fmt.c revision 15344
1/* 2 * Copyright (c) 1980, 1993 3 * The Regents of the University of California. All rights reserved. 4 * 5 * Redistribution and use in source and binary forms, with or without 6 * modification, are permitted provided that the following conditions 7 * are met: 8 * 1. Redistributions of source code must retain the above copyright 9 * notice, this list of conditions and the following disclaimer. 10 * 2. Redistributions in binary form must reproduce the above copyright 11 * notice, this list of conditions and the following disclaimer in the 12 * documentation and/or other materials provided with the distribution. 13 * 3. All advertising materials mentioning features or use of this software 14 * must display the following acknowledgement: 15 * This product includes software developed by the University of 16 * California, Berkeley and its contributors. 17 * 4. Neither the name of the University nor the names of its contributors 18 * may be used to endorse or promote products derived from this software 19 * without specific prior written permission. 20 * 21 * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND 22 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE 23 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE 24 * ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE 25 * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL 26 * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS 27 * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) 28 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT 29 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY 30 * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF 31 * SUCH DAMAGE. 32 */ 33 34#ifndef lint 35static char copyright[] = 36"@(#) Copyright (c) 1980, 1993\n\ 37 The Regents of the University of California. All rights reserved.\n"; 38#endif /* not lint */ 39 40#ifndef lint 41static char sccsid[] = "@(#)fmt.c 8.1 (Berkeley) 7/20/93"; 42#endif /* not lint */ 43 44#include <stdio.h> 45#include <ctype.h> 46#include <locale.h> 47#include <stdlib.h> 48 49/* 50 * fmt -- format the concatenation of input files or standard input 51 * onto standard output. Designed for use with Mail ~| 52 * 53 * Syntax : fmt [ goal [ max ] ] [ name ... ] 54 * Authors: Kurt Shoens (UCB) 12/7/78; 55 * Liz Allen (UMCP) 2/24/83 [Addition of goal length concept]. 56 */ 57 58/* LIZ@UOM 6/18/85 -- Don't need LENGTH any more. 59 * #define LENGTH 72 Max line length in output 60 */ 61#define NOSTR ((char *) 0) /* Null string pointer for lint */ 62 63/* LIZ@UOM 6/18/85 --New variables goal_length and max_length */ 64#define GOAL_LENGTH 65 65#define MAX_LENGTH 75 66int goal_length; /* Target or goal line length in output */ 67int max_length; /* Max line length in output */ 68int pfx; /* Current leading blank count */ 69int lineno; /* Current input line */ 70int mark; /* Last place we saw a head line */ 71 72char *malloc(); /* for lint . . . */ 73char *headnames[] = {"To", "Subject", "Cc", 0}; 74 75/* 76 * Drive the whole formatter by managing input files. Also, 77 * cause initialization of the output stuff and flush it out 78 * at the end. 79 */ 80 81main(argc, argv) 82 int argc; 83 char **argv; 84{ 85 register FILE *fi; 86 register int errs = 0; 87 int number; /* LIZ@UOM 6/18/85 */ 88 89 (void) setlocale(LC_CTYPE, ""); 90 91 goal_length = GOAL_LENGTH; 92 max_length = MAX_LENGTH; 93 setout(); 94 lineno = 1; 95 mark = -10; 96 /* 97 * LIZ@UOM 6/18/85 -- Check for goal and max length arguments 98 */ 99 if (argc > 1 && (1 == (sscanf(argv[1], "%d", &number)))) { 100 argv++; 101 argc--; 102 goal_length = number; 103 if (argc > 1 && (1 == (sscanf(argv[1], "%d", &number)))) { 104 argv++; 105 argc--; 106 max_length = number; 107 } 108 } 109 if (max_length <= goal_length) { 110 fprintf(stderr, "Max length must be greater than %s\n", 111 "goal length"); 112 exit(1); 113 } 114 if (argc < 2) { 115 fmt(stdin); 116 oflush(); 117 exit(0); 118 } 119 while (--argc) { 120 if ((fi = fopen(*++argv, "r")) == NULL) { 121 perror(*argv); 122 errs++; 123 continue; 124 } 125 fmt(fi); 126 fclose(fi); 127 } 128 oflush(); 129 exit(errs); 130} 131 132/* 133 * Read up characters from the passed input file, forming lines, 134 * doing ^H processing, expanding tabs, stripping trailing blanks, 135 * and sending each line down for analysis. 