1 /* $NetBSD: col.c,v 1.13 2003/10/16 06:45:22 itojun Exp $ */ 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 * 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 #include <sys/cdefs.h> 36 #ifndef lint 37 __COPYRIGHT("@(#) Copyright (c) 1990, 1993, 1994\n\ 38 The Regents of the University of California. All rights reserved.\n"); 39 #endif /* not lint */ 40 41 #ifndef lint 42 #if 0 43 static char sccsid[] = "@(#)col.c 8.5 (Berkeley) 5/4/95"; 44 #endif 45 __RCSID("$NetBSD: col.c,v 1.13 2003/10/16 06:45:22 itojun Exp $"); 46 #endif /* not lint */ 47 48 #include <ctype.h> 49 #include <err.h> 50 #include <string.h> 51 #include <stdio.h> 52 #include <stdlib.h> 53 #include <unistd.h> 54 55 #define BS '\b' /* backspace */ 56 #define TAB '\t' /* tab */ 57 #define SPACE ' ' /* space */ 58 #define NL '\n' /* newline */ 59 #define CR '\r' /* carriage return */ 60 #define ESC '\033' /* escape */ 61 #define SI '\017' /* shift in to normal character set */ 62 #define SO '\016' /* shift out to alternate character set */ 63 #define VT '\013' /* vertical tab (aka reverse line feed) */ 64 #define RLF '\007' /* ESC-07 reverse line feed */ 65 #define RHLF '\010' /* ESC-010 reverse half-line feed */ 66 #define FHLF '\011' /* ESC-011 forward half-line feed */ 67 68 /* build up at least this many lines before flushing them out */ 69 #define BUFFER_MARGIN 32 70 71 typedef char CSET; 72 73 typedef struct char_str { 74 #define CS_NORMAL 1 75 #define CS_ALTERNATE 2 76 short c_column; /* column character is in */ 77 CSET c_set; /* character set (currently only 2) */ 78 char c_char; /* character in question */ 79 } CHAR; 80 81 typedef struct line_str LINE; 82 struct line_str { 83 CHAR *l_line; /* characters on the line */ 84 LINE *l_prev; /* previous line */ 85 LINE *l_next; /* next line */ 86 int l_lsize; /* allocated sizeof l_line */ 87 int l_line_len; /* strlen(l_line) */ 88 int l_needs_sort; /* set if chars went in out of order */ 89 int l_max_col; /* max column in the line */ 90 }; 91 92 LINE *alloc_line __P((void)); 93 void dowarn __P((int)); 94 void flush_line __P((LINE *)); 95 void flush_lines __P((int)); 96 void flush_blanks __P((void)); 97 void free_line __P((LINE *)); 98 int main __P((int, char **)); 99 void usage __P((void)); 100 void wrerr __P((void)); 101 void *xmalloc __P((void *, size_t)); 102 103 CSET last_set; /* char_set of last char printed */ 104 LINE *lines; 105 int compress_spaces; /* if doing space -> tab conversion */ 106 int fine; /* if `fine' resolution (half lines) */ 107 int max_bufd_lines; /* max # lines to keep in memory */ 108 int nblank_lines; /* # blanks after last flushed line */ 109 int no_backspaces; /* if not to output any backspaces */ 110 int pass_unknown_seqs; /* whether to pass unknown control sequences */ 111 112 #define PUTC(ch) \ 113 if (putchar(ch) == EOF) \ 114 wrerr(); 115 116 int 117 main(argc, argv) 118 int argc; 119 char **argv; 120 { 121 int ch; 122 CHAR *c; 123 CSET cur_set; /* current character set */ 124 LINE *l; /* current line */ 125 int extra_lines; /* # of lines