Home | History | Annotate | Download | only in src
      1 /*	$NetBSD: chartype.c,v 1.10 2011/08/16 16:25:15 christos Exp $	*/
      2 
      3 /*-
      4  * Copyright (c) 2009 The NetBSD Foundation, Inc.
      5  * All rights reserved.
      6  *
      7  * Redistribution and use in source and binary forms, with or without
      8  * modification, are permitted provided that the following conditions
      9  * are met:
     10  * 1. Redistributions of source code must retain the above copyright
     11  *    notice, this list of conditions and the following disclaimer.
     12  * 2. Redistributions in binary form must reproduce the above copyright
     13  *    notice, this list of conditions and the following disclaimer in the
     14  *    documentation and/or other materials provided with the distribution.
     15  * 3. All advertising materials mentioning features or use of this software
     16  *    must display the following acknowledgement:
     17  *        This product includes software developed by the NetBSD
     18  *        Foundation, Inc. and its contributors.
     19  * 4. Neither the name of The NetBSD Foundation nor the names of its
     20  *    contributors may be used to endorse or promote products derived
     21  *    from this software without specific prior written permission.
     22  *
     23  * THIS SOFTWARE IS PROVIDED BY THE NETBSD FOUNDATION, INC. AND CONTRIBUTORS
     24  * ``AS IS'' AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED
     25  * TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR
     26  * PURPOSE ARE DISCLAIMED.  IN NO EVENT SHALL THE FOUNDATION OR CONTRIBUTORS
     27  * BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
     28  * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
     29  * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
     30  * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
     31  * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
     32  * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
     33  * POSSIBILITY OF SUCH DAMAGE.
     34  */
     35 
     36 /*
     37  * chartype.c: character classification and meta information
     38  */
     39 #include "config.h"
     40 #if !defined(lint) && !defined(SCCSID)
     41 __RCSID("$NetBSD: chartype.c,v 1.10 2011/08/16 16:25:15 christos Exp $");
     42 #endif /* not lint && not SCCSID */
     43 #include "el.h"
     44 #include <stdlib.h>
     45 
     46 #define CT_BUFSIZ ((size_t)1024)
     47 
     48 #ifdef WIDECHAR
     49 protected void
     50 ct_conv_buff_resize(ct_buffer_t *conv, size_t mincsize, size_t minwsize)
     51 {
     52 	void *p;
     53 	if (mincsize > conv->csize) {
     54 		conv->csize = mincsize;
     55 		p = el_realloc(conv->cbuff, conv->csize * sizeof(*conv->cbuff));
     56 		if (p == NULL) {
     57 			conv->csize = 0;
     58 			el_free(conv->cbuff);
     59 			conv->cbuff = NULL;
     60 		} else
     61 			conv->cbuff = p;
     62 	}
     63 
     64 	if (minwsize > conv->wsize) {
     65 		conv->wsize = minwsize;
     66 		p = el_realloc(conv->wbuff, conv->wsize * sizeof(*conv->wbuff));
     67 		if (p == NULL) {
     68 			conv->wsize = 0;
     69 			el_free(conv->wbuff);
     70 			conv->wbuff = NULL;
     71 		} else
     72 			conv->wbuff = p;
     73 	}
     74 }
     75 
     76 
     77 public char *
     78 ct_encode_string(const Char *s, ct_buffer_t *conv)
     79 {
     80 	char *dst;
     81 	ssize_t used = 0;
     82 
     83 	if (!s)
     84 		return NULL;
     85 	if (!conv->cbuff)
     86 		ct_conv_buff_resize(conv, CT_BUFSIZ, (size_t)0);
     87 	if (!conv->cbuff)
     88 		return NULL;
     89 
     90 	dst = conv->cbuff;
     91 	while (*s) {
     92 		used = (ssize_t)(conv->csize - (size_t)(dst - conv->cbuff));
     93 		if (used < 5) {
