source: mainline/kernel/generic/src/lib/str.c@ 13d5639

lfn serial ticket/834-toolchain-update topic/msim-upgrade topic/simplify-dev-export
Last change on this file since 13d5639 was 8e893ae, checked in by Martin Decky <martin@…>, 13 years ago

avoid comparison with 0 if the type is unsigned

  • Property mode set to 100644
File size: 23.5 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.
[16da5f8e]458 * The strings are considered equal iff they consist of the same
459 * characters on the minimum of their lengths.
460 *
[b888d5f]461 * @param s1 First string to compare.
462 * @param s2 Second string to compare.
[16da5f8e]463 *
[b888d5f]464 * @return 0 if the strings are equal, -1 if first is smaller,
465 * 1 if second smaller.
[16da5f8e]466 *
467 */
[b888d5f]468int str_cmp(const char *s1, const char *s2)
[16da5f8e]469{
[a7b1071]470 wchar_t c1 = 0;
471 wchar_t c2 = 0;
[b888d5f]472
473 size_t off1 = 0;
474 size_t off2 = 0;
[a7b1071]475
476 while (true) {
477 c1 = str_decode(s1, &off1, STR_NO_LIMIT);
478 c2 = str_decode(s2, &off2, STR_NO_LIMIT);
479
[b888d5f]480 if (c1 < c2)
[16da5f8e]481 return -1;
[b888d5f]482
483 if (c1 > c2)
[16da5f8e]484 return 1;
[a7b1071]485
486 if (c1 == 0 || c2 == 0)
487 break;
[16da5f8e]488 }
[a7b1071]489
490 return 0;
[16da5f8e]491}
492
[b888d5f]493/** Compare two NULL terminated strings with length limit.
[16da5f8e]494 *
[b888d5f]495 * Do a char-by-char comparison of two NULL-terminated strings.
[16da5f8e]496 * The strings are considered equal iff they consist of the same
[b888d5f]497 * characters on the minimum of their lengths and the length limit.
[16da5f8e]498 *
[b888d5f]499 * @param s1 First string to compare.
500 * @param s2 Second string to compare.
501 * @param max_len Maximum number of characters to consider.
502 *
503 * @return 0 if the strings are equal, -1 if first is smaller,
504 * 1 if second smaller.
[16da5f8e]505 *
506 */
[98000fb]507int str_lcmp(const char *s1, const char *s2, size_t max_len)
[16da5f8e]508{
[b888d5f]509 wchar_t c1 = 0;
510 wchar_t c2 = 0;
[16da5f8e]511
[b888d5f]512 size_t off1 = 0;
513 size_t off2 = 0;
514
[98000fb]515 size_t len = 0;
[a7b1071]516
517 while (true) {
518 if (len >= max_len)
[b888d5f]519 break;
[a7b1071]520
521 c1 = str_decode(s1, &off1, STR_NO_LIMIT);
522 c2 = str_decode(s2, &off2, STR_NO_LIMIT);
523
[b888d5f]524 if (c1 < c2)
[16da5f8e]525 return -1;
[a7b1071]526
[b888d5f]527 if (c1 > c2)
[16da5f8e]528 return 1;
[a7b1071]529
530 if (c1 == 0 || c2 == 0)
531 break;
532
533 ++len;
[16da5f8e]534 }
[a7b1071]535
536 return 0;
537
[16da5f8e]538}
539
[f4b1535]540/** Copy string.
[b888d5f]541 *
[f4b1535]542 * Copy source string @a src to destination buffer @a dest.
543 * No more than @a size bytes are written. If the size of the output buffer
544 * is at least one byte, the output string will always be well-formed, i.e.
545 * null-terminated and containing only complete characters.
[b888d5f]546 *
[abf09311]547 * @param dest Destination buffer.
[6700ee2]548 * @param count Size of the destination buffer (must be > 0).
[f4b1535]549 * @param src Source string.
[abf09311]550 *
[b888d5f]551 */
[f4b1535]552void str_cpy(char *dest, size_t size, const char *src)
[b888d5f]553{
[6700ee2]554 /* There must be space for a null terminator in the buffer. */
555 ASSERT(size > 0);
[b888d5f]556
[abf09311]557 size_t src_off = 0;
558 size_t dest_off = 0;
559
560 wchar_t ch;
[f4b1535]561 while ((ch = str_decode(src, &src_off, STR_NO_LIMIT)) != 0) {
562 if (chr_encode(ch, dest, &dest_off, size - 1) != EOK)
563 break;
564 }
[abf09311]565
[f4b1535]566 dest[dest_off] = '\0';
567}
568
569/** Copy size-limited substring.