136 */ 137fmt(fi) 138 FILE *fi; 139{ 140 static char *linebuf = 0, *canonb = 0; 141 register char *cp, *cp2, cc; 142 register int c, col; 143#define CHUNKSIZE 1024 144 static int lbufsize = 0, cbufsize = 0; 145 146 c = getc(fi); 147 while (c != EOF) { 148 /* 149 * Collect a line, doing ^H processing. 150 * Leave tabs for now. 151 */ 152 cp = linebuf; 153 while (c != '\n' && c != EOF) { 154 if (cp - linebuf >= lbufsize) { 155 int offset = cp - linebuf; 156 lbufsize += CHUNKSIZE; 157 linebuf = realloc(linebuf, lbufsize); 158 if(linebuf == 0) 159 abort(); 160 cp = linebuf + offset; 161 } 162 if (c == '\b') { 163 if (cp > linebuf) 164 cp--; 165 c = getc(fi); 166 continue; 167 } 168 if (!isprint(c) && c != '\t') { 169 c = getc(fi); 170 continue; 171 } 172 *cp++ = c; 173 c = getc(fi); 174 } 175 176 /* 177 * Toss anything remaining on the input line. 178 */ 179 while (c != '\n' && c != EOF) 180 c = getc(fi); 181 182 if (cp != NULL) { 183 *cp = '\0'; 184 } else { 185 putchar('\n'); 186 c = getc(fi); 187 continue; 188 } 189 190 /* 191 * Expand tabs on the way to canonb. 192 */ 193 col = 0; 194 cp = linebuf; 195 cp2 = canonb; 196 while (cc = *cp++) { 197 if (cc != '\t') { 198 col++; 199 if (cp2 - canonb >= cbufsize) { 200 int offset = cp2 - canonb; 201 cbufsize += CHUNKSIZE; 202 canonb = realloc(canonb, cbufsize); 203 if(canonb == 0) 204 abort(); 205 cp2 = canonb + offset; 206 } 207 *cp2++ = cc; 208 continue; 209 } 210 do { 211 if (cp2 - canonb >= cbufsize) { 212 int offset = cp2 - canonb; 213 cbufsize += CHUNKSIZE; 214 canonb = realloc(canonb, cbufsize); 215 if(canonb == 0) 216 abort(); 217 cp2 = canonb + offset; 218 } 219 *cp2++ = ' '; 220 col++; 221 } while ((col & 07) != 0); 222 } 223 224 /* 225 * Swipe trailing blanks from the line. 226 */ 227 for (cp2--; cp2 >= canonb && *cp2 == ' '; cp2--) 228 ; 229 *++cp2 = '\0'; 230 prefix(canonb); 231 if (c != EOF) 232 c = getc(fi); 233 } 234} 235 236/* 237 * Take a line devoid of tabs and other garbage and determine its 238 * blank prefix. If the indent changes, call for a linebreak. 239 * If the input line is blank, echo the blank line on the output. 240 * Finally, if the line minus the prefix is a mail header, try to keep 241 * it on a line by itself. 242 */ 243prefix(line) 244 char line[]; 245{ 246 register char *cp, **hp; 247 register int np, h; 248 249 if (!*line) { 250 oflush(); 251 putchar('\n'); 252 return; 253 } 254 for (cp = line; *cp == ' '; cp++) 255 ; 256 np = cp - line; 257 258 /* 259 * The following horrible expression attempts to avoid linebreaks 260 * when the indent changes due to a paragraph. 261 */ 262 if (np != pfx && (np > pfx || abs(pfx-np) > 8)) 263 oflush(); 264 if (h = ishead(cp)) 265 oflush(), mark = lineno; 266 if (lineno - mark < 3 && lineno - mark > 0) 267 for (hp = &headnames[0]; *hp != (char *) 0; hp++) 268 if (ispref(*hp, cp)) { 269 h = 1; 270 oflush(); 271 break; 272 } 273 if (!h && (h = (*cp == '.'))) 274 oflush(); 275 pfx = np; 276 if (h) 277 pack(cp, strlen(cp)); 278 else split(cp); 279 if (h) 280 oflush(); 281 lineno++; 282} 283 284/* 285 * Split up the passed line into output "words" which are 286 * maximal strings of non-blanks with the blank separation 287 * attached at the end. Pass these words along to the output 288 * line packer. 289 */ 290split(line) 291 char line[]; 292{ 293 register char *cp, *cp2; 294 char word[BUFSIZ]; 295 int wordl; /* LIZ@UOM 6/18/85 */ 296 297 cp = line; 298 while (*cp) { 299 cp2 = word; 300 wordl = 0; /* LIZ@UOM 6/18/85 */ 301 302 /* 303 * Collect a 'word,' allowing it to contain escaped white 304 * space. 305 */ 306 while (*cp && *cp != ' ') { 307 if (*cp == '\\' && isspace(cp[1])) 308 *cp2++ = *cp++; 309 *cp2++ = *cp++; 310 wordl++;/* LIZ@UOM 6/18/85 */ 311 } 312 313 /* 314 * Guarantee a space at end of line. Two spaces after end of 315 * sentence punctuation. 