1 /* $Id: mandoc.c,v 1.92 2015/02/20 23:55:10 schwarze Exp $ */
3 * Copyright (c) 2008-2011, 2014 Kristaps Dzonsons <kristaps@bsd.lv>
4 * Copyright (c) 2011-2015 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.
20 #include <sys/types.h>
32 #include "mandoc_aux.h"
33 #include "libmandoc.h"
37 static int a2time(time_t *, const char *, const char *);
38 static char *time2a(time_t);
42 mandoc_escape(const char **end, const char **start, int *sz)
44 const char *local_start;
50 * When the caller doesn't provide return storage,
60 * Beyond the backslash, at least one input character
61 * is part of the escape sequence. With one exception
62 * (see below), that character won't be returned.
70 switch ((*start)[-1]) {
72 * First the glyphs. There are several different forms of
73 * these, but each eventually returns a substring of the glyph
93 * Escapes taking no arguments at all.
98 return(ESCAPE_IGNORE);
101 * The \z escape is supposed to output the following
102 * character without advancing the cursor position.
103 * Since we are mostly dealing with terminal mode,
104 * let us just skip the next character.
107 return(ESCAPE_SKIPCHAR);
110 * Handle all triggers matching \X(xy, \Xx, and \X[xxxx], where
111 * 'X' is the trigger. These have opaque sub-strings.
131 if (ESCAPE_ERROR == gly)
149 * These escapes are of the form \X'Y', where 'X' is the trigger
150 * and 'Y' is any string. These have opaque sub-strings.
151 * The \B and \w escapes are handled in roff.c, roff_res().
168 return(ESCAPE_ERROR);
169 if (gly == ESCAPE_ERROR)
170 gly = ESCAPE_OVERSTRIKE;
176 * These escapes are of the form \X'N', where 'X' is the trigger
177 * and 'N' resolves to a numerical expression.
192 if (strchr(" %&()*+-./0123456789:<=>", **start)) {
195 return(ESCAPE_ERROR);
203 * Special handling for the numbered character escape.
204 * XXX Do any other escapes need similar handling?
208 return(ESCAPE_ERROR);
210 if (isdigit((unsigned char)**start)) {
212 return(ESCAPE_IGNORE);
215 while (isdigit((unsigned char)**end))
220 return(ESCAPE_NUMBERED);
223 * Sizes get a special category of their own.
228 /* See +/- counts as a sign. */
229 if ('+' == **end || '-' == **end || ASCII_HYPH == **end)
250 *sz = (*end)[-1] == 's' &&
251 isdigit((unsigned char)(*end)[1]) ? 2 : 1;
261 * Anything else is assumed to be a glyph.
262 * In this case, pass back the character after the backslash.
265 gly = ESCAPE_SPECIAL;
271 assert(ESCAPE_ERROR != gly);
274 * Read up to the terminating character,
275 * paying attention to nested escapes.
279 while (**end != term) {
282 return(ESCAPE_ERROR);
286 mandoc_escape(end, NULL, NULL))
287 return(ESCAPE_ERROR);
294 *sz = (*end)++ - *start;
297 if ((size_t)*sz > strlen(*start))
298 return(ESCAPE_ERROR);
302 /* Run post-processors. */
307 if ('C' == **start) {
309 * Treat constant-width font modes
310 * just like regular font modes.
315 if ('B' == (*start)[0] && 'I' == (*start)[1])
326 gly = ESCAPE_FONTBOLD;
331 gly = ESCAPE_FONTITALIC;
334 gly = ESCAPE_FONTPREV;
339 gly = ESCAPE_FONTROMAN;
344 if (1 == *sz && 'c' == **start)
345 gly = ESCAPE_NOSPACE;
347 * Unicode escapes are defined in groff as \[u0000]
348 * to \[u10FFFF], where the contained value must be
349 * a valid Unicode codepoint. Here, however, only
350 * check the length and range.