570 *
[6700ee2]571 * Copy prefix of string @a src of max. size @a size to destination buffer
572 * @a dest. No more than @a size bytes are written. The output string will
573 * always be well-formed, i.e. null-terminated and containing only complete
574 * characters.
[f4b1535]575 *
576 * No more than @a n bytes are read from the input string, so it does not
577 * have to be null-terminated.
578 *
[abf09311]579 * @param dest Destination buffer.
[6700ee2]580 * @param count Size of the destination buffer (must be > 0).
[f4b1535]581 * @param src Source string.
[abf09311]582 * @param n Maximum number of bytes to read from @a src.
583 *
[f4b1535]584 */
585void str_ncpy(char *dest, size_t size, const char *src, size_t n)
586{
[6700ee2]587 /* There must be space for a null terminator in the buffer. */
588 ASSERT(size > 0);
[b888d5f]589
[abf09311]590 size_t src_off = 0;
591 size_t dest_off = 0;
592
593 wchar_t ch;
[f4b1535]594 while ((ch = str_decode(src, &src_off, n)) != 0) {
595 if (chr_encode(ch, dest, &dest_off, size - 1) != EOK)
[b888d5f]596 break;
597 }
[abf09311]598
[f4b1535]599 dest[dest_off] = '\0';
[b888d5f]600}
[16da5f8e]601
[abf09311]602/** Duplicate string.
603 *
604 * Allocate a new string and copy characters from the source
605 * string into it. The duplicate string is allocated via sleeping
606 * malloc(), thus this function can sleep in no memory conditions.
607 *
608 * The allocation cannot fail and the return value is always
609 * a valid pointer. The duplicate string is always a well-formed
610 * null-terminated UTF-8 string, but it can differ from the source
611 * string on the byte level.
612 *
613 * @param src Source string.
614 *
615 * @return Duplicate string.
616 *
617 */
618char *str_dup(const char *src)
619{
620 size_t size = str_size(src) + 1;
621 char *dest = malloc(size, 0);
622 ASSERT(dest);
623
624 str_cpy(dest, size, src);
625 return dest;
626}
627
628/** Duplicate string with size limit.
629 *
630 * Allocate a new string and copy up to @max_size bytes from the source
631 * string into it. The duplicate string is allocated via sleeping
632 * malloc(), thus this function can sleep in no memory conditions.
633 * No more than @max_size + 1 bytes is allocated, but if the size
634 * occupied by the source string is smaller than @max_size + 1,
635 * less is allocated.
636 *
637 * The allocation cannot fail and the return value is always
638 * a valid pointer. The duplicate string is always a well-formed
639 * null-terminated UTF-8 string, but it can differ from the source
640 * string on the byte level.
641 *
642 * @param src Source string.
643 * @param n Maximum number of bytes to duplicate.
644 *
645 * @return Duplicate string.
646 *
647 */
648char *str_ndup(const char *src, size_t n)
649{
650 size_t size = str_size(src);
651 if (size > n)
652 size = n;
653
654 char *dest = malloc(size + 1, 0);
655 ASSERT(dest);
656
657 str_ncpy(dest, size + 1, src, size);
658 return dest;
659}
660
[0f06dbc]661/** Convert wide string to string.
[b888d5f]662 *
[0f06dbc]663 * Convert wide string @a src to string. The output is written to the buffer
664 * specified by @a dest and @a size. @a size must be non-zero and the string
665 * written will always be well-formed.
[16da5f8e]666 *
[0f06dbc]667 * @param dest Destination buffer.
668 * @param size Size of the destination buffer.
669 * @param src Source wide string.
[16da5f8e]670 */
[0f06dbc]671void wstr_to_str(char *dest, size_t size, const wchar_t *src)
[16da5f8e]672{
[b888d5f]673 wchar_t ch;
[0f06dbc]674 size_t src_idx;
675 size_t dest_off;
676
677 /* There must be space for a null terminator in the buffer. */
678 ASSERT(size > 0);
679
680 src_idx = 0;
681 dest_off = 0;
[b888d5f]682
683 while ((ch = src[src_idx++]) != 0) {
[0f06dbc]684 if (chr_encode(ch, dest, &dest_off, size - 1) != EOK)
[b888d5f]685 break;
[16da5f8e]686 }
[0f06dbc]687
688 dest[dest_off] = '\0';
[16da5f8e]689}
690
[20f1597]691/** Find first occurence of character in string.
