source: mainline/kernel/generic/src/lib/str.c@ bab75df6

lfn serial ticket/834-toolchain-update topic/msim-upgrade topic/simplify-dev-export
Last change on this file since bab75df6 was bab75df6, checked in by Jiri Svoboda <jiri@…>, 7 years ago

Let kernel code get printf via the standard stdio header. Clean up unused includes.

  • Property mode set to 100644
File size: 24.1 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
[174156fd]29/** @addtogroup kernel_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 <cpu.h>
106#include <arch/asm.h>
107#include <arch.h>
[d09f84e6]108#include <errno.h>
[b888d5f]109#include <align.h>
[63e27ef]110#include <assert.h>
[30a5470]111#include <macros.h>
[1066041]112#include <mm/slab.h>
[16da5f8e]113
[8e893ae]114/** Check the condition if wchar_t is signed */
[002fd5f]115#ifdef __WCHAR_UNSIGNED__
[1433ecda]116#define WCHAR_SIGNED_CHECK(cond) (true)
[8e893ae]117#else
[1433ecda]118#define WCHAR_SIGNED_CHECK(cond) (cond)
[8e893ae]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;
[a35b458]152
[b888d5f]153 /* First byte read from string */
154 uint8_t b0 = (uint8_t) str[(*offset)++];
[a35b458]155
[b888d5f]156 /* Determine code length */
[a35b458]157
[b888d5f]158 unsigned int b0_bits; /* Data bits in first byte */
159 unsigned int cbytes; /* Number of continuation bytes */
[a35b458]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 }
[a35b458]181
[b888d5f]182 if (*offset + cbytes > size)
[c8bf88d]183 return U_SPECIAL;
[a35b458]184
[b888d5f]185 wchar_t ch = b0 & LO_MASK_8(b0_bits);
[a35b458]186
[b888d5f]187 /* Decode continuation bytes */
[0dd1d444]188 while (cbytes > 0) {
[b888d5f]189 uint8_t b = (uint8_t) str[(*offset)++];
[a35b458]190
[b888d5f]191 /* Must be 10xxxxxx */
192 if ((b & 0xc0) != 0x80)
[c8bf88d]193 return U_SPECIAL;
[a35b458]194
[b888d5f]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 }
[a35b458]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 */
[b7fd2a0]218errno_t 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;
[a35b458]222
[b888d5f]223 if (!chr_check(ch))
[d09f84e6]224 return EINVAL;
[a35b458]225
[7c3fb9b]226 /*
227 * Unsigned version of ch (bit operations should only be done
228 * on unsigned types).
229 */
[b888d5f]230 uint32_t cc = (uint32_t) ch;
[a35b458]231
[b888d5f]232 /* Determine how many continuation bytes are needed */
[a35b458]233
[b888d5f]234 unsigned int b0_bits; /* Data bits in first byte */
235 unsigned int cbytes; /* Number of continuation bytes */
[a35b458]236
[32704cb]237 if ((cc & ~LO_MASK_32(7)) == 0) {
238 b0_bits = 7;
239 cbytes = 0;
240 } else if ((cc & ~LO_MASK_32(11)) == 0) {
241 b0_bits = 5;
242 cbytes = 1;
243 } else if ((cc & ~LO_MASK_32(16)) == 0) {
244 b0_bits = 4;
245 cbytes = 2;
246 } else if ((cc & ~LO_MASK_32(21)) == 0) {
247 b0_bits = 3;
248 cbytes = 3;
249 } else {
[b888d5f]250 /* Codes longer than 21 bits are not supported */
[d09f84e6]251 return EINVAL;
[74c8da2c]252 }
[a35b458]253
[b888d5f]254 /* Check for available space in buffer */
255 if (*offset + cbytes >= size)
[d09f84e6]256 return EOVERFLOW;
[a35b458]257
[b888d5f]258 /* Encode continuation bytes */
259 unsigned int i;
260 for (i = cbytes; i > 0; i--) {
[e1813cf]261 str[*offset + i] = 0x80 | (cc & LO_MASK_32(CONT_BITS));
[32704cb]262 cc = cc >> CONT_BITS;
[74c8da2c]263 }
[a35b458]264
[b888d5f]265 /* Encode first byte */
[e1813cf]266 str[*offset] = (cc & LO_MASK_32(b0_bits)) | HI_MASK_8(8 - b0_bits - 1);
[a35b458]267
[b888d5f]268 /* Advance offset */
269 *offset += cbytes + 1;
[a35b458]270
[d09f84e6]271 return EOK;
[74c8da2c]272}
273
[b888d5f]274/** Get size of string.
