1 /* $Id: mandoc.c,v 1.74 2013/12/30 18:30:32 schwarze Exp $ */
3 * Copyright (c) 2008, 2009, 2010, 2011 Kristaps Dzonsons <kristaps@bsd.lv>
4 * Copyright (c) 2011, 2012, 2013 Ingo Schwarze <schwarze@openbsd.org>
6 * Permission to use, copy, modify, and distribute this software for any
7 * purpose with or without fee is hereby granted, provided that the above
8 * copyright notice and this permission notice appear in all copies.
10 * THE SOFTWARE IS PROVIDED "AS IS" AND THE AUTHORS DISCLAIM ALL WARRANTIES
11 * WITH REGARD TO THIS SOFTWARE INCLUDING ALL IMPLIED WARRANTIES OF
12 * MERCHANTABILITY AND FITNESS. IN NO EVENT SHALL THE AUTHORS BE LIABLE FOR
13 * ANY SPECIAL, DIRECT, INDIRECT, OR CONSEQUENTIAL DAMAGES OR ANY DAMAGES
14 * WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS, WHETHER IN AN
15 * ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING OUT OF
16 * OR IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE.
22 #include <sys/types.h>
34 #include "libmandoc.h"
38 static int a2time(time_t *, const char *, const char *);
39 static char *time2a(time_t);
43 mandoc_escape(const char const **end, const char const **start, int *sz)
45 const char *local_start;
51 * When the caller doesn't provide return storage,
61 * Beyond the backslash, at least one input character
62 * is part of the escape sequence. With one exception
63 * (see below), that character won't be returned.
71 switch ((*start)[-1]) {
73 * First the glyphs. There are several different forms of
74 * these, but each eventually returns a substring of the glyph
84 * Unicode escapes are defined in groff as \[uXXXX] to
85 * \[u10FFFF], where the contained value must be a valid
86 * Unicode codepoint. Here, however, only check whether
87 * it's not a zero-width escape.
89 if ('u' == (*start)[0] && ']' != (*start)[1])
97 if ('u' == (*start)[0] && '\'' != (*start)[1])
100 gly = ESCAPE_SPECIAL;
105 * Escapes taking no arguments at all.
110 return(ESCAPE_IGNORE);
113 * The \z escape is supposed to output the following
114 * character without advancing the cursor position.
115 * Since we are mostly dealing with terminal mode,
116 * let us just skip the next character.
119 return(ESCAPE_SKIPCHAR);
122 * Handle all triggers matching \X(xy, \Xx, and \X[xxxx], where
123 * 'X' is the trigger. These have opaque sub-strings.
143 if (ESCAPE_ERROR == gly)
161 * These escapes are of the form \X'Y', where 'X' is the trigger
162 * and 'Y' is any string. These have opaque sub-strings.
182 return(ESCAPE_ERROR);
189 * These escapes are of the form \X'N', where 'X' is the trigger
190 * and 'N' resolves to a numerical expression.
206 return(ESCAPE_ERROR);
213 * Special handling for the numbered character escape.
214 * XXX Do any other escapes need similar handling?
218 return(ESCAPE_ERROR);
220 if (isdigit((unsigned char)**start)) {
222 return(ESCAPE_IGNORE);
225 while (isdigit((unsigned char)**end))
230 return(ESCAPE_NUMBERED);
233 * Sizes get a special category of their own.
238 /* See +/- counts as a sign. */
239 if ('+' == **end || '-' == **end || ASCII_HYPH == **end)
263 * Anything else is assumed to be a glyph.
264 * In this case, pass back the character after the backslash.
267 gly = ESCAPE_SPECIAL;
273 assert(ESCAPE_ERROR != gly);
276 * Read up to the terminating character,
277 * paying attention to nested escapes.
281 while (**end != term) {
284 return(ESCAPE_ERROR);
288 mandoc_escape(end, NULL, NULL))
289 return(ESCAPE_ERROR);
296 *sz = (*end)++ - *start;
299 if ((size_t)*sz > strlen(*start))
300 return(ESCAPE_ERROR);
304 /* Run post-processors. */
309 if ('C' == **start) {
311 * Treat constant-width font modes
312 * just like regular font modes.
