1 /* $NetBSD: col.c,v 1.11 1999/09/23 08:47:47 tron 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. All advertising materials mentioning features or use of this software 19 * must display the following acknowledgement: 20 * This product includes software developed by the University of 21 * California, Berkeley and its contributors. 22 * 4. Neither the name of the University nor the names of its contributors 23 * may be used to endorse or promote products derived from this software 24 * without specific prior written permission. 25 * 26 * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND 27 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE 28 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE 29 * ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE 30 * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL 31 * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS 32 * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) 33 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT 34 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY 35 * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF 36 * SUCH DAMAGE. 37 */ 38 39 #include <sys/cdefs.h> 40 #ifndef lint 41 __COPYRIGHT("@(#) Copyright (c) 1990, 1993, 1994\n\ 42 The Regents of the University of California. All rights reserved.\n"); 43 #endif /* not lint */ 44 45 #ifndef lint 46 #if 0 47 static char sccsid[] = "@(#)col.c 8.5 (Berkeley) 5/4/95"; 48 #endif 49 __RCSID("$NetBSD: col.c,v 1.11 1999/09/23 08:47:47 tron Exp $"); 50 #endif /* not lint */ 51 52 #include <ctype.h> 53 #include <err.h> 54 #include <string.h> 55 #include <stdio.h> 56 #include <stdlib.h> 57 #include <unistd.h> 58 59 #define BS '\b' /* backspace */ 60 #define TAB '\t' /* tab */ 61 #define SPACE ' ' /* space */ 62 #define NL '\n' /* newline */ 63 #define CR '\r' /* carriage return */ 64 #define ESC '\033' /* escape */ 65 #define SI '\017' /* shift in to normal character set */ 66 #define SO '\016' /* shift out to alternate character set */ 67 #define VT '\013' /* vertical tab (aka reverse line feed) */ 68 #define RLF '\007' /* ESC-07 reverse line feed */ 69 #define RHLF '\010' /* ESC-010 reverse half-line feed */ 70 #define FHLF '\011' /* ESC-011 forward half-line feed */ 71 72 /* build up at least this many lines before flushing them out */ 73 #define BUFFER_MARGIN 32 74 75 typedef char CSET; 76 77 typedef struct char_str { 78 #define CS_NORMAL 1 79 #define CS_ALTERNATE 2 80 short c_column; /* column character is in */ 81 CSET c_set; /* character set (currently only 2) */ 82 char c_char; /* character in question */ 83 } CHAR; 84 85 typedef struct line_str LINE; 86 struct line_str { 87 CHAR *l_line; /* characters on the line */ 88 LINE *l_prev; /* previous line */ 89 LINE *l_next; /* next line */ 90 int l_lsize; /* allocated sizeof l_line */ 91 int l_line_len; /* strlen(l_line) */ 92 int l_needs_sort; /* set if chars went in out of order */ 93 int l_max_col; /* max column in the line */ 94 }; 95 96 LINE *alloc_line __P((void)); 97 void dowarn __P((int)); 98 void flush_line __P((LINE *)); 99 void flush_lines __P((int)); 100 void flush_blanks __P((void)); 101 void free_line __P((LINE *)); 102 int main __P((int, char **)); 103 void usage __P((void)); 104 void wrerr __P((void)); 105 void *xmalloc __P((void *, size_t)); 106 107 CSET last_set; /* char_set of last char printed */ 108 LINE *lines; 109 int compress_spaces; /* if doing space -> tab conversion */ 110 int fine; /* if `fine' resolution (half lines) */ 111 int max_bufd_lines; /* max # lines to keep in memory */ 112 int nblank_lines; /* # blanks after last flushed line */ 113 int no_backspaces; /* if not to output any backspaces */ 114 int pass_unknown_seqs; /* whether to pass unknown control sequences */ 115 116 #define PUTC(ch) \ 117 if (putchar(ch) == EOF) \ 118 wrerr(); 119 120 int 121 main(argc, argv) 122 int argc; 123 char **argv; 124 { 125 int ch; 126 CHAR *c; 127 CSET cur_set; /* current character set */ 128 LINE *l; /* current line */ 129 int extra_lines; /* # of lines above first line */ 130 int cur_col; /* current column */ 131 int cur_line; /* line number of current position */ 132 int max_line; /* max value of cur_line */ 133 int this_line; /* line l points to */ 134 int nflushd_lines; /* number of lines that were flushed */ 135 int adjust, opt, warned; 136 137 max_bufd_lines = 128; 138 compress_spaces = 1; /* compress spaces into tabs */ 139 pass_unknown_seqs = 0; /* remove unknown escape sequences */ 140 while ((opt = getopt(argc, argv, "bfhl:px")) != -1) 141 switch (opt) { 142 case 'b': /* do not output backspaces */ 143 no_backspaces = 1; 144 break; 145 case 'f': /* allow half forward line feeds */ 146 fine = 1; 147 break; 148 case 'h': /* compress spaces into tabs */ 149 compress_spaces = 1; 150 break; 151 case 'l': /* buffered line count */ 152 if ((max_bufd_lines = atoi(optarg)) <= 0) { 153 (void)fprintf(stderr, 154 "col: bad -l argument %s.\n", optarg); 155 exit(EXIT_FAILURE); 156 } 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 /* this value is in half lines */ 173 max_bufd_lines *= 2; 174 175 adjust = cur_col = extra_lines = warned = 0; 176 cur_line = max_line = nflushd_lines = this_line = 0; 177 cur_set = last_set = CS_NORMAL; 178 lines = l = alloc_line(); 179 180 while ((ch = getchar()) != EOF) { 181 if (!isgraph(ch)) { 182 switch (ch) { 183 case BS: /* can't go back further */ 184 if (cur_col == 0) 185 continue; 186 --cur_col; 187 continue; 188 case CR: 189 cur_col = 0; 190 continue; 191 case ESC: /* just ignore EOF */ 192 switch(getchar()) { 193 case RLF: 194 cur_line -= 2; 195 break; 196 case RHLF: 197 cur_line--; 198 break; 199 case FHLF: 200 cur_line++; 201 if (cur_line > max_line) 202 max_line = cur_line; 203 } 204 continue; 205 case NL: 206 cur_line += 2; 207 if (cur_line > max_line) 208 max_line = cur_line; 209 cur_col = 0; 210 continue; 211 case SPACE: 212 ++cur_col; 213 continue; 214 case SI: 215 cur_set = CS_NORMAL; 216 continue; 217 case SO: 218 cur_set = CS_ALTERNATE; 219 continue; 220 case TAB: /* adjust column */ 221 cur_col |= 7; 222 ++cur_col; 223 continue; 224 case VT: 225 cur_line -= 2; 226 continue; 227 } 228 if (!pass_unknown_seqs) 229 continue; 230 } 231 232 /* Must stuff ch in a line - are we at the right one? */ 233 if (cur_line != this_line - adjust) { 234 LINE *lnew; 235 int nmove; 236 237 adjust = 0; 238 nmove = cur_line - this_line; 239 if (!fine) { 240 /* round up to next line */ 241 if (cur_line & 1) { 