utf8.c 12.7 KB
Newer Older
1 2
/* Synchronet UTF-8 encode/decode/translate functions */

3
/* $Id: utf8.c,v 1.9 2019/08/03 08:05:09 rswindell Exp $ */
4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36

/****************************************************************************
 * @format.tab-size 4		(Plain Text/Source Code File Header)			*
 * @format.use-tabs true	(see http://www.synchro.net/ptsc_hdr.html)		*
 *																			*
 * Copyright Rob Swindell - http://www.synchro.net/copyright.html			*
 *																			*
 * This library is free software; you can redistribute it and/or			*
 * modify it under the terms of the GNU Lesser General Public License		*
 * as published by the Free Software Foundation; either version 2			*
 * of the License, or (at your option) any later version.					*
 * See the GNU Lesser General Public License for more details: lgpl.txt or	*
 * http://www.fsf.org/copyleft/lesser.html									*
 *																			*
 * Anonymous FTP access to the most recent released source is available at	*
 * ftp://vert.synchro.net, ftp://cvs.synchro.net and ftp://ftp.synchro.net	*
 *																			*
 * Anonymous CVS access to the development source and modification history	*
 * is available at cvs.synchro.net:/cvsroot/sbbs, example:					*
 * cvs -d :pserver:anonymous@cvs.synchro.net:/cvsroot/sbbs login			*
 *     (just hit return, no password is necessary)							*
 * cvs -d :pserver:anonymous@cvs.synchro.net:/cvsroot/sbbs checkout src		*
 *																			*
 * For Synchronet coding style and modification guidelines, see				*
 * http://www.synchro.net/source.html										*
 *																			*
 * You are encouraged to submit any modifications (preferably in Unix diff	*
 * format) via e-mail to mods@synchro.net									*
 *																			*
 * Note: If this box doesn't appear square, then you need to fix your tabs.	*
 ****************************************************************************/

#include "utf8.h"
rswindell's avatar
rswindell committed
37
#include "unicode.h"
38 39
#include <stdbool.h>
#include <string.h>
40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141

char* utf8_normalize_str(char* str)
{
	char* dest = str;

	for(char* src = str; *src != 0; src++) {
		if(*src == '\xC2' && *(src + 1) == '\xA0') { // NO-BREAK SPACE
			src++;
			*dest++ = ' ';
			continue;
		}
		if(*src == '\xE2') {
			if(*(src + 1) == '\x80') {
				switch(*(src + 2)) {
					case '\x82': // EN SPACE
					case '\x83': // EM SPACE
						src += 2;
						*dest++ = ' ';
						continue;
					case '\x8B': // ZERO WIDTH SPACE
					case '\x8C': // ZERO WIDTH NON-JOINER
					case '\x8D': // ZERO WIDTH JOINER
						src += 2;
						continue;
					case '\x90': // HYPHEN
					case '\x91': // NON-BREAKING HYPHEN
					case '\x92': // FIGURE DASH
					case '\x93': // EN DASH
						src += 2;
						*dest++ = '-';
						continue;
					case '\x98': // LEFT SINGLE QUOTATION MARK
						src += 2;
						*dest++ = '`';
						continue;
					case '\x99': // RIGHT SINGLE QUOTATION MARK
					case '\xB2': // PRIME
						src += 2;
						*dest++ = '\'';
						continue;
					case '\x9C': // LEFT DOUBLE QUOTATION MARK
					case '\x9D': // RIGHT DOUBLE QUOTATION MARK
						src += 2;
						*dest++ = '"';
						continue;
					case '\xA6': // HORIZONTAL ELLIPSIS -> ASCII periods (3)
						src += 2;
						for(int i = 0; i < 3; i++)
							*dest++ =  '.';
						continue;
				}
			}
			else if(*(src + 1) == '\x81') {
				switch(*(src + 2)) {
					case '\x83': // HYPEN BULLET
						src += 2;
						*dest++ = '-';
						continue;
					case '\x84': // FRACTION SLASH
						src += 2;
						*dest++ = '/';
						continue;
				}
			}
			else if(*(src + 1) == '\x88') {
				switch(*(src + 2)) {
					case '\x92': // MINUS SIGN
						src += 2;
						*dest++ = '-';
						continue;
				}
			}
		}
		else if(*src == '\xEF') {
			if(*(src + 1) == '\xBB' && *(src + 2) == '\xBF') {
				// Zero Width No-Break Space (BOM, ZWNBSP)
				src += 2;
				continue;
			}
			if(*(src + 1) == '\xBC') {
				if(*(src + 2) >= '\x81' && *(src + 2) <= '\xBF') { // FULLWIDTH EXCLAMATION MARK through FULLWIDTH LOW LINE
					src += 2;
					*src -= '\x81';
					*dest++ = '!' + *src;
					continue;
				}
			}
			else if(*(src + 1) == '\xBD') {
				if(*(src + 2) >= '\x80' && *(src + 2) <= '\x9E') { // FULLWIDTH GRAVE ACCENT through FULLWIDTH TILDE
					src += 2;
					*src -= '\x80';
					*dest++ = '`' + *src;
					continue;
				}
			}
		}
		*dest++ = *src;
	}
	*dest = 0;
	return str;
}

