source: mainline/kernel/generic/src/lib/str.c@ 550523f5

lfn serial ticket/834-toolchain-update topic/msim-upgrade topic/simplify-dev-export
Last change on this file since 550523f5 was 1772e6d, checked in by Martin Sucha <sucha14@…>, 13 years ago

Update documentation for str_cmp and str_lcmp.

This is a modified version of formulation suggested by Jiri Zarevucky.

  • Property mode set to 100644
File size: 24.2 KB
RevLine 
[16da5f8e]1/*
2 * Copyright (c) 2001-2004 Jakub Jermar
3 * All rights reserved.
4 *
5 * Redistribution and use in source and binary forms, with or without
6 * modification, are permitted provided that the following conditions
7 * are met:
8 *
9 * - Redistributions of source code must retain the above copyright
10 * notice, this list of conditions and the following disclaimer.
11 * - Redistributions in binary form must reproduce the above copyright
12 * notice, this list of conditions and the following disclaimer in the
13 * documentation and/or other materials provided with the distribution.
14 * - The name of the author may not be used to endorse or promote products
15 * derived from this software without specific prior written permission.
16 *
17 * THIS SOFTWARE IS PROVIDED BY THE AUTHOR ``AS IS'' AND ANY EXPRESS OR
18 * IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES
19 * OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED.
20 * IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR ANY DIRECT, INDIRECT,
21 * INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT
22 * NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
23 * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
24 * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
25 * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF
26 * THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
27 */
28
[2f57690]29/** @addtogroup generic
[16da5f8e]30 * @{
31 */
32
33/**
34 * @file
[82bb9c1]35 * @brief String functions.
36 *
37 * Strings and characters use the Universal Character Set (UCS). The standard
38 * strings, called just strings are encoded in UTF-8. Wide strings (encoded
39 * in UTF-32) are supported to a limited degree. A single character is
[b888d5f]40 * represented as wchar_t.@n
[82bb9c1]41 *
[b888d5f]42 * Overview of the terminology:@n
[82bb9c1]43 *
[b888d5f]44 * Term Meaning
45 * -------------------- ----------------------------------------------------
46 * byte 8 bits stored in uint8_t (unsigned 8 bit integer)
[82bb9c1]47 *
[b888d5f]48 * character UTF-32 encoded Unicode character, stored in wchar_t
49 * (signed 32 bit integer), code points 0 .. 1114111
50 * are valid
[82bb9c1]51 *
[b888d5f]52 * ASCII character 7 bit encoded ASCII character, stored in char
53 * (usually signed 8 bit integer), code points 0 .. 127
54 * are valid
55 *
56 * string UTF-8 encoded NULL-terminated Unicode string, char *
57 *
58 * wide string UTF-32 encoded NULL-terminated Unicode string,
59 * wchar_t *
60 *
61 * [wide] string size number of BYTES in a [wide] string (excluding
62 * the NULL-terminator), size_t
63 *
64 * [wide] string length number of CHARACTERS in a [wide] string (excluding
[98000fb]65 * the NULL-terminator), size_t
[b888d5f]66 *
67 * [wide] string width number of display cells on a monospace display taken
[98000fb]68 * by a [wide] string, size_t
[b888d5f]69 *
70 *
71 * Overview of string metrics:@n
72 *
73 * Metric Abbrev. Type Meaning
74 * ------ ------ ------ -------------------------------------------------
75 * size n size_t number of BYTES in a string (excluding the
76 * NULL-terminator)
77 *
[98000fb]78 * length l size_t number of CHARACTERS in a string (excluding the
[b888d5f]79 * null terminator)
80 *
[98000fb]81 * width w size_t number of display cells on a monospace display
[b888d5f]82 * taken by a string
83 *
84 *
85 * Function naming prefixes:@n
86 *
87 * chr_ operate on characters
88 * ascii_ operate on ASCII characters
89 * str_ operate on strings
90 * wstr_ operate on wide strings
91 *
92 * [w]str_[n|l|w] operate on a prefix limited by size, length
93 * or width
94 *
95 *
96 * A specific character inside a [wide] string can be referred to by:@n
97 *
98 * pointer (char *, wchar_t *)
99 * byte offset (size_t)
[98000fb]100 * character index (size_t)
[82bb9c1]101 *
[16da5f8e]102 */
103
[19f857a]104#include <str.h>
[16da5f8e]105#include <print.h>
106#include <cpu.h>
107#include <arch/asm.h>
108#include <arch.h>
[d09f84e6]109#include <errno.h>
[b888d5f]110#include <align.h>
[6700ee2]111#include <debug.h>
[30a5470]112#include <macros.h>
[16da5f8e]113
[8e893ae]114/** Check the condition if wchar_t is signed */
115#ifdef WCHAR_IS_UNSIGNED
116 #define WCHAR_SIGNED_CHECK(cond) (true)
117#else
118 #define WCHAR_SIGNED_CHECK(cond) (cond)
119#endif
120
[b888d5f]121/** Byte mask consisting of lowest @n bits (out of 8) */
122#define LO_MASK_8(n) ((uint8_t) ((1 << (n)) - 1))
[0dd1d444]123
[b888d5f]124/** Byte mask consisting of lowest @n bits (out of 32) */
125#define LO_MASK_32(n) ((uint32_t) ((1 << (n)) - 1))
[32704cb]126
[b888d5f]127/** Byte mask consisting of highest @n bits (out of 8) */
128#define HI_MASK_8(n) (~LO_MASK_8(8 - (n)))
[32704cb]129
[b888d5f]130/** Number of data bits in a UTF-8 continuation byte */
131#define CONT_BITS 6
[0dd1d444]132
[b888d5f]133/** Decode a single character from a string.