275 *
276 * Get the number of bytes which are used by the string @a str (excluding the
277 * NULL-terminator).
278 *
279 * @param str String to consider.
280 *
281 * @return Number of bytes used by the string
[82bb9c1]282 *
283 */
[b888d5f]284size_t str_size(const char *str)
[82bb9c1]285{
[b888d5f]286 size_t size = 0;
[a35b458]287
[b888d5f]288 while (*str++ != 0)
289 size++;
[a35b458]290
[b888d5f]291 return size;
[82bb9c1]292}
293
[b888d5f]294/** Get size of wide string.
295 *
296 * Get the number of bytes which are used by the wide string @a str (excluding the
297 * NULL-terminator).
298 *
299 * @param str Wide string to consider.
300 *
301 * @return Number of bytes used by the wide string
302 *
303 */
304size_t wstr_size(const wchar_t *str)
305{
306 return (wstr_length(str) * sizeof(wchar_t));
307}
308
309/** Get size of string with length limit.
[74c8da2c]310 *
[f25b2819]311 * Get the number of bytes which are used by up to @a max_len first
312 * characters in the string @a str. If @a max_len is greater than
[b888d5f]313 * the length of @a str, the entire string is measured (excluding the
314 * NULL-terminator).
315 *
316 * @param str String to consider.
317 * @param max_len Maximum number of characters to measure.
[74c8da2c]318 *
[b888d5f]319 * @return Number of bytes used by the characters.
[74c8da2c]320 *
321 */
[98000fb]322size_t str_lsize(const char *str, size_t max_len)
[74c8da2c]323{
[98000fb]324 size_t len = 0;
[b888d5f]325 size_t offset = 0;
[a35b458]326
[b888d5f]327 while (len < max_len) {
328 if (str_decode(str, &offset, STR_NO_LIMIT) == 0)
[b54d2f1]329 break;
[a35b458]330
[f25b2819]331 len++;
[21a639b7]332 }
[a35b458]333
[b888d5f]334 return offset;
[74c8da2c]335}
336
[b888d5f]337/** Get size of wide string with length limit.
[82bb9c1]338 *
[b888d5f]339 * Get the number of bytes which are used by up to @a max_len first
340 * wide characters in the wide string @a str. If @a max_len is greater than
341 * the length of @a str, the entire wide string is measured (excluding the
342 * NULL-terminator).
343 *
344 * @param str Wide string to consider.
345 * @param max_len Maximum number of wide characters to measure.
[82bb9c1]346 *
[b888d5f]347 * @return Number of bytes used by the wide characters.
[82bb9c1]348 *
349 */
[98000fb]350size_t wstr_lsize(const wchar_t *str, size_t max_len)
[82bb9c1]351{
[b888d5f]352 return (wstr_nlength(str, max_len * sizeof(wchar_t)) * sizeof(wchar_t));
[82bb9c1]353}
354
[b888d5f]355/** Get number of characters in a string.
[82bb9c1]356 *
[b888d5f]357 * @param str NULL-terminated string.
[82bb9c1]358 *
[b888d5f]359 * @return Number of characters in string.
[82bb9c1]360 *
361 */
[98000fb]362size_t str_length(const char *str)
[82bb9c1]363{
[98000fb]364 size_t len = 0;
[b888d5f]365 size_t offset = 0;
[a35b458]366
[b888d5f]367 while (str_decode(str, &offset, STR_NO_LIMIT) != 0)
368 len++;
[a35b458]369
[b888d5f]370 return len;
[82bb9c1]371}
372
[b888d5f]373/** Get number of characters in a wide string.
[74c8da2c]374 *
[b888d5f]375 * @param str NULL-terminated wide string.
376 *
377 * @return Number of characters in @a str.
[74c8da2c]378 *
379 */
[98000fb]380size_t wstr_length(const wchar_t *wstr)
[74c8da2c]381{
[98000fb]382 size_t len = 0;
[a35b458]383
[b888d5f]384 while (*wstr++ != 0)
385 len++;
[a35b458]386
[b888d5f]387 return len;
[74c8da2c]388}
389
[b888d5f]390/** Get number of characters in a string with size limit.
391 *
392 * @param str NULL-terminated string.