142 143
/* Replace all multi-byte UTF-8 sequences with 'ch' or 'zwch' (when non-zero) */
/* When ch and zwch are 0, effectively strips all UTF-8 chars from str */
144
char* utf8_replace_chars(char* str, char (*lookup)(enum unicode_codepoint), char unsupported_ch, char unsupported_zwch, char error_ch)
145 146 147 148 149 150 151 152 153 154 155
{
	char* end = str + strlen(str);
	char* dest = str;

	int len ;
	for(char* src= str; src < end; src += len) {
		if(!(*src & 0x80)) {
			*dest++ = *src;
			len = 1;
			continue;
		}
156
		enum unicode_codepoint codepoint = 0;
157 158 159 160 161 162 163 164 165 166 167 168 169 170
		len = utf8_getc(src, end - src, &codepoint);
		if(len < 2) {
			if(error_ch)
				*dest++ = error_ch;
			len = 1;
			continue;
		}
		if(lookup != NULL) {
			char ch = lookup(codepoint);
			if(ch) {
				*dest++ = ch;
				continue;
			}
		}
171
		if(unicode_width(codepoint) == 0) {
172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193
			if(unsupported_zwch)
				*dest++ = unsupported_zwch;
		} 
		else if(unsupported_ch)
			*dest++ = unsupported_ch;
	}
	*dest = 0;
	return str;
}

bool utf8_str_is_valid(const char* str)
{
	const char* end = str + strlen(str);
	while (str < end) {
		int len = utf8_getc(str, end - str, NULL);
		if (len < 1)
			return false;
		str += len;
	}
	return true;
}

194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227
// Return the total printed-width of UTF-8 string (str) accounting for zero/half/full-width codepoints
size_t utf8_str_total_width(const char* str)
{
	size_t count = 0;
	const char* end = str + strlen(str);
	while (str < end) {
		enum unicode_codepoint codepoint = 0;
		int len = utf8_getc(str, end - str, &codepoint);
		if (len < 1)
			break;
		count += unicode_width(codepoint);
		str += len;
	}
	return count;
}

// Return the count of chars within the specified width range in UTF-8 string (str)
size_t utf8_str_count_width(const char* str, size_t min_width, size_t max_width)
{
	size_t count = 0;
	const char* end = str + strlen(str);
	while (str < end) {
		enum unicode_codepoint codepoint = 0;
		int len = utf8_getc(str, end - str, &codepoint);
		if (len < 1)
			break;
		size_t width = unicode_width(codepoint);
		if(width >= min_width && width <= max_width)
			count++;
		str += len;
	}
	return count;
}

rswindell's avatar
rswindell committed
228 229 230 231
int cp437_to_utf8_str(const char* str, char* dest, size_t maxlen, unsigned char minval)
{
	int retval = 0;
	size_t outlen = 0;
232
	for(const unsigned char* p = (const unsigned char*)str; *p != 0; p++) {
rswindell's avatar
rswindell committed
233 234 235 236
		if(outlen >= maxlen) {
			retval = -1;
			break;
		}
237
		enum unicode_codepoint codepoint = 0;
rswindell's avatar
rswindell committed
238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253
		if(*p >= minval)
			codepoint = cp437_unicode_tbl[*p];
		if(codepoint) {
			retval = utf8_putc(dest + outlen, maxlen - outlen, codepoint);
			if(retval < 1)
				break;
			outlen += retval;
		} else {
			*(dest + outlen) = *p;
			outlen++;
		}
	}
	*(dest + outlen) = 0;
	return retval;
}

