1 /*- 2 * SPDX-License-Identifier: BSD-3-Clause 3 * 4 * Copyright (c) 1990, 1993, 1994 5 * The Regents of the University of California. All rights reserved. 6 * 7 * This code is derived from software contributed to Berkeley by 8 * Michael Rendell of the Memorial University of Newfoundland. 9 * 10 * Redistribution and use in source and binary forms, with or without 11 * modification, are permitted provided that the following conditions 12 * are met: 13 * 1. Redistributions of source code must retain the above copyright 14 * notice, this list of conditions and the following disclaimer. 15 * 2. Redistributions in binary form must reproduce the above copyright 16 * notice, this list of conditions and the following disclaimer in the 17 * documentation and/or other materials provided with the distribution. 18 * 3. Neither the name of the University nor the names of its contributors 19 * may be used to endorse or promote products derived from this software 20 * without specific prior written permission. 21 * 22 * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND 23 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE 24 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE 25 * ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE 26 * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL 27 * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS 28 * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) 29 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT 30 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY 31 * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF 32 * SUCH DAMAGE. 33 */ 34 35 #ifndef lint 36 static const char copyright[] = 37 "@(#) Copyright (c) 1990, 1993, 1994\n\ 38 The Regents of the University of California. All rights reserved.\n"; 39 #endif 40 41 #if 0 42 #ifndef lint 43 static char sccsid[] = "@(#)col.c 8.5 (Berkeley) 5/4/95"; 44 #endif 45 #endif 46 47 #include <sys/cdefs.h> 48 #include <sys/capsicum.h> 49 50 #include <capsicum_helpers.h> 51 #include <err.h> 52 #include <errno.h> 53 #include <locale.h> 54 #include <stdio.h> 55 #include <stdlib.h> 56 #include <string.h> 57 #include <termios.h> 58 #include <unistd.h> 59 #include <wchar.h> 60 #include <wctype.h> 61 62 #define BS '\b' /* backspace */ 63 #define TAB '\t' /* tab */ 64 #define SPACE ' ' /* space */ 65 #define NL '\n' /* newline */ 66 #define CR '\r' /* carriage return */ 67 #define ESC '\033' /* escape */ 68 #define SI '\017' /* shift in to normal character set */ 69 #define SO '\016' /* shift out to alternate character set */ 70 #define VT '\013' /* vertical tab (aka reverse line feed) */ 71 #define RLF '7' /* ESC-7 reverse line feed */ 72 #define RHLF '8' /* ESC-8 reverse half-line feed */ 73 #define FHLF '9' /* ESC-9 forward half-line feed */ 74 75 /* build up at least this many lines before flushing them out */ 76 #define BUFFER_MARGIN 32 77 78 typedef char CSET; 79 80 typedef struct char_str { 81 #define CS_NORMAL 1 82 #define CS_ALTERNATE 2 83 short c_column; /* column character is in */ 84 CSET c_set; /* character set (currently only 2) */ 85 wchar_t c_char; /* character in question */ 86 int c_width; /* character width */ 87 } CHAR; 88 89 typedef struct line_str LINE; 90 struct line_str { 91 CHAR *l_line; /* characters on the line */ 92 LINE *l_prev; /* previous line */ 93 LINE *l_next; /* next line */ 94 int l_lsize; /* allocated sizeof l_line */ 95 int l_line_len; /* strlen(l_line) */ 96 int l_needs_sort; /* set if chars went in out of order */ 97 int l_max_col; /* max column in the line */ 98 }; 99 100 static void addto_lineno(int *, int); 101 static LINE *alloc_line(void); 102 static void dowarn(int); 103 static void flush_line(LINE *); 104 static void flush_lines(int); 105 static void flush_blanks(void); 106 static void free_line(LINE *); 107 static void usage(void); 108 