| 1 | /* From https://svnweb.freebsd.org/base/head/usr.bin/col/, r282722 */ |
| 2 | |
| 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 | * 4. 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 | |
| 49 | #include <err.h> |
| 50 | #include <errno.h> |
| 51 | #include <limits.h> |
| 52 | #include <locale.h> |
| 53 | #include <stdio.h> |
| 54 | #include <stdlib.h> |
| 55 | #include <string.h> |
| 56 | #include <termios.h> |
| 57 | #include <unistd.h> |
| 58 | #include <wchar.h> |
| 59 | #include <wctype.h> |
| 60 | |
| 61 | #define BS '\b' /* backspace */ |
| 62 | #define TAB '\t' /* tab */ |
| 63 | #define SPACE ' ' /* space */ |
| 64 | #define NL '\n' /* newline */ |
| 65 | #define CR '\r' /* carriage return */ |
| 66 | #define ESC '\033' /* escape */ |
| 67 | #define SI '\017' /* shift in to normal character set */ |
| 68 | #define SO '\016' /* shift out to alternate character set */ |
| 69 | #define VT '\013' /* vertical tab (aka reverse line feed) */ |
| 70 | #define RLF '7' /* ESC-7 reverse line feed */ |
| 71 | #define RHLF '8' /* ESC-8 reverse half-line feed */ |
| 72 | #define FHLF '9' /* ESC-9 forward half-line feed */ |
| 73 | |
| 74 | /* build up at least this many lines before flushing them out */ |
| 75 | #define BUFFER_MARGIN 32 |
| 76 | |
| 77 | typedef char CSET; |
| 78 | |
| 79 | typedef struct char_str { |
| 80 | #define CS_NORMAL 1 |
| 81 | #define CS_ALTERNATE 2 |
| 82 | short c_column; /* column character is in */ |
| 83 | CSET c_set; /* character set (currently only 2) */ |
| 84 | wchar_t c_char; /* character in question */ |
| 85 | int c_width; /* character width */ |
| 86 | } CHAR; |
| 87 | |
| 88 | typedef struct line_str LINE; |
| 89 | struct line_str { |
| 90 | CHAR *l_line; /* characters on the line */ |
| 91 | LINE *l_prev; /* previous line */ |
| 92 | LINE *l_next; /* next line */ |
| 93 | int l_lsize; /* allocated sizeof l_line */ |
| 94 | int l_line_len; /* strlen(l_line) */ |
| 95 | int l_needs_sort; /* set if chars went in out of order */ |
| 96 | int l_max_col; /* max column in the line */ |
| 97 | }; |
| 98 | |
| 99 | static void addto_lineno(int *, int); |
| 100 | static LINE *alloc_line(void); |
| 101 | static void dowarn(int); |
| 102 | static void flush_line(LINE *); |
| 103 | static void flush_lines(int); |
| 104 | static void flush_blanks(void); |
| 105 | static void free_line(LINE *); |
| 106 | static void usage(void); |
| 107 | |
| 108 | static CSET last_set; /* char_set of last char printed */ |
| 109 | static LINE *lines; |
| 110 | static int compress_spaces; /* if doing space -> tab conversion */ |
| 111 | static int fine; /* if `fine' resolution (half lines) */ |
| 112 | static int max_bufd_lines; /* max # of half lines to keep in memory */ |
| 113 | static int nblank_lines; /* # blanks after last flushed line */ |
| 114 | static int no_backspaces; /* if not to output any backspaces */ |
| 115 | static int pass_unknown_seqs; /* pass unknown control sequences */ |
| 116 | |
| 117 | #define PUTC(ch) \ |
| 118 | do { \ |
| 119 | if (putwchar(ch) == WEOF) \ |
| 120 | errx(1, "write error"); \ |
| 121 | } while (0) |
| 122 | |
| 123 | int |
| 124 | main(int argc, char **argv) |
| 125 | { |
| 126 | wint_t ch; |