242 adjust = 1; 243 nmove++; 244 } 245 } 246 if (nmove < 0) { 247 for (; nmove < 0 && l->l_prev; nmove++) 248 l = l->l_prev; 249 if (nmove) { 250 if (nflushd_lines == 0) { 251 /* 252 * Allow backup past first 253 * line if nothing has been 254 * flushed yet. 255 */ 256 for (; nmove < 0; nmove++) { 257 lnew = alloc_line(); 258 l->l_prev = lnew; 259 lnew->l_next = l; 260 l = lines = lnew; 261 extra_lines++; 262 } 263 } else { 264 if (!warned++) 265 dowarn(cur_line); 266 cur_line -= nmove; 267 } 268 } 269 } else { 270 /* may need to allocate here */ 271 for (; nmove > 0 && l->l_next; nmove--) 272 l = l->l_next; 273 for (; nmove > 0; nmove--) { 274 lnew = alloc_line(); 275 lnew->l_prev = l; 276 l->l_next = lnew; 277 l = lnew; 278 } 279 } 280 this_line = cur_line + adjust; 281 nmove = this_line - nflushd_lines; 282 if (nmove >= max_bufd_lines + BUFFER_MARGIN) { 283 nflushd_lines += nmove - max_bufd_lines; 284 flush_lines(nmove - max_bufd_lines); 285 } 286 } 287 /* grow line's buffer? */ 288 if (l->l_line_len + 1 >= l->l_lsize) { 289 int need; 290 291 need = l->l_lsize ? l->l_lsize * 2 : 90; 292 l->l_line = (CHAR *)xmalloc((void *) l->l_line, 293 (unsigned) need * sizeof(CHAR)); 294 l->l_lsize = need; 295 } 296 c = &l->l_line[l->l_line_len++]; 297 c->c_char = ch; 298 c->c_set = cur_set; 299 c->c_column = cur_col; 300 /* 301 * If things are put in out of order, they will need sorting 302 * when it is flushed. 303 */ 304 if (cur_col < l->l_max_col) 305 l->l_needs_sort = 1; 306 else 307 l->l_max_col = cur_col; 308 cur_col++; 309 } 310 if (max_line == 0) 311 exit(EXIT_SUCCESS); /* no lines, so just exit */ 312 313 /* goto the last line that had a character on it */ 314 for (; l->l_next; l = l->l_next) 315 this_line++; 316 flush_lines(this_line - nflushd_lines + extra_lines + 1); 317 318 /* make sure we leave things in a sane state */ 319 if (last_set != CS_NORMAL) 320 PUTC('\017'); 321 322 /* flush out the last few blank lines */ 323 nblank_lines = max_line - this_line; 324 if (max_line & 1) 325 nblank_lines++; 326 else if (!nblank_lines) 327 /* missing a \n on the last line? */ 328 nblank_lines = 2; 329 flush_blanks(); 330 exit(EXIT_SUCCESS); 331 /* NOTREACHED */ 332 } 333 334 void 335 flush_lines(nflush) 336 int nflush; 337 { 338 LINE *l; 339 340 while (--nflush >= 0) { 341 l = lines; 342 lines = l->l_next; 343 if (l->l_line) { 344 flush_blanks(); 345 flush_line(l); 346 } 347 nblank_lines++; 348 if (l->l_line) 349 (void)free((void *)l->l_line); 350 free_line(l); 351 } 352 if (lines) 353 lines->l_prev = NULL; 354 } 355 356 /* 357 * Print a number of newline/half newlines. If fine flag is set, nblank_lines 358 * is the number of half line feeds, otherwise it is the number of whole line 359 * feeds. 360 */ 361 void 362 flush_blanks() 363 { 364 int half, i, nb; 365 366 half = 0; 367 nb = nblank_lines; 368 if (nb & 1) { 369 if (fine) 370 half = 1; 371 else 372 nb++; 373 } 374 nb /= 2; 375 for (i = nb; --i >= 0;) 376 PUTC('\n'); 377 if (half) { 378 PUTC('\033'); 379 PUTC('9'); 380 if (!nb) 381 PUTC('\r'); 382 } 383 nblank_lines = 0; 384 } 385 386 /* 387 * Write a line to stdout taking care of space to tab conversion (-h flag) 388 * and character set shifts. 