[21a639b7]134 *
[b888d5f]135 * Decode a single character from a string of size @a size. Decoding starts
[e1813cf]136 * at @a offset and this offset is moved to the beginning of the next
137 * character. In case of decoding error, offset generally advances at least
[b888d5f]138 * by one. However, offset is never moved beyond size.
[21a639b7]139 *
[b888d5f]140 * @param str String (not necessarily NULL-terminated).
141 * @param offset Byte offset in string where to start decoding.
142 * @param size Size of the string (in bytes).
143 *
[c8bf88d]144 * @return Value of decoded character, U_SPECIAL on decoding error or
[b888d5f]145 * NULL if attempt to decode beyond @a size.
[21a639b7]146 *
147 */
[b888d5f]148wchar_t str_decode(const char *str, size_t *offset, size_t size)
[21a639b7]149{
[b888d5f]150 if (*offset + 1 > size)
151 return 0;
152
153 /* First byte read from string */
154 uint8_t b0 = (uint8_t) str[(*offset)++];
155
156 /* Determine code length */
157
158 unsigned int b0_bits; /* Data bits in first byte */
159 unsigned int cbytes; /* Number of continuation bytes */
160
[0dd1d444]161 if ((b0 & 0x80) == 0) {
162 /* 0xxxxxxx (Plain ASCII) */
163 b0_bits = 7;
164 cbytes = 0;
165 } else if ((b0 & 0xe0) == 0xc0) {
166 /* 110xxxxx 10xxxxxx */
167 b0_bits = 5;
168 cbytes = 1;
169 } else if ((b0 & 0xf0) == 0xe0) {
170 /* 1110xxxx 10xxxxxx 10xxxxxx */
171 b0_bits = 4;
172 cbytes = 2;
173 } else if ((b0 & 0xf8) == 0xf0) {
174 /* 11110xxx 10xxxxxx 10xxxxxx 10xxxxxx */
175 b0_bits = 3;
176 cbytes = 3;
177 } else {
[b888d5f]178 /* 10xxxxxx -- unexpected continuation byte */
[c8bf88d]179 return U_SPECIAL;
[74c8da2c]180 }
[b888d5f]181
182 if (*offset + cbytes > size)
[c8bf88d]183 return U_SPECIAL;
[b888d5f]184
185 wchar_t ch = b0 & LO_MASK_8(b0_bits);
186
187 /* Decode continuation bytes */
[0dd1d444]188 while (cbytes > 0) {
[b888d5f]189 uint8_t b = (uint8_t) str[(*offset)++];
190
191 /* Must be 10xxxxxx */
192 if ((b & 0xc0) != 0x80)
[c8bf88d]193 return U_SPECIAL;
[b888d5f]194
195 /* Shift data bits to ch */
[0dd1d444]196 ch = (ch << CONT_BITS) | (wchar_t) (b & LO_MASK_8(CONT_BITS));
[b888d5f]197 cbytes--;
[74c8da2c]198 }
[b888d5f]199
[0dd1d444]200 return ch;
[74c8da2c]201}
202
[e1813cf]203/** Encode a single character to string representation.
[74c8da2c]204 *
[e1813cf]205 * Encode a single character to string representation (i.e. UTF-8) and store
206 * it into a buffer at @a offset. Encoding starts at @a offset and this offset
207 * is moved to the position where the next character can be written to.
