1 /**************************************************************************** 2 * Copyright 2020 Thomas E. Dickey * 3 * Copyright 2005-2012,2017 Free Software Foundation, Inc. * 4 * * 5 * Permission is hereby granted, free of charge, to any person obtaining a * 6 * copy of this software and associated documentation files (the * 7 * "Software"), to deal in the Software without restriction, including * 8 * without limitation the rights to use, copy, modify, merge, publish, * 9 * distribute, distribute with modifications, sublicense, and/or sell * 10 * copies of the Software, and to permit persons to whom the Software is * 11 * furnished to do so, subject to the following conditions: * 12 * * 13 * The above copyright notice and this permission notice shall be included * 14 * in all copies or substantial portions of the Software. * 15 * * 16 * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS * 17 * OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF * 18 * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. * 19 * IN NO EVENT SHALL THE ABOVE COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, * 20 * DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR * 21 * OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR * 22 * THE USE OR OTHER DEALINGS IN THE SOFTWARE. * 23 * * 24 * Except as contained in this notice, the name(s) of the above copyright * 25 * holders shall not be used in advertising or otherwise to promote the * 26 * sale, use or other dealings in this Software without prior written * 27 * authorization. * 28 ****************************************************************************/ 29 30 /**************************************************************************** 31 * Author: Thomas Dickey * 32 ****************************************************************************/ 33 34 #include <curses.priv.h> 35 36 #include <ctype.h> 37 38 #include <tic.h> 39 40 MODULE_ID("$Id: trim_sgr0.c,v 1.18 2020/02/02 23:34:34 tom Exp $") 41 42 #undef CUR 43 #define CUR tp-> 44 45 #define CSI 233 46 #define ESC 033 /* ^[ */ 47 #define L_BRACK '[' 48 49 static char * 50 set_attribute_9(TERMTYPE2 *tp, int flag) 51 { 52 const char *value; 53 char *result; 54 55 value = tparm(set_attributes, 0, 0, 0, 0, 0, 0, 0, 0, flag); 56 if (PRESENT(value)) 57 result = strdup(value); 58 else 59 result = 0; 60 return result; 61 } 62 63 static int 64 is_csi(const char *s) 65 { 66 int result = 0; 67 if (s != 0) { 68 if (UChar(s[0]) == CSI) 69 result = 1; 70 else if (s[0] == ESC && s[1] == L_BRACK) 71 result = 2; 72 } 73 return result; 74 } 75 76 static char * 77 skip_zero(char *s) 78 { 79 if (s[0] == '0') { 80 if (s[1] == ';') 81 s += 2; 82 else if (isalpha(UChar(s[1]))) 83 s += 1; 84 } 85 return s; 86 } 87 88 static const char * 89 skip_delay(const char *s) 90 { 91 if (s[0] == '$' && s[1] == '<') { 92 s += 2; 93 while (isdigit(UChar(*s)) || *s == '/') 94 ++s; 95 if (*s == '>') 96 ++s; 97 } 98 return s; 99 } 100 101 /* 102 * Improve similar_sgr a little by moving the attr-string from the beginning 103 * to the end of the s-string. 104 */ 105 static bool 106 rewrite_sgr(char *s, char *attr) 107 { 108 if (s != 0) { 109 if (PRESENT(attr)) { 110 size_t len_s = strlen(s); 111 size_t len_a = strlen(attr); 112 113 if (len_s > len_a && !strncmp(attr, s, len_a)) { 114 unsigned n; 115 TR(TRACE_DATABASE, ("rewrite:\n\t%s", s)); 116 for (n = 0; n < len_s - len_a; ++n) { 117 s[n] = s[n + len_a]; 118 } 119 _nc_STRCPY(s + n, attr, strlen(s) + 1); 120 TR(TRACE_DATABASE, ("to:\n\t%s", s)); 121 } 122 } 123 return TRUE; 124 } 125 return FALSE; /* oops */ 126 } 127 128 static bool 129 similar_sgr(char *a, char *b) 130 { 131 bool result = FALSE; 132 if (a != 0 && b != 0) { 133 int csi_a = is_csi(a); 134 int csi_b = is_csi(b); 135 size_t len_a; 136 size_t len_b; 137 138 TR(TRACE_DATABASE, ("similar_sgr:\n\t%s\n\t%s", 139 _nc_visbuf2(1, a), 140 _nc_visbuf2(2, b))); 141 if (csi_a != 0 && csi_b != 0 && csi_a == csi_b) { 142 a += csi_a; 143 b += csi_b; 144 if (*a != *b) { 145 a = skip_zero(a); 146 b = skip_zero(b); 147 } 148 } 149 len_a = strlen(a); 150 len_b = strlen(b); 151 if (len_a && len_b) { 152 if (len_a > len_b) 153 result = (strncmp(a, b, len_b) == 0); 154 else 155 result = (strncmp(a, b, len_a) == 0); 156 } 157 TR(TRACE_DATABASE, ("...similar_sgr: %d\n\t%s\n\t%s", result, 158 _nc_visbuf2(1, a), 159 _nc_visbuf2(2, b))); 160 } 161 return result; 162 } 163 164 static unsigned 165 chop_out(char *string, unsigned i, unsigned j) 166 { 167 TR(TRACE_DATABASE, ("chop_out %d..