| 127 | CHAR *c; |
| 128 | CSET cur_set; /* current character set */ |
| 129 | LINE *l; /* current line */ |
| 130 | int extra_lines; /* # of lines above first line */ |
| 131 | int cur_col; /* current column */ |
| 132 | int cur_line; /* line number of current position */ |
| 133 | int max_line; /* max value of cur_line */ |
| 134 | int this_line; /* line l points to */ |
| 135 | int nflushd_lines; /* number of lines that were flushed */ |
| 136 | int adjust, opt, warned, width; |
| 137 | const char *errstr; |
| 138 | unsigned long cmd; |
| 139 | |
| 140 | (void)setlocale(LC_CTYPE, ""); |
| 141 | |
| 142 | max_bufd_lines = 256; |
| 143 | compress_spaces = 1; /* compress spaces into tabs */ |
| 144 | while ((opt = getopt(argc, argv, "bfhl:px")) != -1) |
| 145 | switch (opt) { |
| 146 | case 'b': /* do not output backspaces */ |
| 147 | no_backspaces = 1; |
| 148 | break; |
| 149 | case 'f': /* allow half forward line feeds */ |
| 150 | fine = 1; |
| 151 | break; |
| 152 | case 'h': /* compress spaces into tabs */ |
| 153 | compress_spaces = 1; |
| 154 | break; |
| 155 | case 'l': /* buffered line count */ |
| 156 | max_bufd_lines = atoi(optarg); |
| 157 | break; |
| 158 | case 'p': /* pass unknown control sequences */ |
| 159 | pass_unknown_seqs = 1; |
| 160 | break; |
| 161 | case 'x': /* do not compress spaces into tabs */ |
| 162 | compress_spaces = 0; |
| 163 | break; |
| 164 | case '?': |
| 165 | default: |
| 166 | usage(); |
| 167 | } |
| 168 | |
| 169 | if (optind != argc) |
| 170 | usage(); |
| 171 | |
| 172 | adjust = cur_col = extra_lines = warned = 0; |
| 173 | cur_line = max_line = nflushd_lines = this_line = 0; |
| 174 | cur_set = last_set = CS_NORMAL; |
| 175 | lines = l = alloc_line(); |
| 176 | |
| 177 | while ((ch = getwchar()) != WEOF) { |
| 178 | if (!iswgraph(ch)) { |
| 179 | switch (ch) { |
| 180 | case BS: /* can't go back further */ |
| 181 | if (cur_col == 0) |
| 182 | continue; |
| 183 | --cur_col; |
| 184 | continue; |
| 185 | case CR: |
| 186 | cur_col = 0; |
| 187 | continue; |
| 188 | case ESC: /* just ignore EOF */ |
| 189 | switch(getwchar()) { |
| 190 | /* |
| 191 | * In the input stream, accept both the |
| 192 | * XPG5 sequences ESC-digit and the |
| 193 | * traditional BSD sequences ESC-ctrl. |
| 194 | */ |
| 195 | case '\007': |
| 196 | /* FALLTHROUGH */ |
| 197 | case RLF: |
| 198 | addto_lineno(&cur_line, -2); |
| 199 | break; |
| 200 | case '\010': |
| 201 | /* FALLTHROUGH */ |
| 202 | case RHLF: |
| 203 | addto_lineno(&cur_line, -1); |
| 204 | break; |
| 205 | case '\011': |
| 206 | /* FALLTHROUGH */ |
| 207 | case FHLF: |
| 208 | addto_lineno(&cur_line, 1); |
| 209 | if (cur_line > max_line) |
| 210 | max_line = cur_line; |
| 211 | } |
| 212 | continue; |
| 213 | case NL: |
| 214 | addto_lineno(&cur_line, 2); |
| 215 | if (cur_line > max_line) |
| 216 | max_line = cur_line; |
| 217 | cur_col = 0; |
| 218 | continue; |
| 219 | case SPACE: |
| 220 | ++cur_col; |
| 221 | continue; |
| 222 | case SI: |
| 223 | cur_set = CS_NORMAL; |
| 224 | continue; |
| 225 | case SO: |
| 226 | cur_set = CS_ALTERNATE; |
| 227 | continue; |
| 228 | case TAB: /* adjust column */ |
| 229 | cur_col |= 7; |
| 230 | ++cur_col; |