[74c8da2c]208 *
[b888d5f]209 * @param ch Input character.
210 * @param str Output buffer.
211 * @param offset Byte offset where to start writing.
212 * @param size Size of the output buffer (in bytes).
[74c8da2c]213 *
[d09f84e6]214 * @return EOK if the character was encoded successfully, EOVERFLOW if there
[8e893ae]215 * was not enough space in the output buffer or EINVAL if the character
216 * code was invalid.
[74c8da2c]217 */
[8e893ae]218int chr_encode(const wchar_t ch, char *str, size_t *offset, size_t size)
[74c8da2c]219{
[b888d5f]220 if (*offset >= size)
[d09f84e6]221 return EOVERFLOW;
[b888d5f]222
223 if (!chr_check(ch))
[d09f84e6]224 return EINVAL;
[b888d5f]225
226 /* Unsigned version of ch (bit operations should only be done
227 on unsigned types). */
228 uint32_t cc = (uint32_t) ch;
229
230 /* Determine how many continuation bytes are needed */
231
232 unsigned int b0_bits; /* Data bits in first byte */
233 unsigned int cbytes; /* Number of continuation bytes */
234
[32704cb]235 if ((cc & ~LO_MASK_32(7)) == 0) {
236 b0_bits = 7;
237 cbytes = 0;
238 } else if ((cc & ~LO_MASK_32(11)) == 0) {
239 b0_bits = 5;
240 cbytes = 1;
241 } else if ((cc & ~LO_MASK_32(16)) == 0) {
242 b0_bits = 4;
243 cbytes = 2;
244 } else if ((cc & ~LO_MASK_32(21)) == 0) {
245 b0_bits = 3;
246 cbytes = 3;
247 } else {
[b888d5f]248 /* Codes longer than 21 bits are not supported */
[d09f84e6]249 return EINVAL;
[74c8da2c]250 }
[b888d5f]251
252 /* Check for available space in buffer */
253 if (*offset + cbytes >= size)
[d09f84e6]254 return EOVERFLOW;
[b888d5f]255
256 /* Encode continuation bytes */
257 unsigned int i;
258 for (i = cbytes; i > 0; i--) {
[e1813cf]259 str[*offset + i] = 0x80 | (cc & LO_MASK_32(CONT_BITS));
[32704cb]260 cc = cc >> CONT_BITS;
[74c8da2c]261 }
[b888d5f]262
263 /* Encode first byte */
[e1813cf]264 str[*offset] = (cc & LO_MASK_32(b0_bits)) | HI_MASK_8(8 - b0_bits - 1);
[b888d5f]265
266 /* Advance offset */
267 *offset += cbytes + 1;
[74c8da2c]268
[d09f84e6]269 return EOK;
[74c8da2c]270}
271
[b888d5f]272/** Get size of string.
273 *
274 * Get the number of bytes which are used by the string @a str (excluding the
275 * NULL-terminator).
276 *
277 * @param str String to consider.
278 *
279 * @return Number of bytes used by the string
[82bb9c1]280 *
281 */
[b888d5f]282size_t str_size(const char *str)
[82bb9c1]283{
[b888d5f]284 size_t size = 0;
285
286 while (*str++ != 0)
287 size++;
288
289 return size;
[82bb9c1]290}
291
[b888d5f]292/** Get size of wide string.
293 *
294 * Get the number of bytes which are used by the wide string @a str (excluding the
295 * NULL-terminator).
296 *
297 * @param str Wide string to consider.
298 *
299 * @return Number of bytes used by the wide string
300 *
301 */
302size_t wstr_size(const wchar_t *str)
303{
304 return (wstr_length(str) * sizeof(wchar_t));
305}
306
307/** Get size of string with length limit.
[74c8da2c]308 *
[f25b2819]309 * Get the number of bytes which are used by up to @a max_len first
310 * characters in the string @a str. If @a max_len is greater than
[b888d5f]311 * the length of @a str, the entire string is measured (excluding the
312 * NULL-terminator).
313 *
314 * @param str String to consider.
315 * @param max_len Maximum number of characters to measure.
[74c8da2c]316 *
[b888d5f]317 * @return Number of bytes used by the characters.
[74c8da2c]318 *
319 */
[98000fb]320size_t str_lsize(const char *str, size_t max_len)
[74c8da2c]321{
[98000fb]322 size_t len = 0;
[b888d5f]323 size_t offset = 0;
324
325 while (len < max_len) {
326 if (str_decode(str, &offset, STR_NO_LIMIT) == 0)
[b54d2f1]327 break;
[b888d5f]328
[f25b2819]329 len++;
[21a639b7]330 }
[b888d5f]331
332 return offset;
[74c8da2c]333}
334
[b888d5f]335/** Get size of wide string with length limit.