389 */ 390 void 391 flush_line(l) 392 LINE *l; 393 { 394 CHAR *c, *endc; 395 int nchars, last_col, this_col; 396 397 last_col = 0; 398 nchars = l->l_line_len; 399 400 if (l->l_needs_sort) { 401 static CHAR *sorted; 402 static int count_size, *count, i, save, sorted_size, tot; 403 404 /* 405 * Do an O(n) sort on l->l_line by column being careful to 406 * preserve the order of characters in the same column. 407 */ 408 if (l->l_lsize > sorted_size) { 409 sorted_size = l->l_lsize; 410 sorted = (CHAR *)xmalloc((void *)sorted, 411 (unsigned)sizeof(CHAR) * sorted_size); 412 } 413 if (l->l_max_col >= count_size) { 414 count_size = l->l_max_col + 1; 415 count = (int *)xmalloc((void *)count, 416 (unsigned)sizeof(int) * count_size); 417 } 418 (void)memset(count, 0, sizeof(int) * l->l_max_col + 1); 419 for (i = nchars, c = l->l_line; --i >= 0; c++) 420 count[c->c_column]++; 421 422 /* 423 * calculate running total (shifted down by 1) to use as 424 * indices into new line. 425 */ 426 for (tot = 0, i = 0; i <= l->l_max_col; i++) { 427 save = count[i]; 428 count[i] = tot; 429 tot += save; 430 } 431 432 for (i = nchars, c = l->l_line; --i >= 0; c++) 433 sorted[count[c->c_column]++] = *c; 434 c = sorted; 435 } else 436 c = l->l_line; 437 while (nchars > 0) { 438 this_col = c->c_column; 439 endc = c; 440 do { 441 ++endc; 442 } while (--nchars > 0 && this_col == endc->c_column); 443 444 /* if -b only print last character */ 445 if (no_backspaces) 446 c = endc - 1; 447 448 if (this_col > last_col) { 449 int nspace = this_col - last_col; 450 451 if (compress_spaces && nspace > 1) { 452 int ntabs; 453 454 ntabs = ((last_col % 8) + nspace) / 8; 455 if (ntabs) { 456 nspace -= (ntabs * 8) - (last_col % 8); 457 while (--ntabs >= 0) 458 PUTC('\t'); 459 } 460 } 461 while (--nspace >= 0) 462 PUTC(' '); 463 last_col = this_col; 464 } 465 last_col++; 466 467 for (;;) { 468 if (c->c_set != last_set) { 469 switch (c->c_set) { 470 case CS_NORMAL: 471 PUTC('\017'); 472 break; 473 case CS_ALTERNATE: 474 PUTC('\016'); 475 } 476 last_set = c->c_set; 477 } 478 PUTC(c->c_char); 479 if (++c >= endc) 480 break; 481 PUTC('\b'); 482 } 483 } 484 } 485 486 #define NALLOC 64 487 488 static LINE *line_freelist; 489 490 LINE * 491 alloc_line() 492 { 493 LINE *l; 494 int i; 495 496 if (!line_freelist) { 497 l = (LINE *)xmalloc((void *)NULL, sizeof(LINE) * NALLOC); 498 line_freelist = l; 499 for (i = 1; i < NALLOC; i++, l++) 500 l->l_next = l + 1; 501 l->l_next = NULL; 502 } 503 l = line_freelist; 504 line_freelist = l->l_next; 505 506 (void)memset(l, 0, sizeof(LINE)); 507 return (l); 508 } 509 510 void 511 free_line(l) 512 LINE *l; 513 { 514 515 l->l_next = line_freelist; 516 line_freelist = l; 517 } 518 519 void * 520 xmalloc(p, size) 521 void *p; 522 size_t size; 523 { 524 525 if (!(p = (void *)realloc(p, size))) 526 err(EXIT_FAILURE, "realloc"); 527 return (p); 528 } 529 530 void 531 usage() 532 { 533 534 (void)fprintf(stderr, "usage: col [-bfpx] [-l nline]\n"); 535 exit(EXIT_FAILURE); 536 } 537 538 void 539 wrerr() 540 { 541 542 (void)fprintf(stderr, "col: write error.\n"); 543 exit(EXIT_FAILURE); 544 } 545 546 void 547 dowarn(line) 548 int line; 549 { 550 551 warnx("warning: can't back up %s", 552 line < 0 ? "past first line" : "-- line already flushed"); 553 } 554