| 231 | continue; |
| 232 | case VT: |
| 233 | addto_lineno(&cur_line, -2); |
| 234 | continue; |
| 235 | } |
| 236 | if (iswspace(ch)) { |
| 237 | if ((width = wcwidth(ch)) > 0) |
| 238 | cur_col += width; |
| 239 | continue; |
| 240 | } |
| 241 | if (!pass_unknown_seqs) |
| 242 | continue; |
| 243 | } |
| 244 | |
| 245 | /* Must stuff ch in a line - are we at the right one? */ |
| 246 | if (cur_line + adjust != this_line) { |
| 247 | LINE *lnew; |
| 248 | |
| 249 | /* round up to next line */ |
| 250 | adjust = !fine && (cur_line & 1); |
| 251 | |
| 252 | if (cur_line + adjust < this_line) { |
| 253 | while (cur_line + adjust < this_line && |
| 254 | l->l_prev != NULL) { |
| 255 | l = l->l_prev; |
| 256 | this_line--; |
| 257 | } |
| 258 | if (cur_line + adjust < this_line) { |
| 259 | if (nflushd_lines == 0) { |
| 260 | /* |
| 261 | * Allow backup past first |
| 262 | * line if nothing has been |
| 263 | * flushed yet. |
| 264 | */ |
| 265 | while (cur_line + adjust |
| 266 | < this_line) { |
| 267 | lnew = alloc_line(); |
| 268 | l->l_prev = lnew; |
| 269 | lnew->l_next = l; |
| 270 | l = lines = lnew; |
| 271 | extra_lines++; |
| 272 | this_line--; |
| 273 | } |
| 274 | } else { |
| 275 | if (!warned++) |
| 276 | dowarn(cur_line); |
| 277 | cur_line = this_line - adjust; |
| 278 | } |
| 279 | } |
| 280 | } else { |
| 281 | /* may need to allocate here */ |
| 282 | while (cur_line + adjust > this_line) { |
| 283 | if (l->l_next == NULL) { |
| 284 | l->l_next = alloc_line(); |
| 285 | l->l_next->l_prev = l; |
| 286 | } |
| 287 | l = l->l_next; |
| 288 | this_line++; |
| 289 | } |
| 290 | } |
| 291 | if (this_line > nflushd_lines && |
| 292 | this_line - nflushd_lines >= |
| 293 | max_bufd_lines + BUFFER_MARGIN) { |
| 294 | if (extra_lines) { |
| 295 | flush_lines(extra_lines); |
| 296 | extra_lines = 0; |
| 297 | } |
| 298 | flush_lines(this_line - nflushd_lines - |
| 299 | max_bufd_lines); |
| 300 | nflushd_lines = this_line - max_bufd_lines; |
| 301 | } |
| 302 | } |
| 303 | /* grow line's buffer? */ |
| 304 | if (l->l_line_len + 1 >= l->l_lsize) { |
| 305 | int need; |
| 306 | |
| 307 | need = l->l_lsize ? l->l_lsize * 2 : 90; |
| 308 | if ((l->l_line = realloc(l->l_line, |
| 309 | (unsigned)need * sizeof(CHAR))) == NULL) |
| 310 | err(1, NULL); |
| 311 | l->l_lsize = need; |
| 312 | } |
| 313 | c = &l->l_line[l->l_line_len++]; |
| 314 | c->c_char = ch; |
| 315 | c->c_set = cur_set; |
| 316 | c->c_column = cur_col; |
| 317 | c->c_width = wcwidth(ch); |
| 318 | /* |
| 319 | * If things are put in out of order, they will need sorting |
| 320 | * when it is flushed. |
| 321 | */ |
| 322 | if (cur_col < l->l_max_col) |
| 323 | l->l_needs_sort = 1; |
| 324 | else |
| 325 | l->l_max_col = cur_col; |
| 326 | if (c->c_width > 0) |
| 327 | cur_col += c->c_width; |
| 328 | } |
| 329 | if (ferror(stdin)) |
| 330 | err(1, NULL); |
| 331 | if (extra_lines) |
| 332 | flush_lines(extra_lines); |
| 333 | |
| 334 | /* goto the last line that had a character on it */ |
| 335 | for (; l->l_next; l = l->l_next) |
| 336 | this_line++; |
| 337 | flush_lines(this_line - nflushd_lines + 1); |
| 338 | |
| 339 | /* make sure we leave things in a sane state */ |