393 * @param size Maximum number of bytes to consider.
394 *
395 * @return Number of characters in string.
[74c8da2c]396 *
397 */
[98000fb]398size_t str_nlength(const char *str, size_t size)
[74c8da2c]399{
[98000fb]400 size_t len = 0;
[b888d5f]401 size_t offset = 0;
[a35b458]402
[b888d5f]403 while (str_decode(str, &offset, size) != 0)
404 len++;
[a35b458]405
[b888d5f]406 return len;
[21a639b7]407}
408
[b888d5f]409/** Get number of characters in a string with size limit.
[2f57690]410 *
[b888d5f]411 * @param str NULL-terminated string.
412 * @param size Maximum number of bytes to consider.
[74c8da2c]413 *
[f25b2819]414 * @return Number of characters in string.
[b888d5f]415 *
[74c8da2c]416 */
[98000fb]417size_t wstr_nlength(const wchar_t *str, size_t size)
[74c8da2c]418{
[98000fb]419 size_t len = 0;
420 size_t limit = ALIGN_DOWN(size, sizeof(wchar_t));
421 size_t offset = 0;
[a35b458]422
[b888d5f]423 while ((offset < limit) && (*str++ != 0)) {
[f25b2819]424 len++;
[b888d5f]425 offset += sizeof(wchar_t);
[74c8da2c]426 }
[a35b458]427
[f25b2819]428 return len;
[74c8da2c]429}
430
[b888d5f]431/** Check whether character is plain ASCII.
432 *
433 * @return True if character is plain ASCII.
[74c8da2c]434 *
435 */
[f2b8cdc]436bool ascii_check(wchar_t ch)
[74c8da2c]437{
[8e893ae]438 if (WCHAR_SIGNED_CHECK(ch >= 0) && (ch <= 127))
[b888d5f]439 return true;
[a35b458]440
[b888d5f]441 return false;
442}
[f25b2819]443
[b888d5f]444/** Check whether character is valid
445 *
446 * @return True if character is a valid Unicode code point.
447 *
448 */
[f2b8cdc]449bool chr_check(wchar_t ch)
[b888d5f]450{
[8e893ae]451 if (WCHAR_SIGNED_CHECK(ch >= 0) && (ch <= 1114111))
[b888d5f]452 return true;
[a35b458]453
[b888d5f]454 return false;
[16da5f8e]455}
456
[b888d5f]457/** Compare two NULL terminated strings.
[16da5f8e]458 *
[b888d5f]459 * Do a char-by-char comparison of two NULL-terminated strings.
[4efeab5]460 * The strings are considered equal iff their length is equal
461 * and both strings consist of the same sequence of characters.
462 *
[1772e6d]463 * A string S1 is less than another string S2 if it has a character with
464 * lower value at the first character position where the strings differ.
465 * If the strings differ in length, the shorter one is treated as if
466 * padded by characters with a value of zero.
[16da5f8e]467 *
[b888d5f]468 * @param s1 First string to compare.
469 * @param s2 Second string to compare.
[16da5f8e]470 *
[1772e6d]471 * @return 0 if the strings are equal, -1 if the first is less than the second,
472 * 1 if the second is less than the first.
[16da5f8e]473 *
474 */
[b888d5f]475int str_cmp(const char *s1, const char *s2)
[16da5f8e]476{
[a7b1071]477 wchar_t c1 = 0;
478 wchar_t c2 = 0;
[a35b458]479
[b888d5f]480 size_t off1 = 0;
481 size_t off2 = 0;
[a7b1071]482
483 while (true) {
484 c1 = str_decode(s1, &off1, STR_NO_LIMIT);
485 c2 = str_decode(s2, &off2, STR_NO_LIMIT);
486
[b888d5f]487 if (c1 < c2)
[16da5f8e]488 return -1;
[a35b458]489
[b888d5f]490 if (c1 > c2)
[16da5f8e]491 return 1;
[a7b1071]492
493 if (c1 == 0 || c2 == 0)
[1b20da0]494 break;
[16da5f8e]495 }
[a7b1071]496
497 return 0;
[16da5f8e]498}
499
[b888d5f]500/** Compare two NULL terminated strings with length limit.
[16da5f8e]501 *
[b888d5f]502 * Do a char-by-char comparison of two NULL-terminated strings.
[4efeab5]503 * The strings are considered equal iff
504 * min(str_length(s1), max_len) == min(str_length(s2), max_len)
505 * and both strings consist of the same sequence of characters,
506 * up to max_len characters.