316 */ 317 if (*cp == '\0') { 318 *cp2++ = ' '; 319 if (index(".:!", cp[-1])) 320 *cp2++ = ' '; 321 } 322 while (*cp == ' ') 323 *cp2++ = *cp++; 324 *cp2 = '\0'; 325 /* 326 * LIZ@UOM 6/18/85 pack(word); 327 */ 328 pack(word, wordl); 329 } 330} 331 332/* 333 * Output section. 334 * Build up line images from the words passed in. Prefix 335 * each line with correct number of blanks. The buffer "outbuf" 336 * contains the current partial line image, including prefixed blanks. 337 * "outp" points to the next available space therein. When outp is NOSTR, 338 * there ain't nothing in there yet. At the bottom of this whole mess, 339 * leading tabs are reinserted. 340 */ 341char outbuf[BUFSIZ]; /* Sandbagged output line image */ 342char *outp; /* Pointer in above */ 343 344/* 345 * Initialize the output section. 346 */ 347setout() 348{ 349 outp = NOSTR; 350} 351 352/* 353 * Pack a word onto the output line. If this is the beginning of 354 * the line, push on the appropriately-sized string of blanks first. 355 * If the word won't fit on the current line, flush and begin a new 356 * line. If the word is too long to fit all by itself on a line, 357 * just give it its own and hope for the best. 358 * 359 * LIZ@UOM 6/18/85 -- If the new word will fit in at less than the 360 * goal length, take it. If not, then check to see if the line 361 * will be over the max length; if so put the word on the next 362 * line. If not, check to see if the line will be closer to the 363 * goal length with or without the word and take it or put it on 364 * the next line accordingly. 365 */ 366 367/* 368 * LIZ@UOM 6/18/85 -- pass in the length of the word as well 369 * pack(word) 370 * char word[]; 371 */ 372pack(word,wl) 373 char word[]; 374 int wl; 375{ 376 register char *cp; 377 register int s, t; 378 379 if (outp == NOSTR) 380 leadin(); 381 /* 382 * LIZ@UOM 6/18/85 -- change condition to check goal_length; s is the 383 * length of the line before the word is added; t is now the length 384 * of the line after the word is added 385 * t = strlen(word); 386 * if (t+s <= LENGTH) 387 */ 388 s = outp - outbuf; 389 t = wl + s; 390 if ((t <= goal_length) || 391 ((t <= max_length) && (t - goal_length <= goal_length - s))) { 392 /* 393 * In like flint! 394 */ 395 for (cp = word; *cp; *outp++ = *cp++); 396 return; 397 } 398 if (s > pfx) { 399 oflush(); 400 leadin(); 401 } 402 for (cp = word; *cp; *outp++ = *cp++); 403} 404 405/* 406 * If there is anything on the current output line, send it on 407 * its way. Set outp to NOSTR to indicate the absence of the current 408 * line prefix. 409 */ 410oflush() 411{ 412 if (outp == NOSTR) 413 return; 414 *outp = '\0'; 415 tabulate(outbuf); 416 outp = NOSTR; 417} 418 419/* 420 * Take the passed line buffer, insert leading tabs where possible, and 421 * output on standard output (finally). 422 */ 423tabulate(line) 424 char line[]; 425{ 426 register char *cp; 427 register int b, t; 428 429 /* 430 * Toss trailing blanks in the output line. 431 */ 432 cp = line + strlen(line) - 1; 433 while (cp >= line && *cp == ' ') 434 cp--; 435 *++cp = '\0'; 436 437 /* 438 * Count the leading blank space and tabulate. 439 */ 440 for (cp = line; *cp == ' '; cp++) 441 ; 442 b = cp-line; 443 t = b >> 3; 444 b &= 07; 445 if (t > 0) 446 do 447 putc('\t', stdout); 448 while (--t); 449 if (b > 0) 450 do 451 putc(' ', stdout); 452 while (--b); 453 while (*cp) 454 putc(*cp++, stdout); 455 putc('\n', stdout); 456} 457 458/* 459 * Initialize the output line with the appropriate number of 460 * leading blanks. 461 */ 462leadin() 463{ 464 register int b; 465 register char *cp; 466 467 for (b = 0, cp = outbuf; b < pfx; b++) 468 *cp++ = ' '; 469 outp = cp; 470} 471 472/* 473 * Save a string in dynamic space. 474 * This little goodie is needed for 475 * a headline detector in head.c 476 */ 477char * 478savestr(str) 479 char str[]; 480{ 481 register char *top; 482 483 top = malloc(strlen(str) + 1); 484 if (top == NOSTR) { 485 fprintf(stderr, "fmt: Ran out of memory\n"); 486 exit(1); 487 } 488 strcpy(top, str); 489 return (top); 490} 491 492/* 493 * Is s1 a prefix of s2?? 494 */ 495ispref(s1, s2) 496 register char *s1, *s2; 497{ 498 499 while (*s1++ == *s2) 500 ; 501 return (*s1 == '\0'); 502} 503