[82bb9c1]336 *
[b888d5f]337 * Get the number of bytes which are used by up to @a max_len first
338 * wide characters in the wide string @a str. If @a max_len is greater than
339 * the length of @a str, the entire wide string is measured (excluding the
340 * NULL-terminator).
341 *
342 * @param str Wide string to consider.
343 * @param max_len Maximum number of wide characters to measure.
[82bb9c1]344 *
[b888d5f]345 * @return Number of bytes used by the wide characters.
[82bb9c1]346 *
347 */
[98000fb]348size_t wstr_lsize(const wchar_t *str, size_t max_len)
[82bb9c1]349{
[b888d5f]350 return (wstr_nlength(str, max_len * sizeof(wchar_t)) * sizeof(wchar_t));
[82bb9c1]351}
352
[b888d5f]353/** Get number of characters in a string.
[82bb9c1]354 *
[b888d5f]355 * @param str NULL-terminated string.
[82bb9c1]356 *
[b888d5f]357 * @return Number of characters in string.
[82bb9c1]358 *
359 */
[98000fb]360size_t str_length(const char *str)
[82bb9c1]361{
[98000fb]362 size_t len = 0;
[b888d5f]363 size_t offset = 0;
364
365 while (str_decode(str, &offset, STR_NO_LIMIT) != 0)
366 len++;
367
368 return len;
[82bb9c1]369}
370
[b888d5f]371/** Get number of characters in a wide string.
[74c8da2c]372 *
[b888d5f]373 * @param str NULL-terminated wide string.
374 *
375 * @return Number of characters in @a str.
[74c8da2c]376 *
377 */
[98000fb]378size_t wstr_length(const wchar_t *wstr)
[74c8da2c]379{
[98000fb]380 size_t len = 0;
[74c8da2c]381
[b888d5f]382 while (*wstr++ != 0)
383 len++;
384
385 return len;
[74c8da2c]386}
387
[b888d5f]388/** Get number of characters in a string with size limit.
389 *
390 * @param str NULL-terminated string.
391 * @param size Maximum number of bytes to consider.
392 *
393 * @return Number of characters in string.
[74c8da2c]394 *
395 */
[98000fb]396size_t str_nlength(const char *str, size_t size)
[74c8da2c]397{
[98000fb]398 size_t len = 0;
[b888d5f]399 size_t offset = 0;
[74c8da2c]400
[b888d5f]401 while (str_decode(str, &offset, size) != 0)
402 len++;
403
404 return len;
[21a639b7]405}
406
[b888d5f]407/** Get number of characters in a string with size limit.
[2f57690]408 *
[b888d5f]409 * @param str NULL-terminated string.
410 * @param size Maximum number of bytes to consider.
[74c8da2c]411 *
[f25b2819]412 * @return Number of characters in string.
[b888d5f]413 *
[74c8da2c]414 */
[98000fb]415size_t wstr_nlength(const wchar_t *str, size_t size)
[74c8da2c]416{
[98000fb]417 size_t len = 0;
418 size_t limit = ALIGN_DOWN(size, sizeof(wchar_t));
419 size_t offset = 0;
[b888d5f]420
421 while ((offset < limit) && (*str++ != 0)) {
[f25b2819]422 len++;
[b888d5f]423 offset += sizeof(wchar_t);
[74c8da2c]424 }
[b888d5f]425
[f25b2819]426 return len;
[74c8da2c]427}
428
[b888d5f]429/** Check whether character is plain ASCII.
430 *
431 * @return True if character is plain ASCII.
[74c8da2c]432 *
433 */
[f2b8cdc]434bool ascii_check(wchar_t ch)
[74c8da2c]435{
[8e893ae]436 if (WCHAR_SIGNED_CHECK(ch >= 0) && (ch <= 127))
[b888d5f]437 return true;
438
439 return false;
440}
[f25b2819]441
[b888d5f]442/** Check whether character is valid
443 *
444 * @return True if character is a valid Unicode code point.
445 *
446 */
[f2b8cdc]447bool chr_check(wchar_t ch)
[b888d5f]448{
[8e893ae]449 if (WCHAR_SIGNED_CHECK(ch >= 0) && (ch <= 1114111))
[b888d5f]450 return true;
451
452 return false;
[16da5f8e]453}
454
[b888d5f]455/** Compare two NULL terminated strings.