507 *
[1772e6d]508 * A string S1 is less than another string S2 if it has a character with
509 * lower value at the first character position where the strings differ.
510 * If the strings differ in length, the shorter one is treated as if
511 * padded by characters with a value of zero. Only the first max_len
512 * characters are considered.
[16da5f8e]513 *
[b888d5f]514 * @param s1 First string to compare.
515 * @param s2 Second string to compare.
516 * @param max_len Maximum number of characters to consider.
517 *
[1772e6d]518 * @return 0 if the strings are equal, -1 if the first is less than the second,
519 * 1 if the second is less than the first.
[16da5f8e]520 *
521 */
[98000fb]522int str_lcmp(const char *s1, const char *s2, size_t max_len)
[16da5f8e]523{
[b888d5f]524 wchar_t c1 = 0;
525 wchar_t c2 = 0;
[a35b458]526
[b888d5f]527 size_t off1 = 0;
528 size_t off2 = 0;
[a35b458]529
[98000fb]530 size_t len = 0;
[a7b1071]531
532 while (true) {
533 if (len >= max_len)
[b888d5f]534 break;
[a7b1071]535
536 c1 = str_decode(s1, &off1, STR_NO_LIMIT);
537 c2 = str_decode(s2, &off2, STR_NO_LIMIT);
538
[b888d5f]539 if (c1 < c2)
[16da5f8e]540 return -1;
[a7b1071]541
[b888d5f]542 if (c1 > c2)
[16da5f8e]543 return 1;
[a7b1071]544
545 if (c1 == 0 || c2 == 0)
546 break;
547
[1b20da0]548 ++len;
[16da5f8e]549 }
[a7b1071]550
551 return 0;
552
[16da5f8e]553}
554
[f4b1535]555/** Copy string.
[b888d5f]556 *
[f4b1535]557 * Copy source string @a src to destination buffer @a dest.
558 * No more than @a size bytes are written. If the size of the output buffer
559 * is at least one byte, the output string will always be well-formed, i.e.
560 * null-terminated and containing only complete characters.
[b888d5f]561 *
[abf09311]562 * @param dest Destination buffer.
[6700ee2]563 * @param count Size of the destination buffer (must be > 0).
[f4b1535]564 * @param src Source string.
[abf09311]565 *
[b888d5f]566 */
[f4b1535]567void str_cpy(char *dest, size_t size, const char *src)
[b888d5f]568{
[6700ee2]569 /* There must be space for a null terminator in the buffer. */
[63e27ef]570 assert(size > 0);
571 assert(src != NULL);
[a35b458]572
[abf09311]573 size_t src_off = 0;
574 size_t dest_off = 0;
[a35b458]575
[abf09311]576 wchar_t ch;
[f4b1535]577 while ((ch = str_decode(src, &src_off, STR_NO_LIMIT)) != 0) {
578 if (chr_encode(ch, dest, &dest_off, size - 1) != EOK)
579 break;
580 }
[a35b458]581
[f4b1535]582 dest[dest_off] = '\0';
583}
584
585/** Copy size-limited substring.
586 *
[6700ee2]587 * Copy prefix of string @a src of max. size @a size to destination buffer
588 * @a dest. No more than @a size bytes are written. The output string will
589 * always be well-formed, i.e. null-terminated and containing only complete
590 * characters.
[f4b1535]591 *
592 * No more than @a n bytes are read from the input string, so it does not
593 * have to be null-terminated.
594 *
[abf09311]595 * @param dest Destination buffer.
[6700ee2]596 * @param count Size of the destination buffer (must be > 0).
[f4b1535]597 * @param src Source string.
[abf09311]598 * @param n Maximum number of bytes to read from @a src.
599 *
[f4b1535]600 */
601void str_ncpy(char *dest, size_t size, const char *src, size_t n)
602{
[6700ee2]603 /* There must be space for a null terminator in the buffer. */
[63e27ef]604 assert(size > 0);
[a35b458]605
[abf09311]606 size_t src_off = 0;
607 size_t dest_off = 0;
[a35b458]608
[abf09311]609 wchar_t ch;
[f4b1535]610 while ((ch = str_decode(src, &src_off, n)) != 0) {
611 if (chr_encode(ch, dest, &dest_off, size - 1) != EOK)
[b888d5f]612 break;
613 }
[a35b458]614
[f4b1535]615 dest[dest_off] = '\0';
[b888d5f]616}
[16da5f8e]617
[abf09311]618/** Duplicate string.