109 static CSET last_set; /* char_set of last char printed */ 110 static LINE *lines; 111 static int compress_spaces; /* if doing space -> tab conversion */ 112 static int fine; /* if `fine' resolution (half lines) */ 113 static int max_bufd_lines; /* max # of half lines to keep in memory */ 114 static int nblank_lines; /* # blanks after last flushed line */ 115 static int no_backspaces; /* if not to output any backspaces */ 116 static int pass_unknown_seqs; /* pass unknown control sequences */ 117 118 #define PUTC(ch) \ 119 do { \ 120 if (putwchar(ch) == WEOF) \ 121 errx(1, "write error"); \ 122 } while (0) 123 124 int 125 main(int argc, char **argv) 126 { 127 wint_t ch; 128 CHAR *c; 129 CSET cur_set; /* current character set */ 130 LINE *l; /* current line */ 131 int extra_lines; /* # of lines above first line */ 132 int cur_col; /* current column */ 133 int cur_line; /* line number of current position */ 134 int max_line; /* max value of cur_line */ 135 int this_line; /* line l points to */ 136 int nflushd_lines; /* number of lines that were flushed */ 137 int adjust, opt, warned, width; 138 const char *errstr; 139 140 (void)setlocale(LC_CTYPE, ""); 141 142 if (caph_limit_stdio() == -1) 143 err(1, "unable to limit stdio"); 144 145 if (caph_enter() < 0) 146 err(1, "unable to enter capability mode"); 147 148 max_bufd_lines = 256; 149 compress_spaces = 1; /* compress spaces into tabs */ 150 while ((opt = getopt(argc, argv, "bfhl:px")) != -1) 151 switch (opt) { 152 case 'b': /* do not output backspaces */ 153 no_backspaces = 1; 154 break; 155 case 'f': /* allow half forward line feeds */ 156 fine = 1; 157 break; 158 case 'h': /* compress spaces into tabs */ 159 compress_spaces = 1; 160 break; 161 case 'l': /* buffered line count */ 162 max_bufd_lines = strtonum(optarg, 1, 163 (INT_MAX - BUFFER_MARGIN) / 2, &errstr) * 2; 164 if (errstr != NULL) 165 errx(1, "bad -l argument, %s: %s", errstr, 166 optarg); 167 break; 168 case 'p': /* pass unknown control sequences */ 169 pass_unknown_seqs = 1; 170 break; 171 case 'x': /* do not compress spaces into tabs */ 172 compress_spaces = 0; 173 break; 174 case '?': 175 default: 176 usage(); 177 } 178 179 if (optind != argc) 180 usage(); 181 182 adjust = cur_col = extra_lines = warned = 0; 183 cur_line = max_line = nflushd_lines = this_line = 0; 184 cur_set = last_set = CS_NORMAL; 185 lines = l = alloc_line(); 186 187 while ((ch = getwchar()) != WEOF) { 188 if (!iswgraph(ch)) { 189 switch (ch) { 190 case BS: /* can't go back further */ 191 if (cur_col == 0) 192 continue; 193 --cur_col; 194 continue; 195 case CR: 196 cur_col = 0; 197 continue; 198 case ESC: /* just ignore EOF */ 199 switch(getwchar()) { 200 /* 201 * In the input stream, accept both the 202 * XPG5 sequences ESC-digit and the 203 * traditional BSD sequences ESC-ctrl. 204 */ 205 case '\007': 206 /* FALLTHROUGH */ 207 case RLF: 208 addto_lineno(&cur_line, -2); 209 break; 210 case '\010': 211 /* FALLTHROUGH */ 212 case RHLF: 213 addto_lineno(&cur_line, -1); 214 break; 215 case '\011': 216 /* FALLTHROUGH */ 217 case FHLF: 218 addto_lineno(&cur_line, 1); 219 if (cur_line > max_line) 220 max_line = cur_line; 221 } 222 continue; 223 case NL: 224 addto_lineno(&cur_line, 2); 225 if (cur_line > max_line) 226 max_line = cur_line; 227 cur_col = 0; 228 continue; 229 case SPACE: 230 ++cur_col; 231 continue; 232 case SI: 233 cur_set = CS_NORMAL; 234 continue; 235 case SO: 236 cur_set = CS_ALTERNATE; 237 continue; 238 case TAB: /* adjust column */ 239 cur_col |= 7; 240 ++cur_col; 241 continue; 242 case VT: 243 addto_lineno(&cur_line, -2); 244 continue; 245 } 246 if (iswspace(ch)) { 247 if ((width = wcwidth(ch)) > 0) 248 cur_col += width; 249 continue; 250 } 251 if (!pass_unknown_seqs) 252 continue; 253 } 254 255 /* Must stuff ch in a line - are we at the right one? */ 256 if (cur_line + adjust != this_line) { 257 LINE *lnew; 258 259 /* round up to next line */ 260 adjust = !fine && (cur_line & 1); 261 262 if (cur_line + adjust < this_line) { 263 while (cur_line + adjust < this_line && 264 l->l_prev != NULL) { 265 l = l->l_prev; 266 this_line--; 267 } 268 if (cur_line + adjust < this_line) { 269 if (nflushd_lines == 0) { 270 /* 271 * Allow backup past first 272 * line if nothing has been 273 * flushed yet. 274 */ 275 while (cur_line + adjust 276 < this_line) { 277 lnew = alloc_line(); 278 l->l_prev = lnew; 279 lnew->l_next = l; 280 l = lines = lnew; 281 extra_lines++; 282 this_line--; 283 } 284 } else { 285 if (!warned++) 286 dowarn(cur_line); 287 cur_line = this_line - adjust; 288 } 289 } 290 } else { 291 /* may need to allocate here */ 292 while (cur_line + adjust > this_line) { 293 if (l->l_next == NULL) { 294 l->l_next = alloc_line(); 295 l->l_next->l_prev = l; 296 } 297 l = l->l_next; 298 this_line++; 299 } 300 } 301 if (this_line > nflushd_lines && 302 this_line - nflushd_lines >= 303 max_bufd_lines + BUFFER_MARGIN) { 304 if (extra_lines) { 305 flush_lines(extra_lines); 306 extra_lines = 0; 307 } 308 flush_lines(this_line - nflushd_lines - 309 max_bufd_lines); 310 nflushd_lines = this_line - max_bufd_lines; 311 } 312 } 313 /* grow line's buffer? */ 314 if (l->l_line_len + 1 >= l->l_lsize) { 315 int need; 316 317 need = l->l_lsize ? l->l_lsize * 2 : 90; 318 if ((l->l_line = realloc(l->l_line, 319 (unsigned)need * sizeof(CHAR))) == NULL) 320 err(1, NULL); 321 l->l_lsize = need; 322 } 323 c = &l->l_line[l->l_line_len++]; 324 c->c_char = ch; 325 c->c_set = cur_set; 326 c->c_column = cur_col; 327 c->c_width = wcwidth(ch); 328 /* 329 * If things are put in out of order, they will need sorting 330 * when it is flushed. 331 */ 332 if (cur_col < l->l_max_col) 333 l->l_needs_sort = 1; 334 else 335 l->l_max_col = cur_col; 336 if (c->c_width > 0) 337 cur_col += c->c_width; 338 } 339 if (ferror(stdin)) 340 err(1, NULL); 341 if (extra_lines) { 342 /* 343 * Extra lines only exist if no lines have been flushed 344 * yet. This means that 'lines' must point to line zero 345 * after we flush the extra lines. 346 */ 347 flush_lines(extra_lines); 348 l = lines; 349 this_line = 0; 350 } 351 352 /* goto the last line that had a character on it */ 353 for (; l->l_next; l = l->l_next) 354 this_line++; 355 flush_lines(this_line - nflushd_lines + 1); 356 357 /* make sure we leave things in a sane state */ 358 if (last_set != CS_NORMAL) 359 PUTC(SI); 360 361 /* flush out the last few blank lines */ 362 if (max_line >= this_line) 363 nblank_lines = max_line - this_line + (max_line & 1); 364 if (nblank_lines == 0) 365 /* end with a newline even if the source doesn't */ 366 nblank_lines = 2; 367 flush_blanks(); 368 exit(0); 369 } 370 371 /* 372 * Prints the first 'nflush' lines. Printed lines are freed. 373 * After this function returns, 'lines' points to the first 374 * of the remaining lines, and 'nblank_lines' will have the 375 * number of half line feeds between the final flushed line 376 * and the first remaining line. 377 */ 378 static void 379 flush_lines(int nflush) 380 { 381 LINE *l; 382 383 while (--nflush >= 0) { 384 l = lines; 385 lines = l->l_next; 386 if (l->l_line) { 387 flush_blanks(); 388 flush_line(l); 389 free(l->l_line); 390 } 391 if (l->l_next) 392 nblank_lines++; 393 free_line(l); 394 } 395 if (lines) 396 lines->l_prev = NULL; 397 } 398 399 /* 400 * Print a number of newline/half newlines. 401 * nblank_lines is the number of half line feeds. 402 */ 403 static void 404 flush_blanks(void) 405 { 406 int half, i, nb; 407 408 half = 0; 409 nb = nblank_lines; 410 if (nb & 1) { 411 if (fine) 412 half = 1; 413 else 414 nb++; 415 } 416 nb /= 2; 417 for (i = nb; --i >= 0;) 418 PUTC('\n'); 419 if (half) { 420 PUTC(ESC); 421 PUTC(FHLF); 422 if (!nb) 423 PUTC('\r'); 424 } 425 nblank_lines = 0; 426 } 427 428 /* 429 * Write a line to stdout taking care of space to tab conversion (-h flag) 430 * and character set shifts. 