above first line */ 126 int cur_col; /* current column */ 127 int cur_line; /* line number of current position */ 128 int max_line; /* max value of cur_line */ 129 int this_line; /* line l points to */ 130 int nflushd_lines; /* number of lines that were flushed */ 131 int adjust, opt, warned; 132 133 max_bufd_lines = 128; 134 compress_spaces = 1; /* compress spaces into tabs */ 135 pass_unknown_seqs = 0; /* remove unknown escape sequences */ 136 while ((opt = getopt(argc, argv, "bfhl:px")) != -1) 137 switch (opt) { 138 case 'b': /* do not output backspaces */ 139 no_backspaces = 1; 140 break; 141 case 'f': /* allow half forward line feeds */ 142 fine = 1; 143 break; 144 case 'h': /* compress spaces into tabs */ 145 compress_spaces = 1; 146 break; 147 case 'l': /* buffered line count */ 148 if ((max_bufd_lines = atoi(optarg)) <= 0) { 149 (void)fprintf(stderr, 150 "col: bad -l argument %s.\n", optarg); 151 exit(EXIT_FAILURE); 152 } 153 break; 154 case 'p': /* pass unknown control sequences */ 155 pass_unknown_seqs = 1; 156 break; 157 case 'x': /* do not compress spaces into tabs */ 158 compress_spaces = 0; 159 break; 160 case '?': 161 default: 162 usage(); 163 } 164 165 if (optind != argc) 166 usage(); 167 168 /* this value is in half lines */ 169 max_bufd_lines *= 2; 170 171 adjust = cur_col = extra_lines = warned = 0; 172 cur_line = max_line = nflushd_lines = this_line = 0; 173 cur_set = last_set = CS_NORMAL; 174 lines = l = alloc_line(); 175 176 while ((ch = getchar()) != EOF) { 177 if (!isgraph(ch)) { 178 switch (ch) { 179 case BS: /* can't go back further */ 180 if (cur_col == 0) 181 continue; 182 --cur_col; 183 continue; 184 case CR: 185 cur_col = 0; 186 continue; 187 case ESC: /* just ignore EOF */ 188 switch(getchar()) { 189 case RLF: 190 cur_line -= 2; 191 break; 192 case RHLF: 193 cur_line--; 194 break; 195 case FHLF: 196 cur_line++; 197 if (cur_line > max_line) 198 max_line = cur_line; 199 } 200 continue; 201 case NL: 202 cur_line += 2; 203 if (cur_line > max_line) 204 max_line = cur_line; 205 cur_col = 0; 206 continue; 207 case SPACE: 208 ++cur_col; 209 continue; 210 case SI: 211 cur_set = CS_NORMAL; 212 continue; 213 case SO: 214 cur_set = CS_ALTERNATE; 215 continue; 216 case TAB: /* adjust column */ 217 cur_col |= 7; 218 ++cur_col; 219 continue; 220 case VT: 221 cur_line -= 2; 222 continue; 223 } 224 if (!pass_unknown_seqs) 225 continue; 226 } 227 228 /* Must stuff ch in a line - are we at the right one? */ 229 if (cur_line != this_line - adjust) { 230 LINE *lnew; 231 int nmove; 232 233 adjust = 0; 234 nmove = cur_line - this_line; 235 if (!fine) { 236 /* round up to next line */ 237 if (cur_line & 1) { 238 adjust = 1; 239 nmove++; 240 } 241 } 242 if (nmove < 0) { 243 for (; nmove < 0 && l->l_prev; nmove++) 244 l = l->l_prev; 245 if (nmove) { 246 if (nflushd_lines == 0) { 247 /* 248 * Allow backup past first 249 * line if nothing has been 250 * flushed yet. 251 */ 252 for (; nmove < 0; nmove++) { 253 lnew = alloc_line(); 254 l->l_prev = lnew; 255 lnew->l_next = l; 256 l = lines = lnew; 257 