619 *
620 * Allocate a new string and copy characters from the source
621 * string into it. The duplicate string is allocated via sleeping
622 * malloc(), thus this function can sleep in no memory conditions.
623 *
624 * The allocation cannot fail and the return value is always
625 * a valid pointer. The duplicate string is always a well-formed
626 * null-terminated UTF-8 string, but it can differ from the source
627 * string on the byte level.
628 *
629 * @param src Source string.
630 *
631 * @return Duplicate string.
632 *
633 */
634char *str_dup(const char *src)
635{
636 size_t size = str_size(src) + 1;
[11b285d]637 char *dest = nfmalloc(size);
[63e27ef]638 assert(dest);
[a35b458]639
[abf09311]640 str_cpy(dest, size, src);
641 return dest;
642}
643
644/** Duplicate string with size limit.
645 *
646 * Allocate a new string and copy up to @max_size bytes from the source
647 * string into it. The duplicate string is allocated via sleeping
648 * malloc(), thus this function can sleep in no memory conditions.
649 * No more than @max_size + 1 bytes is allocated, but if the size
650 * occupied by the source string is smaller than @max_size + 1,
651 * less is allocated.
652 *
653 * The allocation cannot fail and the return value is always
654 * a valid pointer. The duplicate string is always a well-formed
655 * null-terminated UTF-8 string, but it can differ from the source
656 * string on the byte level.
657 *
658 * @param src Source string.
659 * @param n Maximum number of bytes to duplicate.
660 *
661 * @return Duplicate string.
662 *
663 */
664char *str_ndup(const char *src, size_t n)
665{
666 size_t size = str_size(src);
667 if (size > n)
668 size = n;
[a35b458]669
[11b285d]670 char *dest = nfmalloc(size + 1);
[63e27ef]671 assert(dest);
[a35b458]672
[abf09311]673 str_ncpy(dest, size + 1, src, size);
674 return dest;
675}
676
[0f06dbc]677/** Convert wide string to string.
[b888d5f]678 *
[0f06dbc]679 * Convert wide string @a src to string. The output is written to the buffer
680 * specified by @a dest and @a size. @a size must be non-zero and the string
681 * written will always be well-formed.
[16da5f8e]682 *
[0f06dbc]683 * @param dest Destination buffer.
684 * @param size Size of the destination buffer.
685 * @param src Source wide string.
[16da5f8e]686 */
[0f06dbc]687void wstr_to_str(char *dest, size_t size, const wchar_t *src)
[16da5f8e]688{
[b888d5f]689 wchar_t ch;
[0f06dbc]690 size_t src_idx;
691 size_t dest_off;
692
693 /* There must be space for a null terminator in the buffer. */
[63e27ef]694 assert(size > 0);
[0f06dbc]695
696 src_idx = 0;
697 dest_off = 0;
[a35b458]698
[b888d5f]699 while ((ch = src[src_idx++]) != 0) {
[0f06dbc]700 if (chr_encode(ch, dest, &dest_off, size - 1) != EOK)
[b888d5f]701 break;
[16da5f8e]702 }
[0f06dbc]703
704 dest[dest_off] = '\0';
[16da5f8e]705}
706
[20f1597]707/** Find first occurence of character in string.
708 *
[b888d5f]709 * @param str String to search.
710 * @param ch Character to look for.
711 *
712 * @return Pointer to character in @a str or NULL if not found.
[20f1597]713 *
714 */
[dd2cfa7]715char *str_chr(const char *str, wchar_t ch)
[20f1597]716{
[b888d5f]717 wchar_t acc;
718 size_t off = 0;
[f2d2c7ba]719 size_t last = 0;
[a35b458]720
[a7b1071]721 while ((acc = str_decode(str, &off, STR_NO_LIMIT)) != 0) {
[b888d5f]722 if (acc == ch)
[dd2cfa7]723 return (char *) (str + last);
[f2d2c7ba]724 last = off;
[20f1597]725 }
[a35b458]726
[20f1597]727 return NULL;
728}
729
[b888d5f]730/** Insert a wide character into a wide string.
731 *
732 * Insert a wide character into a wide string at position
733 * @a pos. The characters after the position are shifted.
734 *
735 * @param str String to insert to.
736 * @param ch Character to insert to.