692 *
[b888d5f]693 * @param str String to search.
694 * @param ch Character to look for.
695 *
696 * @return Pointer to character in @a str or NULL if not found.
[20f1597]697 *
698 */
[dd2cfa7]699char *str_chr(const char *str, wchar_t ch)
[20f1597]700{
[b888d5f]701 wchar_t acc;
702 size_t off = 0;
[f2d2c7ba]703 size_t last = 0;
[b888d5f]704
[a7b1071]705 while ((acc = str_decode(str, &off, STR_NO_LIMIT)) != 0) {
[b888d5f]706 if (acc == ch)
[dd2cfa7]707 return (char *) (str + last);
[f2d2c7ba]708 last = off;
[20f1597]709 }
[2f57690]710
[20f1597]711 return NULL;
712}
713
[b888d5f]714/** Insert a wide character into a wide string.
715 *
716 * Insert a wide character into a wide string at position
717 * @a pos. The characters after the position are shifted.
718 *
719 * @param str String to insert to.
720 * @param ch Character to insert to.
721 * @param pos Character index where to insert.
722 @ @param max_pos Characters in the buffer.
723 *
724 * @return True if the insertion was sucessful, false if the position
725 * is out of bounds.
726 *
727 */
[98000fb]728bool wstr_linsert(wchar_t *str, wchar_t ch, size_t pos, size_t max_pos)
[b888d5f]729{
[98000fb]730 size_t len = wstr_length(str);
[b888d5f]731
732 if ((pos > len) || (pos + 1 > max_pos))
733 return false;
734
[98000fb]735 size_t i;
[b888d5f]736 for (i = len; i + 1 > pos; i--)
737 str[i + 1] = str[i];
738
739 str[pos] = ch;
740
741 return true;
742}
743
744/** Remove a wide character from a wide string.
745 *
746 * Remove a wide character from a wide string at position
747 * @a pos. The characters after the position are shifted.
748 *
749 * @param str String to remove from.
750 * @param pos Character index to remove.
751 *
752 * @return True if the removal was sucessful, false if the position
753 * is out of bounds.
754 *
755 */
[98000fb]756bool wstr_remove(wchar_t *str, size_t pos)
[b888d5f]757{
[98000fb]758 size_t len = wstr_length(str);
[b888d5f]759
760 if (pos >= len)
761 return false;
762
[98000fb]763 size_t i;
[b888d5f]764 for (i = pos + 1; i <= len; i++)
765 str[i - 1] = str[i];
766
767 return true;
768}
769
[30a5470]770/** Convert string to uint64_t (internal variant).
771 *
772 * @param nptr Pointer to string.
773 * @param endptr Pointer to the first invalid character is stored here.
774 * @param base Zero or number between 2 and 36 inclusive.
775 * @param neg Indication of unary minus is stored here.
776 * @apram result Result of the conversion.
777 *
778 * @return EOK if conversion was successful.
779 *
780 */
781static int str_uint(const char *nptr, char **endptr, unsigned int base,
782 bool *neg, uint64_t *result)
783{
784 ASSERT(endptr != NULL);
785 ASSERT(neg != NULL);
786 ASSERT(result != NULL);
787
788 *neg = false;
789 const char *str = nptr;
790
791 /* Ignore leading whitespace */
792 while (isspace(*str))
793 str++;
794
795 if (*str == '-') {
796 *neg = true;
797 str++;
798 } else if (*str == '+')
799 str++;
800
801 if (base == 0) {
802 /* Decode base if not specified */
803 base = 10;
804
805 if (*str == '0') {
806 base = 8;
807 str++;
808
809 switch (*str) {
810 case 'b':
811 case 'B':
812 base = 2;
813 str++;
814 break;
815 case 'o':
816 case 'O':
817 base = 8;
818 str++;
819 break;
820 case 'd':
821 case 'D':
822 case 't':
823 case 'T':
824 base = 10;
825 str++;
826 break;
827 case 'x':
828 case 'X':
829 base = 16;
830 str++;
831 break;
[4ce914d4]832 default:
833 str--;
[30a5470]834 }
835 }
836 } else {
837 /* Check base range */
838 if ((base < 2) || (base > 36)) {
839 *endptr = (char *) str;
840 return EINVAL;
841 }
842 }
843
844 *result = 0;
845 const char *startstr = str;
846
847 while (*str != 0) {
848 unsigned int digit;
849
850 if ((*str >= 'a') && (*str <= 'z'))
851 digit = *str - 'a' + 10;
852 else if ((*str >= 'A') && (*str <= 'Z'))
853 digit = *str - 'A' + 10;
854 else if ((*str >= '0') && (*str <= '9'))
855 digit = *str - '0';
856 else
857 break;
858
859 if (digit >= base)
860 break;
861
862 uint64_t prev = *result;
863 *result = (*result) * base + digit;
864
865 if (*result < prev) {
866 /* Overflow */
867 *endptr = (char *) str;
868 return EOVERFLOW;
869 }
870
871 str++;
872 }
873
874 if (str == startstr) {
875 /*
876 * No digits were decoded => first invalid character is
877 * the first character of the string.