| 340 | if (last_set != CS_NORMAL) |
| 341 | PUTC(SI); |
| 342 | |
| 343 | /* flush out the last few blank lines */ |
| 344 | if (max_line > this_line) |
| 345 | nblank_lines = max_line - this_line; |
| 346 | if (max_line & 1) |
| 347 | nblank_lines++; |
| 348 | flush_blanks(); |
| 349 | exit(0); |
| 350 | } |
| 351 | |
| 352 | static void |
| 353 | flush_lines(int nflush) |
| 354 | { |
| 355 | LINE *l; |
| 356 | |
| 357 | while (--nflush >= 0) { |
| 358 | l = lines; |
| 359 | lines = l->l_next; |
| 360 | if (l->l_line) { |
| 361 | flush_blanks(); |
| 362 | flush_line(l); |
| 363 | } |
| 364 | if (l->l_line || l->l_next) |
| 365 | nblank_lines++; |
| 366 | if (l->l_line) |
| 367 | (void)free(l->l_line); |
| 368 | free_line(l); |
| 369 | } |
| 370 | if (lines) |
| 371 | lines->l_prev = NULL; |
| 372 | } |
| 373 | |
| 374 | /* |
| 375 | * Print a number of newline/half newlines. If fine flag is set, nblank_lines |
| 376 | * is the number of half line feeds, otherwise it is the number of whole line |
| 377 | * feeds. |
| 378 | */ |
| 379 | static void |
| 380 | flush_blanks(void) |
| 381 | { |
| 382 | int half, i, nb; |
| 383 | |
| 384 | half = 0; |
| 385 | nb = nblank_lines; |
| 386 | if (nb & 1) { |
| 387 | if (fine) |
| 388 | half = 1; |
| 389 | else |
| 390 | nb++; |
| 391 | } |
| 392 | nb /= 2; |
| 393 | for (i = nb; --i >= 0;) |
| 394 | PUTC('\n'); |
| 395 | if (half) { |
| 396 | PUTC(ESC); |
| 397 | PUTC(FHLF); |
| 398 | if (!nb) |
| 399 | PUTC('\r'); |
| 400 | } |
| 401 | nblank_lines = 0; |
| 402 | } |
| 403 | |
| 404 | /* |
| 405 | * Write a line to stdout taking care of space to tab conversion (-h flag) |
| 406 | * and character set shifts. |
| 407 | */ |
| 408 | static void |
| 409 | flush_line(LINE *l) |
| 410 | { |
| 411 | CHAR *c, *endc; |
| 412 | int i, j, nchars, last_col, save, this_col, tot; |
| 413 | |
| 414 | last_col = 0; |
| 415 | nchars = l->l_line_len; |
| 416 | |
| 417 | if (l->l_needs_sort) { |
| 418 | static CHAR *sorted; |
| 419 | static int count_size, *count, sorted_size; |
| 420 | |
| 421 | /* |
| 422 | * Do an O(n) sort on l->l_line by column being careful to |
| 423 | * preserve the order of characters in the same column. |
| 424 | */ |
| 425 | if (l->l_lsize > sorted_size) { |
| 426 | sorted_size = l->l_lsize; |
| 427 | if ((sorted = realloc(sorted, |
| 428 | (unsigned)sizeof(CHAR) * sorted_size)) == NULL) |
| 429 | err(1, NULL); |
| 430 | } |
| 431 | if (l->l_max_col >= count_size) { |
| 432 | count_size = l->l_max_col + 1; |
| 433 | if ((count = realloc(count, |
| 434 | (unsigned)sizeof(int) * count_size)) == NULL) |
| 435 | err(1, NULL); |
| 436 | } |
| 437 | memset(count, 0, sizeof(int) * l->l_max_col + 1); |
| 438 | for (i = nchars, c = l->l_line; --i >= 0; c++) |
| 439 | count[c->c_column]++; |
| 440 | |
| 441 | /* |
| 442 | * calculate running total (shifted down by 1) to use as |
| 443 | * indices into new line. |
| 444 | */ |
| 445 | for (tot = 0, i = 0; i <= l->l_max_col; i++) { |
| 446 | save = count[i]; |
| 447 | count[i] = tot; |
| 448 | tot += save; |
| 449 | } |
| 450 | |
| 451 | for (i = nchars, c = l->l_line; --i >= 0; c++) |
| 452 | sorted[count[c->c_column]++] = *c; |
| 453 | c = sorted; |
| 454 | } else |
| 455 | c = l->l_line; |