254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336
int utf8_to_cp437_str(const char *src, char *dest, size_t maxlen, unsigned char minval, size_t *outlen)
{
	int retval = 0;
	size_t lcl_outlen;
	unsigned char ch;
	if (outlen == NULL)
		outlen = &lcl_outlen;
	*outlen = 0;
	for(const char* p = src; *p != 0; p += retval) {
		if(*outlen >= maxlen) {
			retval = -1;
			break;
		}
		enum unicode_codepoint codepoint;
		retval = utf8_getc(p, maxlen - *outlen, &codepoint);
		if (retval < 1)
			break;
		ch = unicode_to_cp437(codepoint);
		if (ch) {
			*(dest + *outlen) = ch;
			(*outlen)++;
		}
	}
	*(dest + *outlen) = 0;
	return retval;
}

int latin1_to_utf8_str(const char* str, char* dest, size_t maxlen, unsigned char minval, size_t *outlen)
{
	int retval = 0;
	size_t lcl_outlen;
	if (outlen == NULL)
		outlen = &lcl_outlen;
	*outlen = 0;
	for(const unsigned char* p = (const unsigned char *)str; *p != 0; p++) {
		if(*outlen >= maxlen) {
			retval = -1;
			break;
		}
		enum unicode_codepoint codepoint = 0;
		if(*p >= minval)
			codepoint = *p;
		if(codepoint) {
			retval = utf8_putc(dest + *outlen, maxlen - *outlen, codepoint);
			if(retval < 1)
				break;
			*outlen += retval;
		} else {
			*(dest + *outlen) = *p;
			(*outlen)++;
		}
	}
	*(dest + *outlen) = 0;
	return retval;
}

int utf8_to_latin1_str(const char *src, char *dest, size_t maxlen, unsigned char minval, size_t *outlen)
{
	int retval = 0;
	size_t lcl_outlen;
	unsigned char ch;
	if (outlen == NULL)
		outlen = &lcl_outlen;
	*outlen = 0;
	for(const char* p = src; *p != 0; p += retval) {
		if(*outlen >= maxlen) {
			retval = -1;
			break;
		}
		enum unicode_codepoint codepoint;
		retval = utf8_getc(p, maxlen - *outlen, &codepoint);
		if (retval < 1)
			break;
		ch = unicode_to_latin1(codepoint);
		if (ch) {
			*(dest + *outlen) = ch;
			(*outlen)++;
		}
	}
	*(dest + *outlen) = 0;
	return retval;
}

337 338
#define is_unicode_surrogate(value) \
    (value >= UNICODE_BLOCK_SURROGATE_BEGIN && value <= UNICODE_BLOCK_SURROGATE_END)
rswindell's avatar
rswindell committed
339

340 341
// From openssl/crypto/asn1/a_utf8.c:
/*
342
 * Copyright 1995-2021 The OpenSSL Project Authors. All Rights Reserved.
343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362
 *
 * Licensed under the Apache License 2.0 (the "License").  You may not use
 * this file except in compliance with the License.  You can obtain a copy
 * in the file LICENSE in the source distribution or at
 * https://www.openssl.org/source/license.html
 */

/* UTF8 utilities */

/*-
 * This parses a UTF8 string one character at a time. It is passed a pointer
 * to the string and the length of the string. It sets 'value' to the value of
 * the current character. It returns the number of characters read or a
 * negative error code:
 * -1 = string too short
 * -2 = illegal character
 * -3 = subsequent characters not of the form 10xxxxxx
 * -4 = character encoded incorrectly (not minimal length).
 */