%d from %s", i, j, _nc_visbuf(string))); 168 while (string[j] != '\0') { 169 string[i++] = string[j++]; 170 } 171 string[i] = '\0'; 172 return i; 173 } 174 175 /* 176 * Compare, ignoring delays. Some of the delay values are inconsistent, and 177 * we do not want to be stopped by that. 178 * 179 * Returns the number of chars from 'full' that we matched. If any mismatch 180 * occurs, return zero. 181 */ 182 static unsigned 183 compare_part(const char *part, const char *full) 184 { 185 const char *next_part; 186 const char *next_full; 187 unsigned used_full = 0; 188 unsigned used_delay = 0; 189 190 while (*part != 0) { 191 if (*part != *full) { 192 used_full = 0; 193 break; 194 } 195 196 /* 197 * Adjust the return-value to allow the rare case of 198 * string<delay>string 199 * to remove the whole piece. The most common case is a delay at the 200 * end of the string. The adjusted string will retain the delay, which 201 * is conservative. 202 */ 203 if (used_delay != 0) { 204 used_full += used_delay; 205 used_delay = 0; 206 } 207 if (*part == '$' && *full == '$') { 208 next_part = skip_delay(part); 209 next_full = skip_delay(full); 210 if (next_part != part && next_full != full) { 211 used_delay += (unsigned) (next_full - full); 212 full = next_full; 213 part = next_part; 214 continue; 215 } 216 } 217 ++used_full; 218 ++part; 219 ++full; 220 } 221 return used_full; 222 } 223 224 /* 225 * While 'sgr0' is the "same" as termcap 'me', there is a compatibility issue. 226 * The sgr/sgr0 capabilities include setting/clearing alternate character set 227 * mode. A termcap application cannot use sgr, so sgr0 strings that reset 228 * alternate character set mode will be misinterpreted. Here, we remove those 229 * from the more common ISO/ANSI/VT100 entries, which have sgr0 agreeing with 230 * sgr. 231 * 232 * This function returns the modified sgr0 if it can be modified, a null if 233 * an error occurs, or the original sgr0 if no change is needed. 234 */ 235 NCURSES_EXPORT(char *) 236 _nc_trim_sgr0(TERMTYPE2 *tp) 237 { 238 char *result = exit_attribute_mode; 239 240 T((T_CALLED("_nc_trim_sgr0()"))); 241 242 if (PRESENT(exit_attribute_mode) 243 && PRESENT(set_attributes)) { 244 bool found = FALSE; 245 char *on = set_attribute_9(tp, 1); 246 char *off = set_attribute_9(tp, 0); 247 char *end = strdup(exit_attribute_mode); 248 char *tmp; 249 size_t i, j, k; 250 251 TR(TRACE_DATABASE, ("checking if we can trim sgr0 based on sgr")); 252 TR(TRACE_DATABASE, ("sgr0 %s", _nc_visbuf(end))); 253 TR(TRACE_DATABASE, ("sgr(9:off) %s", _nc_visbuf(off))); 254 TR(TRACE_DATABASE, ("sgr(9:on) %s", _nc_visbuf(on))); 255 256 if (!rewrite_sgr(on, enter_alt_charset_mode) 257 || !rewrite_sgr(off, exit_alt_charset_mode) 258 || !rewrite_sgr(end, exit_alt_charset_mode)) { 259 FreeIfNeeded(off); 260 } else if (similar_sgr(off, end) 261 && !similar_sgr(off, on)) { 262 TR(TRACE_DATABASE, ("adjusting sgr(9:off) : %s", _nc_visbuf(off))); 263 result = off; 264 /* 265 * If rmacs is a substring of sgr(0), remove that chunk. 266 */ 267 if (PRESENT(exit_alt_charset_mode)) { 268 TR(TRACE_DATABASE, ("scan for rmacs %s", _nc_visbuf(exit_alt_charset_mode))); 269 j = strlen(off); 270 k = strlen(exit_alt_charset_mode); 271 if (j > k) { 272 for (i = 0; i <= (j - k); ++i) { 273 unsigned k2 = compare_part(exit_alt_charset_mode, 274 off + i); 275 if (k2 != 0) { 276 found = TRUE; 277 chop_out(off, (unsigned) i, (unsigned) (i + k2)); 278 break; 279 } 280 } 281 } 282 } 283 /* 284 * SGR 10 would reset to normal font. 285 */ 286 if (!found) { 287 if ((i = (size_t) is_csi(off)) != 0 288 && off[strlen(off) - 1] == 'm') { 289 TR(TRACE_DATABASE, ("looking for SGR 10 in %s", 290 _nc_visbuf(off))); 291 tmp = skip_zero(off + i); 292 if (tmp[0] == '1' 293 && skip_zero(tmp + 1) != tmp + 1) { 294 i = (size_t) (tmp - off); 295 if (off[i - 1] == ';') 296 i--; 297 j = (size_t) (skip_zero(tmp + 1) - off); 298 (void) chop_out(off, (unsigned) i, (unsigned) j); 299 found = TRUE; 300 } 301 } 302 } 303 if (!found 304 && (tmp = strstr(end, off)) != 0 305 && strcmp(end, off) != 0) { 306 i = (size_t) (tmp - end); 307 j = strlen(off); 308 tmp = strdup(end); 309 chop_out(tmp, (unsigned) i, (unsigned) j); 310 free(off); 311 result = tmp; 312 } 313 TR(TRACE_DATABASE, ("...adjusted sgr0 : %s", _nc_visbuf(result))); 314 if (!strcmp(result, exit_attribute_mode)) { 315 TR(TRACE_DATABASE, ("...same result, discard")); 316 free(result); 317 result = exit_attribute_mode; 318 } 319 } else { 320 /* 321 * Either the sgr does not reference alternate character set, 322 * or it is incorrect. That's too hard to decide right now. 323 */ 324 free(off); 325 } 326 FreeIfNeeded(end); 327 FreeIfNeeded(on); 328 } else { 329 /* 330 * Possibly some applications are confused if sgr0 contains rmacs, 331 * but that would be a different bug report -TD 332 */ 333 } 334 335 returnPtr(result); 336 } 337