsciiOctalDigit(signed char) = delete;
267 template<typename T
> inline bool isAsciiOctalDigit(T code
)
268 { return isAsciiOctalDigit(sal_uInt32(code
)); }
271 /** Check for ASCII white space character.
273 @param code A Unicode code point.
275 @return True if code is an ASCII white space character as defined by C for
276 isspace in the "C" locale (ASCII ' ', '\\f', '\\n', '\\r', '\\t' '\\v').
278 @since LibreOffice 5.4
280 inline bool isAsciiWhiteSpace(sal_uInt32 code
)
282 assert(isUnicodeCodePoint(code
));
283 return code
== ' ' || code
== '\f' || code
== '\n' || code
== '\r'
284 || code
== '\t' || code
== '\v';
287 #if defined LIBO_INTERNAL_ONLY
288 bool isAsciiWhiteSpace(char) = delete;
289 bool isAsciiWhiteSpace(signed char) = delete;
290 template<typename T
> inline bool isAsciiWhiteSpace(T code
)
291 { return isAsciiWhiteSpace(sal_uInt32(code
)); }
294 /** Convert a character, if ASCII, to upper case.
296 @param code A Unicode code point.
298 @return code converted to ASCII upper case.
300 @since LibreOffice 4.2
302 inline sal_uInt32
toAsciiUpperCase(sal_uInt32 code
)
304 assert(isUnicodeCodePoint(code
));
305 return isAsciiLowerCase(code
) ? code
- 32 : code
;
308 #if defined LIBO_INTERNAL_ONLY
309 sal_uInt32
toAsciiUpperCase(char) = delete;
310 sal_uInt32
toAsciiUpperCase(signed char) = delete;
311 template<typename T
> inline sal_uInt32
toAsciiUpperCase(T code
)
312 { return toAsciiUpperCase(sal_uInt32(code
)); }
315 /** Convert a character, if ASCII, to lower case.
317 @param code A Unicode code point.
319 @return code converted to ASCII lower case.
321 @since LibreOffice 4.2
323 inline sal_uInt32
toAsciiLowerCase(sal_uInt32 code
)
325 assert(isUnicodeCodePoint(code
));
326 return isAsciiUpperCase(code
) ? code
+ 32 : code
;
329 #if defined LIBO_INTERNAL_ONLY
330 sal_uInt32
toAsciiLowerCase(char) = delete;
331 sal_uInt32
toAsciiLowerCase(signed char) = delete;
332 template<typename T
> inline sal_uInt32
toAsciiLowerCase(T code
)
333 { return toAsciiLowerCase(sal_uInt32(code
)); }
336 /** Compare two characters ignoring ASCII case.
338 @param code1 A Unicode code point.
340 @param code2 A unicode code point.
342 @return 0 if both code points are equal,
343 < 0 if code1 is less than code2,
344 > 0 if code1 is greater than code2.
346 @since LibreOffice 4.2
348 inline sal_Int32
compareIgnoreAsciiCase(sal_uInt32 code1
, sal_uInt32 code2
)
350 assert(isUnicodeCodePoint(code1
));
351 assert(isUnicodeCodePoint(code2
));
352 return static_cast<sal_Int32
>(toAsciiLowerCase(code1
))
353 - static_cast<sal_Int32
>(toAsciiLowerCase(code2
));
359 sal_uInt32
const surrogatesHighFirst
= 0xD800;
360 sal_uInt32
const surrogatesHighLast
= 0xDBFF;
361 sal_uInt32
const surrogatesLowFirst
= 0xDC00;
362 sal_uInt32
const surrogatesLowLast
= 0xDFFF;
367 /** Check for surrogate.
369 @param code A Unicode code point.
371 @return True if code is a surrogate code point (0xD800--0xDFFF).
373 @since LibreOffice 6.0
375 inline bool isSurrogate(sal_uInt32 code
) {
376 assert(isUnicodeCodePoint(code
));
377 return code
>= detail::surrogatesHighFirst
378 && code
<= detail::surrogatesLowLast
;
381 /** Check for high surrogate.
383 @param code A Unicode code point.
385 @return True if code is a high surrogate code point (0xD800--0xDBFF).
387 @since LibreOffice 5.0
389 inline bool isHighSurrogate(sal_uInt32 code
) {
390 assert(isUnicodeCodePoint(code
));
391 return code
>= detail::surrogatesHighFirst
392 && code
<= detail::surrogatesHighLast
;
395 /** Check for low surrogate.
397 @param code A Unicode code point.
399 @return True if code is a low surrogate code point (0xDC00--0xDFFF).
401 @since LibreOffice 5.0
403 inline bool isLowSurrogate(sal_uInt32 code
) {
404 assert(isUnicodeCodePoint(code
));
405 return code
>= detail::surrogatesLowFirst
406 && code
<= detail::surrogatesLowLast
;
409 /** Get high surrogate half of a non-BMP Unicode code point.
411 @param code A non-BMP Unicode code point.
413 @return The UTF-16 high surrogate half for the give code point.
415 @since LibreOffice 5.0
417 inline sal_Unicode
getHighSurrogate(sal_uInt32 code
) {
418 assert(isUnicodeCodePoint(code
));
419 assert(code
>= 0x10000);
420 return static_cast<sal_Unicode
>(((code
- 0x10000) >> 10) | detail::surrogatesHighFirst
);
423 /** Get low surrogate half of a non-BMP Unicode code point.
425 @param code A non-BMP Unicode code point.
427 @return The UTF-16 low surrogate half for the give code point.
429 @since LibreOffice 5.0
431 inline sal_Unicode
getLowSurrogate(sal_uInt32 code
) {
432 assert(isUnicodeCodePoint(code
));
433 assert(code
>= 0x10000);
434 return static_cast<sal_Unicode
>(((code
- 0x10000) & 0x3FF) | detail::surrogatesLowFirst
);
437 /** Combine surrogates to form a code point.
439 @param high A high surrogate code point.
441 @param low A low surrogate code point.
443 @return The code point represented by the surrogate pair.
445 @since LibreOffice 5.0
447 inline sal_uInt32
combineSurrogates(sal_uInt32 high
, sal_uInt32 low
) {
448 assert(isHighSurrogate(high
));
449 assert(isLowSurrogate(low
));
450 return ((high
- detail::surrogatesHighFirst
) << 10)
451 + (low
- detail::surrogatesLowFirst
) + 0x10000;
454 /** Split a Unicode code point into UTF-16 code units.
456 @param code A Unicode code point.
458 @param output A non-null pointer to an array with space for at least two
459 sal_Unicode UTF-16 code units.
461 @return The number of UTF-16 code units placed into the output (either one
464 @since LibreOffice 5.3
466 inline std::size_t splitSurrogates(sal_uInt32 code
, sal_Unicode
* output
) {
467 assert(isUnicodeCodePoint(code
));
468 assert(output
!= NULL
);
469 if (code
< 0x10000) {
473 output
[0] = getHighSurrogate(code
);
474 output
[1] = getLowSurrogate(code
);
479 /** Check for Unicode scalar value.
481 @param code An integer.
483 @return True if code is a Unicode scalar value.
485 @since LibreOffice 6.0
487 inline bool isUnicodeScalarValue(sal_uInt32 code
)
489 return isUnicodeCodePoint(code
) && !isSurrogate(code
);
496 /* vim:set shiftwidth=4 softtabstop=4 expandtab: */