737 * @param pos Character index where to insert.
[7c3fb9b]738 * @param max_pos Characters in the buffer.
[b888d5f]739 *
740 * @return True if the insertion was sucessful, false if the position
741 * is out of bounds.
742 *
743 */
[98000fb]744bool wstr_linsert(wchar_t *str, wchar_t ch, size_t pos, size_t max_pos)
[b888d5f]745{
[98000fb]746 size_t len = wstr_length(str);
[a35b458]747
[b888d5f]748 if ((pos > len) || (pos + 1 > max_pos))
749 return false;
[a35b458]750
[98000fb]751 size_t i;
[b888d5f]752 for (i = len; i + 1 > pos; i--)
753 str[i + 1] = str[i];
[a35b458]754
[b888d5f]755 str[pos] = ch;
[a35b458]756
[b888d5f]757 return true;
758}
759
760/** Remove a wide character from a wide string.
761 *
762 * Remove a wide character from a wide string at position
763 * @a pos. The characters after the position are shifted.
764 *
765 * @param str String to remove from.
766 * @param pos Character index to remove.
767 *
768 * @return True if the removal was sucessful, false if the position
769 * is out of bounds.
770 *
771 */
[98000fb]772bool wstr_remove(wchar_t *str, size_t pos)
[b888d5f]773{
[98000fb]774 size_t len = wstr_length(str);
[a35b458]775
[b888d5f]776 if (pos >= len)
777 return false;
[a35b458]778
[98000fb]779 size_t i;
[b888d5f]780 for (i = pos + 1; i <= len; i++)
781 str[i - 1] = str[i];
[a35b458]782
[b888d5f]783 return true;
784}
785
[30a5470]786/** Convert string to uint64_t (internal variant).
787 *
788 * @param nptr Pointer to string.
789 * @param endptr Pointer to the first invalid character is stored here.
790 * @param base Zero or number between 2 and 36 inclusive.
791 * @param neg Indication of unary minus is stored here.
792 * @apram result Result of the conversion.
793 *
794 * @return EOK if conversion was successful.
795 *
796 */
[b7fd2a0]797static errno_t str_uint(const char *nptr, char **endptr, unsigned int base,
[30a5470]798 bool *neg, uint64_t *result)
799{
[63e27ef]800 assert(endptr != NULL);
801 assert(neg != NULL);
802 assert(result != NULL);
[a35b458]803
[30a5470]804 *neg = false;
805 const char *str = nptr;
[a35b458]806
[30a5470]807 /* Ignore leading whitespace */
808 while (isspace(*str))
809 str++;
[a35b458]810
[30a5470]811 if (*str == '-') {
812 *neg = true;
813 str++;
814 } else if (*str == '+')
815 str++;
[a35b458]816
[30a5470]817 if (base == 0) {
818 /* Decode base if not specified */
819 base = 10;
[a35b458]820
[30a5470]821 if (*str == '0') {
822 base = 8;
823 str++;
[a35b458]824
[30a5470]825 switch (*str) {
826 case 'b':
827 case 'B':
828 base = 2;
829 str++;
830 break;
831 case 'o':
832 case 'O':
833 base = 8;
834 str++;
835 break;
836 case 'd':
837 case 'D':
838 case 't':
839 case 'T':
840 base = 10;
841 str++;
842 break;
843 case 'x':
844 case 'X':
845 base = 16;
846 str++;
847 break;
[4ce914d4]848 default:
849 str--;
[30a5470]850 }
851 }
852 } else {
853 /* Check base range */
854 if ((base < 2) || (base > 36)) {
855 *endptr = (char *) str;
856 return EINVAL;
857 }
858 }
[a35b458]859
[30a5470]860 *result = 0;
861 const char *startstr = str;
[a35b458]862
[30a5470]863 while (*str != 0) {
864 unsigned int digit;
[a35b458]865
[30a5470]866 if ((*str >= 'a') && (*str <= 'z'))
867 digit = *str - 'a' + 10;
868 else if ((*str >= 'A') && (*str <= 'Z'))
869 digit = *str - 'A' + 10;
870 else if ((*str >= '0') && (*str <= '9'))
871 digit = *str - '0';
872 else
873 break;
[a35b458]874
[30a5470]875 if (digit >= base)
876 break;
[a35b458]877
[30a5470]878 uint64_t prev = *result;
879 *result = (*result) * base + digit;
[a35b458]880
[30a5470]881 if (*result < prev) {
882 /* Overflow */
883 *endptr = (char *) str;
884 return EOVERFLOW;
885 }
[a35b458]886
[30a5470]887 str++;
888 }
[a35b458]889
[30a5470]890 if (str == startstr) {
891 /*
892 * No digits were decoded => first invalid character is
893 * the first character of the string.
