| 1 |
/* $NetBSD: chartype.c,v 1.30 2016/05/09 21:46:56 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 |
* |
| 16 |
* THIS SOFTWARE IS PROVIDED BY THE NETBSD FOUNDATION, INC. AND CONTRIBUTORS |
| 17 |
* ``AS IS'' AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED |
| 18 |
* TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR |
| 19 |
* PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE FOUNDATION OR CONTRIBUTORS |
| 20 |
* BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR |
| 21 |
* CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF |
| 22 |
* SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS |
| 23 |
* INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN |
| 24 |
* CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) |
| 25 |
* ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE |
| 26 |
* POSSIBILITY OF SUCH DAMAGE. |
| 27 |
*/ |
| 28 |
|
| 29 |
/* |
| 30 |
* chartype.c: character classification and meta information |
| 31 |
*/ |
| 32 |
#include "config.h" |
| 33 |
#if !defined(lint) && !defined(SCCSID) |
| 34 |
__RCSID("$NetBSD: chartype.c,v 1.30 2016/05/09 21:46:56 christos Exp $"); |
| 35 |
#endif /* not lint && not SCCSID */ |
| 36 |
|
| 37 |
#include <ctype.h> |
| 38 |
#include <stdlib.h> |
| 39 |
#include <string.h> |
| 40 |
|
| 41 |
#include "el.h" |
| 42 |
|
| 43 |
#define CT_BUFSIZ ((size_t)1024) |
| 44 |
|
| 45 |
static int ct_conv_cbuff_resize(ct_buffer_t *, size_t); |
| 46 |
static int ct_conv_wbuff_resize(ct_buffer_t *, size_t); |
| 47 |
|
| 48 |
static int |
| 49 |
ct_conv_cbuff_resize(ct_buffer_t *conv, size_t csize) |
| 50 |
{ |
| 51 |
void *p; |
| 52 |
|
| 53 |
if (csize <= conv->csize) |
| 54 |
return 0; |
| 55 |
|
| 56 |
conv->csize = csize; |
| 57 |
|
| 58 |
p = el_realloc(conv->cbuff, conv->csize * sizeof(*conv->cbuff)); |
| 59 |
if (p == NULL) { |
| 60 |
conv->csize = 0; |
| 61 |
el_free(conv->cbuff); |
| 62 |
conv->cbuff = NULL; |
| 63 |
return -1; |
| 64 |
} |
| 65 |
conv->cbuff = p; |
| 66 |
return 0; |
| 67 |
} |
| 68 |
|
| 69 |
static int |
| 70 |
ct_conv_wbuff_resize(ct_buffer_t *conv, size_t wsize) |
| 71 |
{ |
| 72 |
void *p; |
| 73 |
|
| 74 |
if (wsize <= conv->wsize) |
| 75 |
return 0; |
| 76 |
|
| 77 |
conv->wsize = wsize; |
| 78 |
|
| 79 |
p = el_realloc(conv->wbuff, conv->wsize * sizeof(*conv->wbuff)); |
| 80 |
if (p == NULL) { |
| 81 |
conv->wsize = 0; |
| 82 |
el_free(conv->wbuff); |
| 83 |
conv->wbuff = NULL; |
| 84 |
return -1; |
| 85 |
} |
| 86 |
conv->wbuff = p; |
| 87 |
return 0; |
| 88 |
} |
| 89 |
|
| 90 |
|
| 91 |
char * |
| 92 |
ct_encode_string(const wchar_t *s, ct_buffer_t *conv) |
| 93 |
{ |
| 94 |
char *dst; |
| 95 |
ssize_t used; |
| 96 |
|
| 97 |
if (!s) |
| 98 |
return NULL; |
| 99 |
|
| 100 |
dst = conv->cbuff; |
| 101 |
for (;;) { |
| 102 |
used = (ssize_t)(dst - conv->cbuff); |
| 103 |
if ((conv->csize - (size_t)used) < 5) { |
| 104 |
if (ct_conv_cbuff_resize(conv, |
| 105 |
conv->csize + CT_BUFSIZ) == -1) |
| 106 |
return NULL; |
| 107 |
dst = conv->cbuff + used; |
| 108 |
} |
| 109 |
if (!*s) |
| 110 |
break; |
| 111 |
used = ct_encode_char(dst, (size_t)5, *s); |
| 112 |
if (used == -1) /* failed to encode, need more buffer space */ |
| 113 |
abort(); |
| 114 |
++s; |
| 115 |
dst += used; |
| 116 |
} |
| 117 |