352 if (**start != 'u' || *sz < 5 || *sz > 7)
354 if (*sz == 7 && ((*start)[1] != '1' || (*start)[2] != '0'))
356 if (*sz == 6 && (*start)[1] == '0')
358 if ((int)strspn(*start + 1, "0123456789ABCDEFabcdef")
360 gly = ESCAPE_UNICODE;
370 * Parse a quoted or unquoted roff-style request or macro argument.
371 * Return a pointer to the parsed argument, which is either the original
372 * pointer or advanced by one byte in case the argument is quoted.
373 * NUL-terminate the argument in place.
374 * Collapse pairs of quotes inside quoted arguments.
375 * Advance the argument pointer to the next argument,
376 * or to the NUL byte terminating the argument line.
379 mandoc_getarg(struct mparse *parse, char **cpp, int ln, int *pos)
382 int quoted, pairs, white;
384 /* Quoting can only start with a new word. */
394 for (cp = start; '\0' != *cp; cp++) {
397 * Move the following text left
398 * after quoted quotes and after "\\" and "\t".
405 * In copy mode, translate double to single
406 * backslashes and backslash-t to literal tabs.
417 /* Skip escaped blanks. */
424 } else if (0 == quoted) {
426 /* Unescaped blanks end unquoted args. */
430 } else if ('"' == cp[0]) {
432 /* Quoted quotes collapse. */
436 /* Unquoted quotes end quoted args. */
443 /* Quoted argument without a closing quote. */
445 mandoc_msg(MANDOCERR_ARG_QUOTE, parse, ln, *pos, NULL);
447 /* NUL-terminate this argument and move to the next one. */
455 *pos += (int)(cp - start) + (quoted ? 1 : 0);
458 if ('\0' == *cp && (white || ' ' == cp[-1]))
459 mandoc_msg(MANDOCERR_SPACE_EOL, parse, ln, *pos, NULL);
465 a2time(time_t *t, const char *fmt, const char *p)
470 memset(&tm, 0, sizeof(struct tm));
474 pp = strptime(p, fmt, &tm);
476 if (NULL != pp && '\0' == *pp) {
498 * up to 9 characters for the month (September) + blank
499 * up to 2 characters for the day + comma + blank
500 * 4 characters for the year and a terminating '\0'
502 p = buf = mandoc_malloc(10 + 4 + 4 + 1);
504 if (0 == (ssz = strftime(p, 10 + 1, "%B ", tm)))
508 if (-1 == (isz = snprintf(p, 4 + 1, "%d, ", tm->tm_mday)))
512 if (0 == strftime(p, 4 + 1, "%Y", tm))
522 mandoc_normdate(struct mparse *parse, char *in, int ln, int pos)
527 if (NULL == in || '\0' == *in ||
528 0 == strcmp(in, "$" "Mdocdate$")) {
529 mandoc_msg(MANDOCERR_DATE_MISSING, parse, ln, pos, NULL);
532 else if (a2time(&t, "%Y-%m-%d", in))
534 else if (!a2time(&t, "$" "Mdocdate: %b %d %Y $", in) &&
535 !a2time(&t, "%b %d, %Y", in)) {
536 mandoc_msg(MANDOCERR_DATE_BAD, parse, ln, pos, in);
539 out = t ? time2a(t) : NULL;
540 return(out ? out : mandoc_strdup(in));
544 mandoc_eos(const char *p, size_t sz)
553 * End-of-sentence recognition must include situations where
554 * some symbols, such as `)', allow prior EOS punctuation to
558 enclosed = found = 0;
559 for (q = p + (int)sz - 1; q >= p; q--) {
579 return(found && (!enclosed || isalnum((unsigned char)*q)));
583 return(found && !enclosed);
587 * Convert a string to a long that may not be <0.
588 * If the string is invalid, or is less than 0, return -1.
591 mandoc_strntoi(const char *p, size_t sz, int base)
604 v = strtol(buf, &ep, base);
606 if (buf[0] == '\0' || *ep != '\0')