| 456 | while (nchars > 0) { |
| 457 | this_col = c->c_column; |
| 458 | endc = c; |
| 459 | do { |
| 460 | ++endc; |
| 461 | } while (--nchars > 0 && this_col == endc->c_column); |
| 462 | |
| 463 | /* if -b only print last character */ |
| 464 | if (no_backspaces) { |
| 465 | c = endc - 1; |
| 466 | if (nchars > 0 && |
| 467 | this_col + c->c_width > endc->c_column) |
| 468 | continue; |
| 469 | } |
| 470 | |
| 471 | if (this_col > last_col) { |
| 472 | int nspace = this_col - last_col; |
| 473 | |
| 474 | if (compress_spaces && nspace > 1) { |
| 475 | while (1) { |
| 476 | int tab_col, tab_size; |
| 477 | |
| 478 | tab_col = (last_col + 8) & ~7; |
| 479 | if (tab_col > this_col) |
| 480 | break; |
| 481 | tab_size = tab_col - last_col; |
| 482 | if (tab_size == 1) |
| 483 | PUTC(' '); |
| 484 | else |
| 485 | PUTC('\t'); |
| 486 | nspace -= tab_size; |
| 487 | last_col = tab_col; |
| 488 | } |
| 489 | } |
| 490 | while (--nspace >= 0) |
| 491 | PUTC(' '); |
| 492 | last_col = this_col; |
| 493 | } |
| 494 | |
| 495 | for (;;) { |
| 496 | if (c->c_set != last_set) { |
| 497 | switch (c->c_set) { |
| 498 | case CS_NORMAL: |
| 499 | PUTC(SI); |
| 500 | break; |
| 501 | case CS_ALTERNATE: |
| 502 | PUTC(SO); |
| 503 | } |
| 504 | last_set = c->c_set; |
| 505 | } |
| 506 | PUTC(c->c_char); |
| 507 | if ((c + 1) < endc) |
| 508 | for (j = 0; j < c->c_width; j++) |
| 509 | PUTC('\b'); |
| 510 | if (++c >= endc) |
| 511 | break; |
| 512 | } |
| 513 | last_col += (c - 1)->c_width; |
| 514 | } |
| 515 | } |
| 516 | |
| 517 | /* |
| 518 | * Increment or decrement a line number, checking for overflow. |
| 519 | * Stop one below INT_MAX such that the adjust variable is safe. |
| 520 | */ |
| 521 | void |
| 522 | addto_lineno(int *lno, int offset) |
| 523 | { |
| 524 | if (offset > 0) { |
| 525 | if (*lno >= INT_MAX - offset) |
| 526 | errx(1, "too many lines"); |
| 527 | } else { |
| 528 | if (*lno < INT_MIN - offset) |
| 529 | errx(1, "too many reverse line feeds"); |
| 530 | } |
| 531 | *lno += offset; |
| 532 | } |
| 533 | |
| 534 | #define NALLOC 64 |
| 535 | |
| 536 | static LINE *line_freelist; |
| 537 | |
| 538 | static LINE * |
| 539 | alloc_line(void) |
| 540 | { |
| 541 | LINE *l; |
| 542 | int i; |
| 543 | |
| 544 | if (!line_freelist) { |
| 545 | if ((l = realloc(NULL, sizeof(LINE) * NALLOC)) == NULL) |
| 546 | err(1, NULL); |
| 547 | line_freelist = l; |
| 548 | for (i = 1; i < NALLOC; i++, l++) |
| 549 | l->l_next = l + 1; |
| 550 | l->l_next = NULL; |
| 551 | } |
| 552 | l = line_freelist; |
| 553 | line_freelist = l->l_next; |
| 554 | |
| 555 | memset(l, 0, sizeof(LINE)); |
| 556 | return (l); |
| 557 | } |
| 558 | |
| 559 | static void |
| 560 | free_line(LINE *l) |
| 561 | { |
| 562 | |
| 563 | l->l_next = line_freelist; |
| 564 | line_freelist = l; |
| 565 | } |
| 566 | |
| 567 | static void |
| 568 | usage(void) |
| 569 | { |
| 570 | |
| 571 | (void)fprintf(stderr, "usage: col [-bfhpx] [-l nline]\n"); |
| 572 | exit(1); |
| 573 | } |
| 574 | |
| 575 | static void |
| 576 | dowarn(int line) |
| 577 | { |
| 578 | |
| 579 | warnx("warning: can't back up %s", |
| 580 | line < 0 ? "past first line" : "-- line already flushed"); |
| 581 | } |