431 */ 432 static void 433 flush_line(LINE *l) 434 { 435 CHAR *c, *endc; 436 int i, j, nchars, last_col, save, this_col, tot; 437 438 last_col = 0; 439 nchars = l->l_line_len; 440 441 if (l->l_needs_sort) { 442 static CHAR *sorted; 443 static int count_size, *count, sorted_size; 444 445 /* 446 * Do an O(n) sort on l->l_line by column being careful to 447 * preserve the order of characters in the same column. 448 */ 449 if (l->l_lsize > sorted_size) { 450 sorted_size = l->l_lsize; 451 if ((sorted = realloc(sorted, 452 (unsigned)sizeof(CHAR) * sorted_size)) == NULL) 453 err(1, NULL); 454 } 455 if (l->l_max_col >= count_size) { 456 count_size = l->l_max_col + 1; 457 if ((count = realloc(count, 458 (unsigned)sizeof(int) * count_size)) == NULL) 459 err(1, NULL); 460 } 461 memset(count, 0, sizeof(int) * l->l_max_col + 1); 462 for (i = nchars, c = l->l_line; --i >= 0; c++) 463 count[c->c_column]++; 464 465 /* 466 * calculate running total (shifted down by 1) to use as 467 * indices into new line. 468 */ 469 for (tot = 0, i = 0; i <= l->l_max_col; i++) { 470 save = count[i]; 471 count[i] = tot; 472 tot += save; 473 } 474 475 for (i = nchars, c = l->l_line; --i >= 0; c++) 476 sorted[count[c->c_column]++] = *c; 477 c = sorted; 478 } else 479 c = l->l_line; 480 while (nchars > 0) { 481 this_col = c->c_column; 482 endc = c; 483 do { 484 ++endc; 485 } while (--nchars > 0 && this_col == endc->c_column); 486 487 /* if -b only print last character */ 488 if (no_backspaces) { 489 c = endc - 1; 490 if (nchars > 0 && 491 this_col + c->c_width > endc->c_column) 492 continue; 493 } 494 495 if (this_col > last_col) { 496 int nspace = this_col - last_col; 497 498 if (compress_spaces && nspace > 1) { 499 while (1) { 500 int tab_col, tab_size; 501 502 tab_col = (last_col + 8) & ~7; 503 if (tab_col > this_col) 504 break; 505 tab_size = tab_col - last_col; 506 if (tab_size == 1) 507 PUTC(' '); 508 else 509 PUTC('\t'); 510 nspace -= tab_size; 511 last_col = tab_col; 512 } 513 } 514 while (--nspace >= 0) 515 PUTC(' '); 516 last_col = this_col; 517 } 518 519 for (;;) { 520 if (c->c_set != last_set) { 521 switch (c->c_set) { 522 case CS_NORMAL: 523 PUTC(SI); 524 break; 525 case CS_ALTERNATE: 526 PUTC(SO); 527 } 528 last_set = c->c_set; 529 } 530 PUTC(c->c_char); 531 if ((c + 1) < endc) 532 for (j = 0; j < c->c_width; j++) 533 PUTC('\b'); 534 if (++c >= endc) 535 break; 536 } 537 last_col += (c - 1)->c_width; 538 } 539 } 540 541 /* 542 * Increment or decrement a line number, checking for overflow. 543 * Stop one below INT_MAX such that the adjust variable is safe. 544 */ 545 void 546 addto_lineno(int *lno, int offset) 547 { 548 if (offset > 0) { 549 if (*lno >= INT_MAX - offset) 550 errx(1, "too many lines"); 551 } else { 552 if (*lno < INT_MIN - offset) 553 errx(1, "too many reverse line feeds"); 554 } 555 *lno += offset; 556 } 557 558 #define NALLOC 64 559 560 static LINE *line_freelist; 561 562 static LINE * 563 alloc_line(void) 564 { 565 LINE *l; 566 int i; 567 568 if (!line_freelist) { 569 if ((l = realloc(NULL, sizeof(LINE) * NALLOC)) == NULL) 570 err(1, NULL); 571 line_freelist = l; 572 for (i = 1; i < NALLOC; i++, l++) 573 l->l_next = l + 1; 574 l->l_next = NULL; 575 } 576 l = line_freelist; 577 line_freelist = l->l_next; 578 579 memset(l, 0, sizeof(LINE)); 580 return (l); 581 } 582 583 static void 584 free_line(LINE *l) 585 { 586 587 l->l_next = line_freelist; 588 line_freelist = l; 589 } 590 591 static void 592 usage(void) 593 { 594 595 (void)fprintf(stderr, "usage: col [-bfhpx] [-l nline]\n"); 596 exit(1); 597 } 598 599 static void 600 dowarn(int line) 601 { 602 603 warnx("warning: can't back up %s", 604 line < 0 ? "past first line" : "-- line already flushed"); 605 } 606