[16da5f8e]456 *
[b888d5f]457 * Do a char-by-char comparison of two NULL-terminated strings.
[4efeab5]458 * The strings are considered equal iff their length is equal
459 * and both strings consist of the same sequence of characters.
460 *
[1772e6d]461 * A string S1 is less than another string S2 if it has a character with
462 * lower value at the first character position where the strings differ.
463 * If the strings differ in length, the shorter one is treated as if
464 * padded by characters with a value of zero.
[16da5f8e]465 *
[b888d5f]466 * @param s1 First string to compare.
467 * @param s2 Second string to compare.
[16da5f8e]468 *
[1772e6d]469 * @return 0 if the strings are equal, -1 if the first is less than the second,
470 * 1 if the second is less than the first.
[16da5f8e]471 *
472 */
[b888d5f]473int str_cmp(const char *s1, const char *s2)
[16da5f8e]474{
[a7b1071]475 wchar_t c1 = 0;
476 wchar_t c2 = 0;
[b888d5f]477
478 size_t off1 = 0;
479 size_t off2 = 0;
[a7b1071]480
481 while (true) {
482 c1 = str_decode(s1, &off1, STR_NO_LIMIT);
483 c2 = str_decode(s2, &off2, STR_NO_LIMIT);
484
[b888d5f]485 if (c1 < c2)
[16da5f8e]486 return -1;
[b888d5f]487
488 if (c1 > c2)
[16da5f8e]489 return 1;
[a7b1071]490
491 if (c1 == 0 || c2 == 0)
492 break;
[16da5f8e]493 }
[a7b1071]494
495 return 0;
[16da5f8e]496}
497
[b888d5f]498/** Compare two NULL terminated strings with length limit.
[16da5f8e]499 *
[b888d5f]500 * Do a char-by-char comparison of two NULL-terminated strings.
[4efeab5]501 * The strings are considered equal iff
502 * min(str_length(s1), max_len) == min(str_length(s2), max_len)
503 * and both strings consist of the same sequence of characters,
504 * up to max_len characters.
505 *
[1772e6d]506 * A string S1 is less than another string S2 if it has a character with
507 * lower value at the first character position where the strings differ.
508 * If the strings differ in length, the shorter one is treated as if
509 * padded by characters with a value of zero. Only the first max_len
510 * characters are considered.
[16da5f8e]511 *
[b888d5f]512 * @param s1 First string to compare.
513 * @param s2 Second string to compare.
514 * @param max_len Maximum number of characters to consider.
515 *
[1772e6d]516 * @return 0 if the strings are equal, -1 if the first is less than the second,
517 * 1 if the second is less than the first.
[16da5f8e]518 *
519 */
[98000fb]520int str_lcmp(const char *s1, const char *s2, size_t max_len)
[16da5f8e]521{
[b888d5f]522 wchar_t c1 = 0;
523 wchar_t c2 = 0;
[16da5f8e]524
[b888d5f]525 size_t off1 = 0;
526 size_t off2 = 0;
527
[98000fb]528 size_t len = 0;
[a7b1071]529
530 while (true) {
531 if (len >= max_len)
[b888d5f]532 break;
[a7b1071]533
534 c1 = str_decode(s1, &off1, STR_NO_LIMIT);
535 c2 = str_decode(s2, &off2, STR_NO_LIMIT);
536
[b888d5f]537 if (c1 < c2)
[16da5f8e]538 return -1;
[a7b1071]539
[b888d5f]540 if (c1 > c2)
[16da5f8e]541 return 1;
[a7b1071]542
543 if (c1 == 0 || c2 == 0)
544 break;
545
546 ++len;
[16da5f8e]547 }
[a7b1071]548
549 return 0;
550
[16da5f8e]551}
552
[f4b1535]553/** Copy string.
[b888d5f]554 *
[f4b1535]555 * Copy source string @a src to destination buffer @a dest.
556 * No more than @a size bytes are written. If the size of the output buffer
557 * is at least one byte, the output string will always be well-formed, i.e.
558 * null-terminated and containing only complete characters.
[b888d5f]559 *
[abf09311]560 * @param dest Destination buffer.
[6700ee2]561 * @param count Size of the destination buffer (must be > 0).
[f4b1535]562 * @param src Source string.