878 */
879 str = nptr;
880 }
881
882 *endptr = (char *) str;
883
884 if (str == nptr)
885 return EINVAL;
886
887 return EOK;
888}
889
890/** Convert string to uint64_t.
891 *
892 * @param nptr Pointer to string.
893 * @param endptr If not NULL, pointer to the first invalid character
894 * is stored here.
895 * @param base Zero or number between 2 and 36 inclusive.
896 * @param strict Do not allow any trailing characters.
[4ce914d4]897 * @param result Result of the conversion.
[30a5470]898 *
899 * @return EOK if conversion was successful.
900 *
901 */
[059a8e4]902int str_uint64_t(const char *nptr, char **endptr, unsigned int base,
[30a5470]903 bool strict, uint64_t *result)
904{
905 ASSERT(result != NULL);
906
907 bool neg;
908 char *lendptr;
909 int ret = str_uint(nptr, &lendptr, base, &neg, result);
910
911 if (endptr != NULL)
912 *endptr = (char *) lendptr;
913
914 if (ret != EOK)
915 return ret;
916
917 /* Do not allow negative values */
918 if (neg)
919 return EINVAL;
920
921 /* Check whether we are at the end of
922 the string in strict mode */
923 if ((strict) && (*lendptr != 0))
924 return EINVAL;
925
926 return EOK;
927}
928
[e535eeb]929void order_suffix(const uint64_t val, uint64_t *rv, char *suffix)
930{
[933cadf]931 if (val > UINT64_C(10000000000000000000)) {
932 *rv = val / UINT64_C(1000000000000000000);
[e535eeb]933 *suffix = 'Z';
[933cadf]934 } else if (val > UINT64_C(1000000000000000000)) {
935 *rv = val / UINT64_C(1000000000000000);
[e535eeb]936 *suffix = 'E';
[933cadf]937 } else if (val > UINT64_C(1000000000000000)) {
938 *rv = val / UINT64_C(1000000000000);
[e535eeb]939 *suffix = 'T';
[933cadf]940 } else if (val > UINT64_C(1000000000000)) {
941 *rv = val / UINT64_C(1000000000);
[e535eeb]942 *suffix = 'G';
[933cadf]943 } else if (val > UINT64_C(1000000000)) {
944 *rv = val / UINT64_C(1000000);
[e535eeb]945 *suffix = 'M';
[933cadf]946 } else if (val > UINT64_C(1000000)) {
947 *rv = val / UINT64_C(1000);
[e535eeb]948 *suffix = 'k';
949 } else {
950 *rv = val;
951 *suffix = ' ';
952 }
953}
954
[933cadf]955void bin_order_suffix(const uint64_t val, uint64_t *rv, const char **suffix,
956 bool fixed)
957{
958 if (val > UINT64_C(1152921504606846976)) {
959 *rv = val / UINT64_C(1125899906842624);
960 *suffix = "EiB";
961 } else if (val > UINT64_C(1125899906842624)) {
962 *rv = val / UINT64_C(1099511627776);
963 *suffix = "TiB";
964 } else if (val > UINT64_C(1099511627776)) {
965 *rv = val / UINT64_C(1073741824);
966 *suffix = "GiB";
967 } else if (val > UINT64_C(1073741824)) {
968 *rv = val / UINT64_C(1048576);
969 *suffix = "MiB";
970 } else if (val > UINT64_C(1048576)) {
971 *rv = val / UINT64_C(1024);
972 *suffix = "KiB";
973 } else {
974 *rv = val;
975 if (fixed)
976 *suffix = "B ";
977 else
978 *suffix = "B";
979 }
980}
981
[16da5f8e]982/** @}
983 */
Note: See TracBrowser for help on using the repository browser.