363
int utf8_getc(const char *str, size_t len, enum unicode_codepoint* val)
364 365 366 367 368 369
{
    const unsigned char *p;
    unsigned long value;
    int ret;
    if (len <= 0)
        return 0;
rswindell's avatar
rswindell committed
370
    p = (const unsigned char*)str;
371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396

    /* Check syntax and work out the encoded value (if correct) */
    if ((*p & 0x80) == 0) {
        value = *p++ & 0x7f;
        ret = 1;
    } else if ((*p & 0xe0) == 0xc0) {
        if (len < 2)
            return -1;
        if ((p[1] & 0xc0) != 0x80)
            return -3;
        value = (*p++ & 0x1f) << 6;
        value |= *p++ & 0x3f;
        if (value < 0x80)
            return -4;
        ret = 2;
    } else if ((*p & 0xf0) == 0xe0) {
        if (len < 3)
            return -1;
        if (((p[1] & 0xc0) != 0x80)
            || ((p[2] & 0xc0) != 0x80))
            return -3;
        value = (*p++ & 0xf) << 12;
        value |= (*p++ & 0x3f) << 6;
        value |= *p++ & 0x3f;
        if (value < 0x800)
            return -4;
397 398
		if (is_unicode_surrogate(value))
			return -2;
399 400 401 402 403 404 405 406 407 408 409 410 411 412 413
        ret = 3;
    } else if ((*p & 0xf8) == 0xf0) {
        if (len < 4)
            return -1;
        if (((p[1] & 0xc0) != 0x80)
            || ((p[2] & 0xc0) != 0x80)
            || ((p[3] & 0xc0) != 0x80))
            return -3;
        value = ((unsigned long)(*p++ & 0x7)) << 18;
        value |= (*p++ & 0x3f) << 12;
        value |= (*p++ & 0x3f) << 6;
        value |= *p++ & 0x3f;
        if (value < 0x10000)
            return -4;
        ret = 4;
414
	} else
415
        return -2;
416 417
	if(val != NULL)
		*val = value;
418 419 420 421 422 423
    return ret;
}

/*
 * This takes a character 'value' and writes the UTF8 encoded value in 'str'
 * where 'str' is a buffer containing 'len' characters. Returns the number of
424 425 426
 * characters written, -1 if 'len' is too small or -2 if 'value' is out of
 * range. 'str' can be set to NULL in which case it just returns the number of
 * characters. It will need at most 4 characters.
427 428
 */

429
int utf8_putc(char *str, size_t len, enum unicode_codepoint value)
430 431
{
    if (!str)
432
        len = 4;                /* Maximum we will need */
433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449
    else if (len <= 0)
        return -1;
    if (value < 0x80) {
        if (str)
            *str = (unsigned char)value;
        return 1;
    }
    if (value < 0x800) {
        if (len < 2)
            return -1;
        if (str) {
            *str++ = (unsigned char)(((value >> 6) & 0x1f) | 0xc0);
            *str = (unsigned char)((value & 0x3f) | 0x80);
        }
        return 2;
    }
    if (value < 0x10000) {
450 451
		if (is_unicode_surrogate(value))
			return -2;
452 453 454 455 456 457 458 459 460
        if (len < 3)
            return -1;
        if (str) {
            *str++ = (unsigned char)(((value >> 12) & 0xf) | 0xe0);
            *str++ = (unsigned char)(((value >> 6) & 0x3f) | 0x80);
            *str = (unsigned char)((value & 0x3f) | 0x80);
        }
        return 3;
    }
461
    if (value < UNICODE_LIMIT) {
462 463 464 465 466 467 468 469 470 471
        if (len < 4)
            return -1;
        if (str) {
            *str++ = (unsigned char)(((value >> 18) & 0x7) | 0xf0);
            *str++ = (unsigned char)(((value >> 12) & 0x3f) | 0x80);
            *str++ = (unsigned char)(((value >> 6) & 0x3f) | 0x80);
            *str = (unsigned char)((value & 0x3f) | 0x80);
        }
        return 4;
    }
472
    return -2;
473
}