*dst = '\0'; |
| 118 |
return conv->cbuff; |
| 119 |
} |
| 120 |
|
| 121 |
wchar_t * |
| 122 |
ct_decode_string(const char *s, ct_buffer_t *conv) |
| 123 |
{ |
| 124 |
size_t len; |
| 125 |
|
| 126 |
if (!s) |
| 127 |
return NULL; |
| 128 |
|
| 129 |
len = mbstowcs(NULL, s, (size_t)0); |
| 130 |
if (len == (size_t)-1) |
| 131 |
return NULL; |
| 132 |
|
| 133 |
if (conv->wsize < ++len) |
| 134 |
if (ct_conv_wbuff_resize(conv, len + CT_BUFSIZ) == -1) |
| 135 |
return NULL; |
| 136 |
|
| 137 |
mbstowcs(conv->wbuff, s, conv->wsize); |
| 138 |
return conv->wbuff; |
| 139 |
} |
| 140 |
|
| 141 |
|
| 142 |
libedit_private wchar_t ** |
| 143 |
ct_decode_argv(int argc, const char *argv[], ct_buffer_t *conv) |
| 144 |
{ |
| 145 |
size_t bufspace; |
| 146 |
int i; |
| 147 |
wchar_t *p; |
| 148 |
wchar_t **wargv; |
| 149 |
ssize_t bytes; |
| 150 |
|
| 151 |
/* Make sure we have enough space in the conversion buffer to store all |
| 152 |
* the argv strings. */ |
| 153 |
for (i = 0, bufspace = 0; i < argc; ++i) |
| 154 |
bufspace += argv[i] ? strlen(argv[i]) + 1 : 0; |
| 155 |
if (conv->wsize < ++bufspace) |
| 156 |
if (ct_conv_wbuff_resize(conv, bufspace + CT_BUFSIZ) == -1) |
| 157 |
return NULL; |
| 158 |
|
| 159 |
wargv = el_malloc((size_t)argc * sizeof(*wargv)); |
| 160 |
|
| 161 |
for (i = 0, p = conv->wbuff; i < argc; ++i) { |
| 162 |
if (!argv[i]) { /* don't pass null pointers to mbstowcs */ |
| 163 |
wargv[i] = NULL; |
| 164 |
continue; |
| 165 |
} else { |
| 166 |
wargv[i] = p; |
| 167 |
bytes = (ssize_t)mbstowcs(p, argv[i], bufspace); |
| 168 |
} |
| 169 |
if (bytes == -1) { |
| 170 |
el_free(wargv); |
| 171 |
return NULL; |
| 172 |
} else |
| 173 |
bytes++; /* include '\0' in the count */ |
| 174 |
bufspace -= (size_t)bytes; |
| 175 |
p += bytes; |
| 176 |
} |
| 177 |
|
| 178 |
return wargv; |
| 179 |
} |
| 180 |
|
| 181 |
|
| 182 |
libedit_private size_t |
| 183 |
ct_enc_width(wchar_t c) |
| 184 |
{ |
| 185 |
/* UTF-8 encoding specific values */ |
| 186 |
if (c < 0x80) |
| 187 |
return 1; |
| 188 |
else if (c < 0x0800) |
| 189 |
return 2; |
| 190 |
else if (c < 0x10000) |
| 191 |
return 3; |
| 192 |
else if (c < 0x110000) |
| 193 |
return 4; |
| 194 |
else |
| 195 |
return 0; /* not a valid codepoint */ |
| 196 |
} |
| 197 |
|
| 198 |
libedit_private ssize_t |
| 199 |
ct_encode_char(char *dst, size_t len, wchar_t c) |
| 200 |
{ |
| 201 |
ssize_t l = 0; |
| 202 |
if (len < ct_enc_width(c)) |
| 203 |
return -1; |
| 204 |
l = wctomb(dst, c); |
| 205 |
|
| 206 |
if (l < 0) { |
| 207 |
wctomb(NULL, L'\0'); |
| 208 |
l = 0; |
| 209 |
} |
| 210 |
return l; |
| 211 |
} |
| 212 |
|
| 213 |
libedit_private const wchar_t * |
| 214 |
ct_visual_string(const wchar_t *s, ct_buffer_t *conv) |
| 215 |
{ |
| 216 |
wchar_t *dst; |
| 217 |
ssize_t used; |
| 218 |
|
| 219 |
if (!s) |
| 220 |
return NULL; |
| 221 |
|
| 222 |
if (ct_conv_wbuff_resize(conv, CT_BUFSIZ) == -1) |
| 223 |
return NULL; |
| 224 |
|
| 225 |
used = 0; |
| 226 |
dst = conv->wbuff; |
| 227 |
while (*s) { |
| 228 |
used = ct_visual_char(dst, |
| 229 |
conv->wsize - (size_t)(dst - conv->wbuff), *s); |
| 230 |
if (used != -1) { |
| 231 |
++s; |
| 232 |
dst += used; |
| 233 |
continue; |
| 234 |
} |
| 235 |
|
| 236 |
/* failed to encode, need more buffer space */ |
| 237 |