317 if ('B' == (*start)[0] && 'I' == (*start)[1])
328 gly = ESCAPE_FONTBOLD;
333 gly = ESCAPE_FONTITALIC;
336 gly = ESCAPE_FONTPREV;
341 gly = ESCAPE_FONTROMAN;
345 case (ESCAPE_SPECIAL):
346 if (1 == *sz && 'c' == **start)
347 gly = ESCAPE_NOSPACE;
357 mandoc_calloc(size_t num, size_t size)
361 ptr = calloc(num, size);
364 exit((int)MANDOCLEVEL_SYSERR);
372 mandoc_malloc(size_t size)
379 exit((int)MANDOCLEVEL_SYSERR);
387 mandoc_realloc(void *ptr, size_t size)
390 ptr = realloc(ptr, size);
393 exit((int)MANDOCLEVEL_SYSERR);
400 mandoc_strndup(const char *ptr, size_t sz)
404 p = mandoc_malloc(sz + 1);
411 mandoc_strdup(const char *ptr)
418 exit((int)MANDOCLEVEL_SYSERR);
425 * Parse a quoted or unquoted roff-style request or macro argument.
426 * Return a pointer to the parsed argument, which is either the original
427 * pointer or advanced by one byte in case the argument is quoted.
428 * NUL-terminate the argument in place.
429 * Collapse pairs of quotes inside quoted arguments.
430 * Advance the argument pointer to the next argument,
431 * or to the NUL byte terminating the argument line.
434 mandoc_getarg(struct mparse *parse, char **cpp, int ln, int *pos)
437 int quoted, pairs, white;
439 /* Quoting can only start with a new word. */
449 for (cp = start; '\0' != *cp; cp++) {
452 * Move the following text left
453 * after quoted quotes and after "\\" and "\t".
460 * In copy mode, translate double to single
461 * backslashes and backslash-t to literal tabs.
472 /* Skip escaped blanks. */
479 } else if (0 == quoted) {
481 /* Unescaped blanks end unquoted args. */
485 } else if ('"' == cp[0]) {
487 /* Quoted quotes collapse. */
491 /* Unquoted quotes end quoted args. */
498 /* Quoted argument without a closing quote. */
500 mandoc_msg(MANDOCERR_BADQUOTE, parse, ln, *pos, NULL);
502 /* NUL-terminate this argument and move to the next one. */
510 *pos += (int)(cp - start) + (quoted ? 1 : 0);
513 if ('\0' == *cp && (white || ' ' == cp[-1]))
514 mandoc_msg(MANDOCERR_EOLNSPACE, parse, ln, *pos, NULL);
520 a2time(time_t *t, const char *fmt, const char *p)
525 memset(&tm, 0, sizeof(struct tm));
529 pp = strptime(p, fmt, &tm);
531 if (NULL != pp && '\0' == *pp) {
551 * up to 9 characters for the month (September) + blank
552 * up to 2 characters for the day + comma + blank
553 * 4 characters for the year and a terminating '\0'
555 p = buf = mandoc_malloc(10 + 4 + 4 + 1);
557 if (0 == (ssz = strftime(p, 10 + 1, "%B ", tm)))
561 if (-1 == (isz = snprintf(p, 4 + 1, "%d, ", tm->tm_mday)))
565 if (0 == strftime(p, 4 + 1, "%Y", tm))
575 mandoc_normdate(struct mparse *parse, char *in, int ln, int pos)
580 if (NULL == in || '\0' == *in ||
581 0 == strcmp(in, "$" "Mdocdate$")) {
582 mandoc_msg(MANDOCERR_NODATE, parse, ln, pos, NULL);
585 else if (a2time(&t, "%Y-%m-%d", in))
587 else if (!a2time(&t, "$" "Mdocdate: %b %d %Y $", in) &&
588 !a2time(&t, "%b %d, %Y", in)) {
589 mandoc_msg(MANDOCERR_BADDATE, parse, ln, pos, NULL);
592 out = t ? time2a(t) : NULL;
593 return(out ? out : mandoc_strdup(in));
597 mandoc_eos(const char *p, size_t sz, int enclosed)
606 * End-of-sentence recognition must include situations where
607 * some symbols, such as `)', allow prior EOS punctuation to
612 for (q = p + (int)sz - 1; q >= p; q--) {
632 return(found && (!enclosed || isalnum((unsigned char)*q)));
636 return(found && !enclosed);
640 * Convert a string to a long that may not be <0.
641 * If the string is invalid, or is less than 0, return -1.
644 mandoc_strntoi(const char *p, size_t sz, int base)
657 v = strtol(buf, &ep, base);
659 if (buf[0] == '\0' || *ep != '\0')