extra_lines++; 258 } 259 } else { 260 if (!warned++) 261 dowarn(cur_line); 262 cur_line -= nmove; 263 } 264 } 265 } else { 266 /* may need to allocate here */ 267 for (; nmove > 0 && l->l_next; nmove--) 268 l = l->l_next; 269 for (; nmove > 0; nmove--) { 270 lnew = alloc_line(); 271 lnew->l_prev = l; 272 l->l_next = lnew; 273 l = lnew; 274 } 275 } 276 this_line = cur_line + adjust; 277 nmove = this_line - nflushd_lines; 278 if (nmove >= max_bufd_lines + BUFFER_MARGIN) { 279 nflushd_lines += nmove - max_bufd_lines; 280 flush_lines(nmove - max_bufd_lines); 281 } 282 } 283 /* grow line's buffer? */ 284 if (l->l_line_len + 1 >= l->l_lsize) { 285 int need; 286 287 need = l->l_lsize ? l->l_lsize * 2 : 90; 288 l->l_line = (CHAR *)xmalloc((void *) l->l_line, 289 (unsigned) need * sizeof(CHAR)); 290 l->l_lsize = need; 291 } 292 c = &l->l_line[l->l_line_len++]; 293 c->c_char = ch; 294 c->c_set = cur_set; 295 c->c_column = cur_col; 296 /* 297 * If things are put in out of order, they will need sorting 298 * when it is flushed. 299 */ 300 if (cur_col < l->l_max_col) 301 l->l_needs_sort = 1; 302 else 303 l->l_max_col = cur_col; 304 cur_col++; 305 } 306 if (max_line == 0) 307 exit(EXIT_SUCCESS); /* no lines, so just exit */ 308 309 /* goto the last line that had a character on it */ 310 for (; l->l_next; l = l->l_next) 311 this_line++; 312 flush_lines(this_line - nflushd_lines + extra_lines + 1); 313 314 /* make sure we leave things in a sane state */ 315 if (last_set != CS_NORMAL) 316 PUTC('\017'); 317 318 /* flush out the last few blank lines */ 319 nblank_lines = max_line - this_line; 320 if (max_line & 1) 321 nblank_lines++; 322 else if (!nblank_lines) 323 /* missing a \n on the last line? */ 324 nblank_lines = 2; 325 flush_blanks(); 326 exit(EXIT_SUCCESS); 327 /* NOTREACHED */ 328 } 329 330 void 331 flush_lines(nflush) 332 int nflush; 333 { 334 LINE *l; 335 336 while (--nflush >= 0) { 337 l = lines; 338 lines = l->l_next; 339 if (l->l_line) { 340 flush_blanks(); 341 flush_line(l); 342 } 343 nblank_lines++; 344 if (l->l_line) 345 (void)free((void *)l->l_line); 346 free_line(l); 347 } 348 if (lines) 349 lines->l_prev = NULL; 350 } 351 352 /* 353 * Print a number of newline/half newlines. If fine flag is set, nblank_lines 354 * is the number of half line feeds, otherwise it is the number of whole line 355 * feeds. 356 */ 357 void 358 flush_blanks() 359 { 360 int half, i, nb; 361 362 half = 0; 363 nb = nblank_lines; 364 if (nb & 1) { 365 if (fine) 366 half = 1; 367 else 368 nb++; 369 } 370 nb /= 2; 371 for (i = nb; --i >= 0;) 372 PUTC('\n'); 373 if (half) { 374 PUTC('\033'); 375 PUTC('9'); 376 if (!nb) 377 PUTC('\r'); 378 } 379 nblank_lines = 0; 380 } 381 382 /* 383 * Write a line to stdout taking care of space to tab conversion (-h flag) 384 * and character set shifts. 385 */ 386 void 387 flush_line(l) 388 LINE *l; 389 { 390 CHAR *c, *endc; 391 int nchars, last_col, this_col; 392 393 last_col = 0; 394 nchars = l->l_line_len; 395 396 if (l->l_needs_sort) { 397 static CHAR *sorted; 398 static int count_size, *count, i, save, sorted_size, tot; 399 400 /* 401 * Do an O(n) sort on l->l_line by column being careful to 402 * preserve the order of characters in the same column. 