[abf09311]563 *
[b888d5f]564 */
[f4b1535]565void str_cpy(char *dest, size_t size, const char *src)
[b888d5f]566{
[6700ee2]567 /* There must be space for a null terminator in the buffer. */
568 ASSERT(size > 0);
[b888d5f]569
[abf09311]570 size_t src_off = 0;
571 size_t dest_off = 0;
572
573 wchar_t ch;
[f4b1535]574 while ((ch = str_decode(src, &src_off, STR_NO_LIMIT)) != 0) {
575 if (chr_encode(ch, dest, &dest_off, size - 1) != EOK)
576 break;
577 }
[abf09311]578
[f4b1535]579 dest[dest_off] = '\0';
580}
581
582/** Copy size-limited substring.
583 *
[6700ee2]584 * Copy prefix of string @a src of max. size @a size to destination buffer
585 * @a dest. No more than @a size bytes are written. The output string will
586 * always be well-formed, i.e. null-terminated and containing only complete
587 * characters.
[f4b1535]588 *
589 * No more than @a n bytes are read from the input string, so it does not
590 * have to be null-terminated.
591 *
[abf09311]592 * @param dest Destination buffer.
[6700ee2]593 * @param count Size of the destination buffer (must be > 0).
[f4b1535]594 * @param src Source string.
[abf09311]595 * @param n Maximum number of bytes to read from @a src.
596 *
[f4b1535]597 */
598void str_ncpy(char *dest, size_t size, const char *src, size_t n)
599{
[6700ee2]600 /* There must be space for a null terminator in the buffer. */
601 ASSERT(size > 0);
[b888d5f]602
[abf09311]603 size_t src_off = 0;
604 size_t dest_off = 0;
605
606 wchar_t ch;
[f4b1535]607 while ((ch = str_decode(src, &src_off, n)) != 0) {
608 if (chr_encode(ch, dest, &dest_off, size - 1) != EOK)
[b888d5f]609 break;
610 }
[abf09311]611
[f4b1535]612 dest[dest_off] = '\0';
[b888d5f]613}
[16da5f8e]614
[abf09311]615/** Duplicate string.
616 *
617 * Allocate a new string and copy characters from the source
618 * string into it. The duplicate string is allocated via sleeping
619 * malloc(), thus this function can sleep in no memory conditions.
620 *
621 * The allocation cannot fail and the return value is always
622 * a valid pointer. The duplicate string is always a well-formed
623 * null-terminated UTF-8 string, but it can differ from the source
624 * string on the byte level.
625 *
626 * @param src Source string.
627 *
628 * @return Duplicate string.
629 *
630 */
631char *str_dup(const char *src)
632{
633 size_t size = str_size(src) + 1;
634 char *dest = malloc(size, 0);
635 ASSERT(dest);
636
637 str_cpy(dest, size, src);
638 return dest;
639}
640
641/** Duplicate string with size limit.
642 *
643 * Allocate a new string and copy up to @max_size bytes from the source
644 * string into it. The duplicate string is allocated via sleeping
645 * malloc(), thus this function can sleep in no memory conditions.
646 * No more than @max_size + 1 bytes is allocated, but if the size
647 * occupied by the source string is smaller than @max_size + 1,
648 * less is allocated.
649 *
650 * The allocation cannot fail and the return value is always
651 * a valid pointer. The duplicate string is always a well-formed
652 * null-terminated UTF-8 string, but it can differ from the source
653 * string on the byte level.
654 *
655 * @param src Source string.
656 * @param n Maximum number of bytes to duplicate.
657 *
658 * @return Duplicate string.
659 *
660 */
661char *str_ndup(const char *src, size_t n)
662{
663 size_t size = str_size(src);
664 if (size > n)
665 size = n;
666
667 char *dest = malloc(size + 1, 0);
668 ASSERT(dest);
669
670 str_ncpy(dest, size + 1, src, size);
671 return dest;
672}
673
[0f06dbc]674/** Convert wide string to string.
[b888d5f]675 *
[0f06dbc]676 * Convert wide string @a src to string. The output is written to the buffer
677 * specified by @a dest and @a size. @a size must be non-zero and the string
678 * written will always be well-formed.
[16da5f8e]679 *
[0f06dbc]680 * @param dest Destination buffer.
681 * @param size Size of the destination buffer.
682 * @param src Source wide string.
[16da5f8e]683 */
[0f06dbc]684void wstr_to_str(char *dest, size_t size, const wchar_t *src)
[16da5f8e]685{
[b888d5f]686 wchar_t ch;
[0f06dbc]687 size_t src_idx;
688 size_t dest_off;
689
690 /* There must be space for a null terminator in the buffer. */
691 ASSERT(size > 0);
692
693 src_idx = 0;
694 dest_off = 0;
[b888d5f]695
696 while ((ch = src[src_idx++]) != 0) {
[0f06dbc]697 if (chr_encode(ch, dest, &dest_off, size - 1) != EOK)
[b888d5f]698 break;
[16da5f8e]699 }
[0f06dbc]700
701 dest[dest_off] = '\0';
[16da5f8e]702}
703
[20f1597]704/** Find first occurence of character in string.