     94 			used = dst - conv->cbuff;
     95 			ct_conv_buff_resize(conv, conv->csize + CT_BUFSIZ,
     96 			    (size_t)0);
     97 			if (!conv->cbuff)
     98 				return NULL;
     99 			dst = conv->cbuff + used;
    100 		}
    101 		used = ct_encode_char(dst, (size_t)5, *s);
    102 		if (used == -1) /* failed to encode, need more buffer space */
    103 			abort();
    104 		++s;
    105 		dst += used;
    106 	}
    107 	*dst = '\0';
    108 	return conv->cbuff;
    109 }
    110 
    111 public Char *
    112 ct_decode_string(const char *s, ct_buffer_t *conv)
    113 {
    114 	size_t len = 0;
    115 
    116 	if (!s)
    117 		return NULL;
    118 	if (!conv->wbuff)
    119 		ct_conv_buff_resize(conv, (size_t)0, CT_BUFSIZ);
    120 	if (!conv->wbuff)
    121 		return NULL;
    122 
    123 	len = ct_mbstowcs(NULL, s, (size_t)0);
    124 	if (len == (size_t)-1)
    125 		return NULL;
    126 	if (len > conv->wsize)
    127 		ct_conv_buff_resize(conv, (size_t)0, len + 1);
    128 	if (!conv->wbuff)
    129 		return NULL;
    130 	ct_mbstowcs(conv->wbuff, s, conv->wsize);
    131 	return conv->wbuff;
    132 }
    133 
    134 
    135 protected Char **
    136 ct_decode_argv(int argc, const char *argv[], ct_buffer_t *conv)
    137 {
    138 	size_t bufspace;
    139 	int i;
    140 	Char *p;
    141 	Char **wargv;
    142 	ssize_t bytes;
    143 
    144 	/* Make sure we have enough space in the conversion buffer to store all
    145 	 * the argv strings. */
    146 	for (i = 0, bufspace = 0; i < argc; ++i)
    147 		bufspace += argv[i] ? strlen(argv[i]) + 1 : 0;
    148 	ct_conv_buff_resize(conv, (size_t)0, bufspace);
    149 	if (!conv->wsize)
    150 		return NULL;
    151 
    152 	wargv = el_malloc((size_t)argc * sizeof(*wargv));
    153 
    154 	for (i = 0, p = conv->wbuff; i < argc; ++i) {
    155 		if (!argv[i]) {   /* don't pass null pointers to mbstowcs */
    156 			wargv[i] = NULL;
    157 			continue;
    158 		} else {
    159 			wargv[i] = p;
    160 			bytes = (ssize_t)mbstowcs(p, argv[i], bufspace);
    161 		}
    162 		if (bytes == -1) {
    163 			el_free(wargv);
    164 			return NULL;
    165 		} else
    166 			bytes++;  /* include '\0' in the count */
    167 		bufspace -= (size_t)bytes;
    168 		p += bytes;
    169 	}
    170 
    171 	return wargv;
    172 }
    173 
    174 
    175 protected size_t
    176 ct_enc_width(Char c)
    177 {
    178 	/* UTF-8 encoding specific values */
    179 	if (c < 0x80)
    180 		return 1;
    181 	else if (c < 0x0800)
    182 		return 2;
    183 	else if (c < 0x10000)
    184 		return 3;
    185 	else if (c < 0x110000)
    186 		return 4;
    187 	else
    188 		return 0; /* not a valid codepoint */
    189 }
    190 
    191 protected ssize_t
    192 ct_encode_char(char *dst, size_t len, Char c)
    193 {
    194 	ssize_t l = 0;
    195 	if (len < ct_enc_width(c))
    196 		return -1;
    197 	l = ct_wctomb(dst, c);
    198 
    199 	if (l < 0) {
    200 		ct_wctomb_reset;
    201 		l = 0;
    202 	}
    203 	return l;
    204 }
    205 #endif
    206 
    207 protected const Char *
    208 ct_visual_string(const Char *s)
    209 {
    210 	static Char *buff = NULL;
    211 	static size_t buffsize = 0;
    212 	void *p;
    213 	Char *dst;
    214 	ssize_t used = 0;
    215 
    216 	if (!s)
    217 		return NULL;
    218 	if (!buff) {
    219 	    buffsize = CT_BUFSIZ;
    220 	    buff = el_malloc(buffsize * sizeof(*buff));
    221 	}
    222 	dst = buff;
    223 	while (*s) {
    224 		used = ct_visual_char(dst, buffsize - (size_t)(dst - buff), *s);
    225 		if (used == -1) { /* failed to encode, need more buffer space */
    226 			used = dst - buff;
    227 			buffsize += CT_BUFSIZ;
    228 			p = el_realloc(buff, buffsize * sizeof(*buff));
    229 			if (p == NULL)