403 */ 404 if (l->l_lsize > sorted_size) { 405 sorted_size = l->l_lsize; 406 sorted = (CHAR *)xmalloc((void *)sorted, 407 (unsigned)sizeof(CHAR) * sorted_size); 408 } 409 if (l->l_max_col >= count_size) { 410 count_size = l->l_max_col + 1; 411 count = (int *)xmalloc((void *)count, 412 (unsigned)sizeof(int) * count_size); 413 } 414 (void)memset(count, 0, sizeof(int) * l->l_max_col + 1); 415 for (i = nchars, c = l->l_line; --i >= 0; c++) 416 count[c->c_column]++; 417 418 /* 419 * calculate running total (shifted down by 1) to use as 420 * indices into new line. 421 */ 422 for (tot = 0, i = 0; i <= l->l_max_col; i++) { 423 save = count[i]; 424 count[i] = tot; 425 tot += save; 426 } 427 428 for (i = nchars, c = l->l_line; --i >= 0; c++) 429 sorted[count[c->c_column]++] = *c; 430 c = sorted; 431 } else 432 c = l->l_line; 433 while (nchars > 0) { 434 this_col = c->c_column; 435 endc = c; 436 do { 437 ++endc; 438 } while (--nchars > 0 && this_col == endc->c_column); 439 440 /* if -b only print last character */ 441 if (no_backspaces) 442 c = endc - 1; 443 444 if (this_col > last_col) { 445 int nspace = this_col - last_col; 446 447 if (compress_spaces && nspace > 1) { 448 int ntabs; 449 450 ntabs = ((last_col % 8) + nspace) / 8; 451 if (ntabs) { 452 nspace -= (ntabs * 8) - (last_col % 8); 453 while (--ntabs >= 0) 454 PUTC('\t'); 455 } 456 } 457 while (--nspace >= 0) 458 PUTC(' '); 459 last_col = this_col; 460 } 461 last_col++; 462 463 for (;;) { 464 if (c->c_set != last_set) { 465 switch (c->c_set) { 466 case CS_NORMAL: 467 PUTC('\017'); 468 break; 469 case CS_ALTERNATE: 470 PUTC('\016'); 471 } 472 last_set = c->c_set; 473 } 474 PUTC(c->c_char); 475 if (++c >= endc) 476 break; 477 PUTC('\b'); 478 } 479 } 480 } 481 482 #define NALLOC 64 483 484 static LINE *line_freelist; 485 486 LINE * 487 alloc_line() 488 { 489 LINE *l; 490 int i; 491 492 if (!line_freelist) { 493 l = (LINE *)xmalloc((void *)NULL, sizeof(LINE) * NALLOC); 494 line_freelist = l; 495 for (i = 1; i < NALLOC; i++, l++) 496 l->l_next = l + 1; 497 l->l_next = NULL; 498 } 499 l = line_freelist; 500 line_freelist = l->l_next; 501 502 (void)memset(l, 0, sizeof(LINE)); 503 return (l); 504 } 505 506 void 507 free_line(l) 508 LINE *l; 509 { 510 511 l->l_next = line_freelist; 512 line_freelist = l; 513 } 514 515 void * 516 xmalloc(p, size) 517 void *p; 518 size_t size; 519 { 520 void *q; 521 522 if (!(q = (void *)realloc(p, size))) 523 err(EXIT_FAILURE, "realloc"); 524 p = q; 525 return (p); 526 } 527 528 void 529 usage() 530 { 531 532 (void)fprintf(stderr, "usage: col [-bfpx] [-l nline]\n"); 533 exit(EXIT_FAILURE); 534 } 535 536 void 537 wrerr() 538 { 539 540 (void)fprintf(stderr, "col: write error.\n"); 541 exit(EXIT_FAILURE); 542 } 543 544 void 545 dowarn(line) 546 int line; 547 { 548 549 warnx("warning: can't back up %s", 550 line < 0 ? "past first line" : "-- line already flushed"); 551 } 552