705 *
[b888d5f]706 * @param str String to search.
707 * @param ch Character to look for.
708 *
709 * @return Pointer to character in @a str or NULL if not found.
[20f1597]710 *
711 */
[dd2cfa7]712char *str_chr(const char *str, wchar_t ch)
[20f1597]713{
[b888d5f]714 wchar_t acc;
715 size_t off = 0;
[f2d2c7ba]716 size_t last = 0;
[b888d5f]717
[a7b1071]718 while ((acc = str_decode(str, &off, STR_NO_LIMIT)) != 0) {
[b888d5f]719 if (acc == ch)
[dd2cfa7]720 return (char *) (str + last);
[f2d2c7ba]721 last = off;
[20f1597]722 }
[2f57690]723
[20f1597]724 return NULL;
725}
726
[b888d5f]727/** Insert a wide character into a wide string.
728 *
729 * Insert a wide character into a wide string at position
730 * @a pos. The characters after the position are shifted.
731 *
732 * @param str String to insert to.
733 * @param ch Character to insert to.
734 * @param pos Character index where to insert.
735 @ @param max_pos Characters in the buffer.
736 *
737 * @return True if the insertion was sucessful, false if the position
738 * is out of bounds.
739 *
740 */
[98000fb]741bool wstr_linsert(wchar_t *str, wchar_t ch, size_t pos, size_t max_pos)
[b888d5f]742{
[98000fb]743 size_t len = wstr_length(str);
[b888d5f]744
745 if ((pos > len) || (pos + 1 > max_pos))
746 return false;
747
[98000fb]748 size_t i;
[b888d5f]749 for (i = len; i + 1 > pos; i--)
750 str[i + 1] = str[i];
751
752 str[pos] = ch;
753
754 return true;
755}
756
757/** Remove a wide character from a wide string.
758 *
759 * Remove a wide character from a wide string at position
760 * @a pos. The characters after the position are shifted.
761 *
762 * @param str String to remove from.
763 * @param pos Character index to remove.
764 *
765 * @return True if the removal was sucessful, false if the position
766 * is out of bounds.
767 *
768 */
[98000fb]769bool wstr_remove(wchar_t *str, size_t pos)
[b888d5f]770{
[98000fb]771 size_t len = wstr_length(str);
[b888d5f]772
773 if (pos >= len)
774 return false;
775
[98000fb]776 size_t i;
[b888d5f]777 for (i = pos + 1; i <= len; i++)
778 str[i - 1] = str[i];
779
780 return true;
781}
782
[30a5470]783/** Convert string to uint64_t (internal variant).
784 *
785 * @param nptr Pointer to string.
786 * @param endptr Pointer to the first invalid character is stored here.
787 * @param base Zero or number between 2 and 36 inclusive.
788 * @param neg Indication of unary minus is stored here.
789 * @apram result Result of the conversion.
790 *
791 * @return EOK if conversion was successful.
792 *
793 */
794static int str_uint(const char *nptr, char **endptr, unsigned int base,
795 bool *neg, uint64_t *result)
796{
797 ASSERT(endptr != NULL);
798 ASSERT(neg != NULL);
799 ASSERT(result != NULL);
800
801 *neg = false;
802 const char *str = nptr;
803
804 /* Ignore leading whitespace */
805 while (isspace(*str))
806 str++;
807
808 if (*str == '-') {
809 *neg = true;
810 str++;
811 } else if (*str == '+')
812 str++;
813
814 if (base == 0) {
815 /* Decode base if not specified */
816 base = 10;
817
818 if (*str == '0') {
819 base = 8;
820 str++;
821
822 switch (*str) {
823 case 'b':
824 case 'B':
825 base = 2;
826 str++;
827 break;
828 case 'o':
829 case 'O':
830 base = 8;
831 str++;
832 break;
833 case 'd':
834 case 'D':
835 case 't':
836 case 'T':
837 base = 10;
838 str++;
839 break;
840 case 'x':
841 case 'X':
842 base = 16;
843 str++;
844 break;
[4ce914d4]845 default:
846 str--;
[30a5470]847 }
848 }
849 } else {
850 /* Check base range */
851 if ((base < 2) || (base > 36)) {
852 *endptr = (char *) str;
853 return EINVAL;
854 }
855 }
856
857 *result = 0;
858 const char *startstr = str;
859
860 while (*str != 0) {
861 unsigned int digit;
862
863 if ((*str >= 'a') && (*str <= 'z'))
864 digit = *str - 'a' + 10;
865 else if ((*str >= 'A') && (*str <= 'Z'))
866 digit = *str - 'A' + 10;
867 else if ((*str >= '0') && (*str <= '9'))
868 digit = *str - '0';
869 else
870 break;
871
872 if (digit >= base)
873 break;
874
875 uint64_t prev = *result;
876 *result = (*result) * base + digit;
877
878 if (*result < prev) {
879 /* Overflow */
880 *endptr = (char *) str;
881 return EOVERFLOW;
882 }
883
884 str++;
885 }
886
887 if (str == startstr) {
888 /*
889 * No digits were decoded => first invalid character is
890 * the first character of the string.