    230 				goto out;
    231 			buff = p;
    232 			dst = buff + used;
    233 			/* don't increment s here - we want to retry it! */
    234 		}
    235 		else
    236 		    ++s;
    237 		dst += used;
    238 	}
    239 	if (dst >= (buff + buffsize)) { /* sigh */
    240 		buffsize += 1;
    241 		p = el_realloc(buff, buffsize * sizeof(*buff));
    242 		if (p == NULL)
    243 			goto out;
    244 		buff = p;
    245 		dst = buff + buffsize - 1;
    246 	}
    247 	*dst = 0;
    248 	return buff;
    249 out:
    250 	el_free(buff);
    251 	buffsize = 0;
    252 	return NULL;
    253 }
    254 
    255 
    256 
    257 protected int
    258 ct_visual_width(Char c)
    259 {
    260 	int t = ct_chr_class(c);
    261 	switch (t) {
    262 	case CHTYPE_ASCIICTL:
    263 		return 2; /* ^@ ^? etc. */
    264 	case CHTYPE_TAB:
    265 		return 1; /* Hmm, this really need to be handled outside! */
    266 	case CHTYPE_NL:
    267 		return 0; /* Should this be 1 instead? */
    268 #ifdef WIDECHAR
    269 	case CHTYPE_PRINT:
    270 		return wcwidth(c);
    271 	case CHTYPE_NONPRINT:
    272 		if (c > 0xffff) /* prefer standard 4-byte display over 5-byte */
    273 			return 8; /* \U+12345 */
    274 		else
    275 			return 7; /* \U+1234 */
    276 #else
    277 	case CHTYPE_PRINT:
    278 		return 1;
    279 	case CHTYPE_NONPRINT:
    280 		return 4; /* \123 */
    281 #endif
    282 	default:
    283 		return 0; /* should not happen */
    284 	}
    285 }
    286 
    287 
    288 protected ssize_t
    289 ct_visual_char(Char *dst, size_t len, Char c)
    290 {
    291 	int t = ct_chr_class(c);
    292 	switch (t) {
    293 	case CHTYPE_TAB:
    294 	case CHTYPE_NL:
    295 	case CHTYPE_ASCIICTL:
    296 		if (len < 2)
    297 			return -1;   /* insufficient space */
    298 		*dst++ = '^';
    299 		if (c == '\177')
    300 			*dst = '?'; /* DEL -> ^? */
    301 		else
    302 			*dst = c | 0100;    /* uncontrolify it */
    303 		return 2;
    304 	case CHTYPE_PRINT:
    305 		if (len < 1)
    306 			return -1;  /* insufficient space */
    307 		*dst = c;
    308 		return 1;
    309 	case CHTYPE_NONPRINT:
    310 		/* we only use single-width glyphs for display,
    311 		 * so this is right */
    312 		if ((ssize_t)len < ct_visual_width(c))
    313 			return -1;   /* insufficient space */
    314 #ifdef WIDECHAR
    315 		*dst++ = '\\';
    316 		*dst++ = 'U';
    317 		*dst++ = '+';
    318 #define tohexdigit(v) "0123456789ABCDEF"[v]
    319 		if (c > 0xffff) /* prefer standard 4-byte display over 5-byte */
    320 			*dst++ = tohexdigit(((unsigned int) c >> 16) & 0xf);
    321 		*dst++ = tohexdigit(((unsigned int) c >> 12) & 0xf);
    322 		*dst++ = tohexdigit(((unsigned int) c >>  8) & 0xf);
    323 		*dst++ = tohexdigit(((unsigned int) c >>  4) & 0xf);
    324 		*dst   = tohexdigit(((unsigned int) c      ) & 0xf);
    325 		return c > 0xffff ? 8 : 7;
    326 #else
    327 		*dst++ = '\\';
    328 #define tooctaldigit(v) ((v) + '0')
    329 		*dst++ = tooctaldigit(((unsigned int) c >> 6) & 0x7);
    330 		*dst++ = tooctaldigit(((unsigned int) c >> 3) & 0x7);
    331 		*dst++ = tooctaldigit(((unsigned int) c     ) & 0x7);
    332 #endif
    333 		/*FALLTHROUGH*/
    334 	/* these two should be handled outside this function */
    335 	default:            /* we should never hit the default */
    336 		return 0;
    337 	}
    338 }
    339 
    340 
    341 
    342 
    343 protected int
    344 ct_chr_class(Char c)
    345 {
    346 	if (c == '\t')
    347 		return CHTYPE_TAB;
    348 	else if (c == '\n')
    349 		return CHTYPE_NL;
    350 	else if (IsASCII(c) && Iscntrl(c))
    351 		return CHTYPE_ASCIICTL;
    352 	else if (Isprint(c))
    353 		return CHTYPE_PRINT;
    354 	else
    355 		return CHTYPE_NONPRINT;
    356 }
    357