894 */
895 str = nptr;
896 }
[a35b458]897
[30a5470]898 *endptr = (char *) str;
[a35b458]899
[30a5470]900 if (str == nptr)
901 return EINVAL;
[a35b458]902
[30a5470]903 return EOK;
904}
905
906/** Convert string to uint64_t.
907 *
908 * @param nptr Pointer to string.
909 * @param endptr If not NULL, pointer to the first invalid character
910 * is stored here.
911 * @param base Zero or number between 2 and 36 inclusive.
912 * @param strict Do not allow any trailing characters.
[4ce914d4]913 * @param result Result of the conversion.
[30a5470]914 *
915 * @return EOK if conversion was successful.
916 *
917 */
[b7fd2a0]918errno_t str_uint64_t(const char *nptr, char **endptr, unsigned int base,
[30a5470]919 bool strict, uint64_t *result)
920{
[63e27ef]921 assert(result != NULL);
[a35b458]922
[30a5470]923 bool neg;
924 char *lendptr;
[b7fd2a0]925 errno_t ret = str_uint(nptr, &lendptr, base, &neg, result);
[a35b458]926
[30a5470]927 if (endptr != NULL)
928 *endptr = (char *) lendptr;
[a35b458]929
[30a5470]930 if (ret != EOK)
931 return ret;
[a35b458]932
[30a5470]933 /* Do not allow negative values */
934 if (neg)
935 return EINVAL;
[a35b458]936
[7c3fb9b]937 /*
938 * Check whether we are at the end of
939 * the string in strict mode
940 */
[30a5470]941 if ((strict) && (*lendptr != 0))
942 return EINVAL;
[a35b458]943
[30a5470]944 return EOK;
945}
946
[e535eeb]947void order_suffix(const uint64_t val, uint64_t *rv, char *suffix)
948{
[933cadf]949 if (val > UINT64_C(10000000000000000000)) {
950 *rv = val / UINT64_C(1000000000000000000);
[e535eeb]951 *suffix = 'Z';
[933cadf]952 } else if (val > UINT64_C(1000000000000000000)) {
953 *rv = val / UINT64_C(1000000000000000);
[e535eeb]954 *suffix = 'E';
[933cadf]955 } else if (val > UINT64_C(1000000000000000)) {
956 *rv = val / UINT64_C(1000000000000);
[e535eeb]957 *suffix = 'T';
[933cadf]958 } else if (val > UINT64_C(1000000000000)) {
959 *rv = val / UINT64_C(1000000000);
[e535eeb]960 *suffix = 'G';
[933cadf]961 } else if (val > UINT64_C(1000000000)) {
962 *rv = val / UINT64_C(1000000);
[e535eeb]963 *suffix = 'M';
[933cadf]964 } else if (val > UINT64_C(1000000)) {
965 *rv = val / UINT64_C(1000);
[e535eeb]966 *suffix = 'k';
967 } else {
968 *rv = val;
969 *suffix = ' ';
970 }
971}
972
[933cadf]973void bin_order_suffix(const uint64_t val, uint64_t *rv, const char **suffix,
974 bool fixed)
975{
976 if (val > UINT64_C(1152921504606846976)) {
977 *rv = val / UINT64_C(1125899906842624);
978 *suffix = "EiB";
979 } else if (val > UINT64_C(1125899906842624)) {
980 *rv = val / UINT64_C(1099511627776);
981 *suffix = "TiB";
982 } else if (val > UINT64_C(1099511627776)) {
983 *rv = val / UINT64_C(1073741824);
984 *suffix = "GiB";
985 } else if (val > UINT64_C(1073741824)) {
986 *rv = val / UINT64_C(1048576);
987 *suffix = "MiB";
988 } else if (val > UINT64_C(1048576)) {
989 *rv = val / UINT64_C(1024);
990 *suffix = "KiB";
991 } else {
992 *rv = val;
993 if (fixed)
994 *suffix = "B ";
995 else
996 *suffix = "B";
997 }
998}
999
[16da5f8e]1000/** @}
1001 */
Note: See TracBrowser for help on using the repository browser.