891 */
892 str = nptr;
893 }
894
895 *endptr = (char *) str;
896
897 if (str == nptr)
898 return EINVAL;
899
900 return EOK;
901}
902
903/** Convert string to uint64_t.
904 *
905 * @param nptr Pointer to string.
906 * @param endptr If not NULL, pointer to the first invalid character
907 * is stored here.
908 * @param base Zero or number between 2 and 36 inclusive.
909 * @param strict Do not allow any trailing characters.
[4ce914d4]910 * @param result Result of the conversion.
[30a5470]911 *
912 * @return EOK if conversion was successful.
913 *
914 */
[059a8e4]915int str_uint64_t(const char *nptr, char **endptr, unsigned int base,
[30a5470]916 bool strict, uint64_t *result)
917{
918 ASSERT(result != NULL);
919
920 bool neg;
921 char *lendptr;
922 int ret = str_uint(nptr, &lendptr, base, &neg, result);
923
924 if (endptr != NULL)
925 *endptr = (char *) lendptr;
926
927 if (ret != EOK)
928 return ret;
929
930 /* Do not allow negative values */
931 if (neg)
932 return EINVAL;
933
934 /* Check whether we are at the end of
935 the string in strict mode */
936 if ((strict) && (*lendptr != 0))
937 return EINVAL;
938
939 return EOK;
940}
941
[e535eeb]942void order_suffix(const uint64_t val, uint64_t *rv, char *suffix)
943{
[933cadf]944 if (val > UINT64_C(10000000000000000000)) {
945 *rv = val / UINT64_C(1000000000000000000);
[e535eeb]946 *suffix = 'Z';
[933cadf]947 } else if (val > UINT64_C(1000000000000000000)) {
948 *rv = val / UINT64_C(1000000000000000);
[e535eeb]949 *suffix = 'E';
[933cadf]950 } else if (val > UINT64_C(1000000000000000)) {
951 *rv = val / UINT64_C(1000000000000);
[e535eeb]952 *suffix = 'T';
[933cadf]953 } else if (val > UINT64_C(1000000000000)) {
954 *rv = val / UINT64_C(1000000000);
[e535eeb]955 *suffix = 'G';
[933cadf]956 } else if (val > UINT64_C(1000000000)) {
957 *rv = val / UINT64_C(1000000);
[e535eeb]958 *suffix = 'M';
[933cadf]959 } else if (val > UINT64_C(1000000)) {
960 *rv = val / UINT64_C(1000);
[e535eeb]961 *suffix = 'k';
962 } else {
963 *rv = val;
964 *suffix = ' ';
965 }
966}
967
[933cadf]968void bin_order_suffix(const uint64_t val, uint64_t *rv, const char **suffix,
969 bool fixed)
970{
971 if (val > UINT64_C(1152921504606846976)) {
972 *rv = val / UINT64_C(1125899906842624);
973 *suffix = "EiB";
974 } else if (val > UINT64_C(1125899906842624)) {
975 *rv = val / UINT64_C(1099511627776);
976 *suffix = "TiB";
977 } else if (val > UINT64_C(1099511627776)) {
978 *rv = val / UINT64_C(1073741824);
979 *suffix = "GiB";
980 } else if (val > UINT64_C(1073741824)) {
981 *rv = val / UINT64_C(1048576);
982 *suffix = "MiB";
983 } else if (val > UINT64_C(1048576)) {
984 *rv = val / UINT64_C(1024);
985 *suffix = "KiB";
986 } else {
987 *rv = val;
988 if (fixed)
989 *suffix = "B ";
990 else
991 *suffix = "B";
992 }
993}
994
[16da5f8e]995/** @}
996 */
Note: See TracBrowser for help on using the repository browser.