used = dst - conv->wbuff; |
| 238 |
if (ct_conv_wbuff_resize(conv, conv->wsize + CT_BUFSIZ) == -1) |
| 239 |
return NULL; |
| 240 |
dst = conv->wbuff + used; |
| 241 |
} |
| 242 |
|
| 243 |
if (dst >= (conv->wbuff + conv->wsize)) { /* sigh */ |
| 244 |
used = dst - conv->wbuff; |
| 245 |
if (ct_conv_wbuff_resize(conv, conv->wsize + CT_BUFSIZ) == -1) |
| 246 |
return NULL; |
| 247 |
dst = conv->wbuff + used; |
| 248 |
} |
| 249 |
|
| 250 |
*dst = L'\0'; |
| 251 |
return conv->wbuff; |
| 252 |
} |
| 253 |
|
| 254 |
|
| 255 |
|
| 256 |
libedit_private int |
| 257 |
ct_visual_width(wchar_t c) |
| 258 |
{ |
| 259 |
int t = ct_chr_class(c); |
| 260 |
switch (t) { |
| 261 |
case CHTYPE_ASCIICTL: |
| 262 |
return 2; /* ^@ ^? etc. */ |
| 263 |
case CHTYPE_TAB: |
| 264 |
return 1; /* Hmm, this really need to be handled outside! */ |
| 265 |
case CHTYPE_NL: |
| 266 |
return 0; /* Should this be 1 instead? */ |
| 267 |
case CHTYPE_PRINT: |
| 268 |
return wcwidth(c); |
| 269 |
case CHTYPE_NONPRINT: |
| 270 |
if (c > 0xffff) /* prefer standard 4-byte display over 5-byte */ |
| 271 |
return 8; /* \U+12345 */ |
| 272 |
else |
| 273 |
return 7; /* \U+1234 */ |
| 274 |
default: |
| 275 |
return 0; /* should not happen */ |
| 276 |
} |
| 277 |
} |
| 278 |
|
| 279 |
|
| 280 |
libedit_private ssize_t |
| 281 |
ct_visual_char(wchar_t *dst, size_t len, wchar_t c) |
| 282 |
{ |
| 283 |
int t = ct_chr_class(c); |
| 284 |
switch (t) { |
| 285 |
case CHTYPE_TAB: |
| 286 |
case CHTYPE_NL: |
| 287 |
case CHTYPE_ASCIICTL: |
| 288 |
if (len < 2) |
| 289 |
return -1; /* insufficient space */ |
| 290 |
*dst++ = '^'; |
| 291 |
if (c == '\177') |
| 292 |
*dst = '?'; /* DEL -> ^? */ |
| 293 |
else |
| 294 |
*dst = c | 0100; /* uncontrolify it */ |
| 295 |
return 2; |
| 296 |
case CHTYPE_PRINT: |
| 297 |
if (len < 1) |
| 298 |
return -1; /* insufficient space */ |
| 299 |
*dst = c; |
| 300 |
return 1; |
| 301 |
case CHTYPE_NONPRINT: |
| 302 |
/* we only use single-width glyphs for display, |
| 303 |
* so this is right */ |
| 304 |
if ((ssize_t)len < ct_visual_width(c)) |
| 305 |
return -1; /* insufficient space */ |
| 306 |
*dst++ = '\\'; |
| 307 |
*dst++ = 'U'; |
| 308 |
*dst++ = '+'; |
| 309 |
#define tohexdigit(v) "0123456789ABCDEF"[v] |
| 310 |
if (c > 0xffff) /* prefer standard 4-byte display over 5-byte */ |
| 311 |
*dst++ = tohexdigit(((unsigned int) c >> 16) & 0xf); |
| 312 |
*dst++ = tohexdigit(((unsigned int) c >> 12) & 0xf); |
| 313 |
*dst++ = tohexdigit(((unsigned int) c >> 8) & 0xf); |
| 314 |
*dst++ = tohexdigit(((unsigned int) c >> 4) & 0xf); |
| 315 |
*dst = tohexdigit(((unsigned int) c ) & 0xf); |
| 316 |
return c > 0xffff ? 8 : 7; |
| 317 |
/*FALLTHROUGH*/ |
| 318 |
/* these two should be handled outside this function */ |
| 319 |
default: /* we should never hit the default */ |
| 320 |
return 0; |
| 321 |
} |
| 322 |
} |
| 323 |
|
| 324 |
|
| 325 |
|
| 326 |
|
| 327 |
libedit_private int |
| 328 |
ct_chr_class(wchar_t c) |
| 329 |
{ |
| 330 |
if (c == '\t') |
| 331 |
return CHTYPE_TAB; |
| 332 |
else if (c == '\n') |
| 333 |
return CHTYPE_NL; |
| 334 |
else if (c < 0x100 && iswcntrl(c)) |
| 335 |
return CHTYPE_ASCIICTL; |
| 336 |
else if (iswprint(c)) |
| 337 |
return CHTYPE_PRINT; |
| 338 |
else |
| 339 |
return CHTYPE_NONPRINT; |
| 340 |
} |