1 /* $Id: mandoc.c,v 1.68 2013/08/08 20:07:47 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 **end, const char **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])
102 * The \z escape is supposed to output the following
103 * character without advancing the cursor position.
104 * Since we are mostly dealing with terminal mode,
105 * let us just skip the next character.
108 return(ESCAPE_SKIPCHAR);
111 * Handle all triggers matching \X(xy, \Xx, and \X[xxxx], where
112 * 'X' is the trigger. These have opaque sub-strings.
132 if (ESCAPE_ERROR == gly)
150 * These escapes are of the form \X'Y', where 'X' is the trigger
151 * and 'Y' is any string. These have opaque sub-strings.
167 return(ESCAPE_ERROR);
174 * These escapes are of the form \X'N', where 'X' is the trigger
175 * and 'N' resolves to a numerical expression.
186 gly = ESCAPE_NUMBERED;
196 return(ESCAPE_ERROR);
197 if (ESCAPE_ERROR == gly)
204 * Special handling for the numbered character escape.
205 * XXX Do any other escapes need similar handling?
209 return(ESCAPE_ERROR);
211 if (isdigit((unsigned char)**start)) {
213 return(ESCAPE_IGNORE);
216 while (isdigit((unsigned char)**end))
221 return(ESCAPE_NUMBERED);
224 * Sizes get a special category of their own.
229 /* See +/- counts as a sign. */
230 if ('+' == **end || '-' == **end || ASCII_HYPH == **end)
254 * Anything else is assumed to be a glyph.
255 * In this case, pass back the character after the backslash.
258 gly = ESCAPE_SPECIAL;
264 assert(ESCAPE_ERROR != gly);
267 * Read up to the terminating character,
268 * paying attention to nested escapes.
272 while (**end != term) {
275 return(ESCAPE_ERROR);
279 mandoc_escape(end, NULL, NULL))
280 return(ESCAPE_ERROR);
287 *sz = (*end)++ - *start;
290 if ((size_t)*sz > strlen(*start))
291 return(ESCAPE_ERROR);
295 /* Run post-processors. */
300 if ('C' == **start) {
302 * Treat constant-width font modes
303 * just like regular font modes.
308 if ('B' == (*start)[0] && 'I' == (*start)[1])
319 gly = ESCAPE_FONTBOLD;
324 gly = ESCAPE_FONTITALIC;
327 gly = ESCAPE_FONTPREV;
332 gly = ESCAPE_FONTROMAN;
336 case (ESCAPE_SPECIAL):
337 if (1 == *sz && 'c' == **start)
338 gly = ESCAPE_NOSPACE;
348 mandoc_calloc(size_t num, size_t size)
352 ptr = calloc(num, size);
355 exit((int)MANDOCLEVEL_SYSERR);
363 mandoc_malloc(size_t size)
370 exit((int)MANDOCLEVEL_SYSERR);
378 mandoc_realloc(void *ptr, size_t size)
381 ptr = realloc(ptr, size);
384 exit((int)MANDOCLEVEL_SYSERR);
391 mandoc_strndup(const char *ptr, size_t sz)
395 p = mandoc_malloc(sz + 1);
402 mandoc_strdup(const char *ptr)
409 exit((int)MANDOCLEVEL_SYSERR);
416 * Parse a quoted or unquoted roff-style request or macro argument.
417 * Return a pointer to the parsed argument, which is either the original
418 * pointer or advanced by one byte in case the argument is quoted.
419 * Null-terminate the argument in place.
420 * Collapse pairs of quotes inside quoted arguments.
421 * Advance the argument pointer to the next argument,
422 * or to the null byte terminating the argument line.
425 mandoc_getarg(struct mparse *parse, char **cpp, int ln, int *pos)
428 int quoted, pairs, white;
430 /* Quoting can only start with a new word. */
440 for (cp = start; '\0' != *cp; cp++) {
443 * Move the following text left
444 * after quoted quotes and after "\\" and "\t".
451 * In copy mode, translate double to single
452 * backslashes and backslash-t to literal tabs.
463 /* Skip escaped blanks. */
470 } else if (0 == quoted) {
472 /* Unescaped blanks end unquoted args. */
476 } else if ('"' == cp[0]) {
478 /* Quoted quotes collapse. */
482 /* Unquoted quotes end quoted args. */
489 /* Quoted argument without a closing quote. */
491 mandoc_msg(MANDOCERR_BADQUOTE, parse, ln, *pos, NULL);
493 /* Null-terminate this argument and move to the next one. */
501 *pos += (int)(cp - start) + (quoted ? 1 : 0);
504 if ('\0' == *cp && (white || ' ' == cp[-1]))
505 mandoc_msg(MANDOCERR_EOLNSPACE, parse, ln, *pos, NULL);
511 a2time(time_t *t, const char *fmt, const char *p)
516 memset(&tm, 0, sizeof(struct tm));
520 pp = strptime(p, fmt, &tm);
522 if (NULL != pp && '\0' == *pp) {
542 * up to 9 characters for the month (September) + blank
543 * up to 2 characters for the day + comma + blank
544 * 4 characters for the year and a terminating '\0'
546 p = buf = mandoc_malloc(10 + 4 + 4 + 1);
548 if (0 == (ssz = strftime(p, 10 + 1, "%B ", tm)))
552 if (-1 == (isz = snprintf(p, 4 + 1, "%d, ", tm->tm_mday)))
556 if (0 == strftime(p, 4 + 1, "%Y", tm))
566 mandoc_normdate(struct mparse *parse, char *in, int ln, int pos)
571 if (NULL == in || '\0' == *in ||
572 0 == strcmp(in, "$" "Mdocdate$")) {
573 mandoc_msg(MANDOCERR_NODATE, parse, ln, pos, NULL);
576 else if (a2time(&t, "%Y-%m-%d", in))
578 else if (!a2time(&t, "$" "Mdocdate: %b %d %Y $", in) &&
579 !a2time(&t, "%b %d, %Y", in)) {
580 mandoc_msg(MANDOCERR_BADDATE, parse, ln, pos, NULL);
583 out = t ? time2a(t) : NULL;
584 return(out ? out : mandoc_strdup(in));
588 mandoc_eos(const char *p, size_t sz, int enclosed)
597 * End-of-sentence recognition must include situations where
598 * some symbols, such as `)', allow prior EOS punctuation to
603 for (q = p + (int)sz - 1; q >= p; q--) {
623 return(found && (!enclosed || isalnum((unsigned char)*q)));
627 return(found && !enclosed);
631 * Convert a string to a long that may not be <0.
632 * If the string is invalid, or is less than 0, return -1.
635 mandoc_strntoi(const char *p, size_t sz, int base)
648 v = strtol(buf, &ep, base);
650 if (buf[0] == '\0' || *ep != '\0')