-/* $Id: mandoc.c,v 1.21 2010/07/06 22:04:31 kristaps Exp $ */
+/* $Id: mandoc.c,v 1.98 2015/11/12 22:44:27 schwarze Exp $ */
/*
- * Copyright (c) 2008, 2009 Kristaps Dzonsons <kristaps@bsd.lv>
+ * Copyright (c) 2008-2011, 2014 Kristaps Dzonsons <kristaps@bsd.lv>
+ * Copyright (c) 2011-2015 Ingo Schwarze <schwarze@openbsd.org>
*
* Permission to use, copy, modify, and distribute this software for any
* purpose with or without fee is hereby granted, provided that the above
* copyright notice and this permission notice appear in all copies.
*
- * THE SOFTWARE IS PROVIDED "AS IS" AND THE AUTHOR DISCLAIMS ALL WARRANTIES
+ * THE SOFTWARE IS PROVIDED "AS IS" AND THE AUTHORS DISCLAIM ALL WARRANTIES
* WITH REGARD TO THIS SOFTWARE INCLUDING ALL IMPLIED WARRANTIES OF
- * MERCHANTABILITY AND FITNESS. IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR
+ * MERCHANTABILITY AND FITNESS. IN NO EVENT SHALL THE AUTHORS BE LIABLE FOR
* ANY SPECIAL, DIRECT, INDIRECT, OR CONSEQUENTIAL DAMAGES OR ANY DAMAGES
* WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS, WHETHER IN AN
* ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING OUT OF
* OR IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE.
*/
-#ifdef HAVE_CONFIG_H
#include "config.h"
-#endif
#include <sys/types.h>
#include <assert.h>
#include <ctype.h>
+#include <errno.h>
+#include <limits.h>
#include <stdlib.h>
#include <stdio.h>
#include <string.h>
#include <time.h>
#include "mandoc.h"
+#include "mandoc_aux.h"
#include "libmandoc.h"
static int a2time(time_t *, const char *, const char *);
-static int spec_norm(char *, int);
+static char *time2a(time_t);
-/*
- * "Normalise" a special string by converting its ASCII_HYPH entries
- * into actual hyphens.
- */
-static int
-spec_norm(char *p, int sz)
+enum mandoc_esc
+mandoc_escape(const char **end, const char **start, int *sz)
{
- int i;
-
- for (i = 0; i < sz; i++)
- if (ASCII_HYPH == p[i])
- p[i] = '-';
+ const char *local_start;
+ int local_sz;
+ char term;
+ enum mandoc_esc gly;
- return(sz);
-}
-
-
-int
-mandoc_special(char *p)
-{
- int terminator; /* Terminator for \s. */
- int lim; /* Limit for N in \s. */
- int c, i;
- char *sv;
-
- sv = p;
-
- if ('\\' != *p++)
- return(spec_norm(sv, 0));
-
- switch (*p) {
- case ('\''):
- /* FALLTHROUGH */
- case ('`'):
- /* FALLTHROUGH */
- case ('q'):
- /* FALLTHROUGH */
- case (ASCII_HYPH):
- /* FALLTHROUGH */
- case ('-'):
- /* FALLTHROUGH */
- case ('~'):
- /* FALLTHROUGH */
- case ('^'):
- /* FALLTHROUGH */
- case ('%'):
- /* FALLTHROUGH */
- case ('0'):
- /* FALLTHROUGH */
- case (' '):
- /* FALLTHROUGH */
- case ('}'):
- /* FALLTHROUGH */
- case ('|'):
- /* FALLTHROUGH */
- case ('&'):
- /* FALLTHROUGH */
- case ('.'):
- /* FALLTHROUGH */
- case (':'):
- /* FALLTHROUGH */
- case ('c'):
- /* FALLTHROUGH */
- case ('e'):
- return(spec_norm(sv, 2));
- case ('s'):
- if ('\0' == *++p)
- return(spec_norm(sv, 2));
-
- c = 2;
- terminator = 0;
- lim = 1;
-
- if (*p == '\'') {
- lim = 0;
- terminator = 1;
- ++p;
- ++c;
- } else if (*p == '[') {
- lim = 0;
- terminator = 2;
- ++p;
- ++c;
- } else if (*p == '(') {
- lim = 2;
- terminator = 3;
- ++p;
- ++c;
- }
+ /*
+ * When the caller doesn't provide return storage,
+ * use local storage.
+ */
- if (*p == '+' || *p == '-') {
- ++p;
- ++c;
- }
+ if (NULL == start)
+ start = &local_start;
+ if (NULL == sz)
+ sz = &local_sz;
- if (*p == '\'') {
- if (terminator)
- return(spec_norm(sv, 0));
- lim = 0;
- terminator = 1;
- ++p;
- ++c;
- } else if (*p == '[') {
- if (terminator)
- return(spec_norm(sv, 0));
- lim = 0;
- terminator = 2;
- ++p;
- ++c;
- } else if (*p == '(') {
- if (terminator)
- return(spec_norm(sv, 0));
- lim = 2;
- terminator = 3;
- ++p;
- ++c;
- }
+ /*
+ * Beyond the backslash, at least one input character
+ * is part of the escape sequence. With one exception
+ * (see below), that character won't be returned.
+ */
- /* TODO: needs to handle floating point. */
+ gly = ESCAPE_ERROR;
+ *start = ++*end;
+ *sz = 0;
+ term = '\0';
- if ( ! isdigit((u_char)*p))
- return(spec_norm(sv, 0));
+ switch ((*start)[-1]) {
+ /*
+ * First the glyphs. There are several different forms of
+ * these, but each eventually returns a substring of the glyph
+ * name.
+ */
+ case '(':
+ gly = ESCAPE_SPECIAL;
+ *sz = 2;
+ break;
+ case '[':
+ gly = ESCAPE_SPECIAL;
+ term = ']';
+ break;
+ case 'C':
+ if ('\'' != **start)
+ return ESCAPE_ERROR;
+ *start = ++*end;
+ gly = ESCAPE_SPECIAL;
+ term = '\'';
+ break;
- for (i = 0; isdigit((u_char)*p); i++) {
- if (lim && i >= lim)
- break;
- ++p;
- ++c;
- }
+ /*
+ * Escapes taking no arguments at all.
+ */
+ case 'd':
+ case 'u':
+ case ',':
+ case '/':
+ return ESCAPE_IGNORE;
- if (terminator && terminator < 3) {
- if (1 == terminator && *p != '\'')
- return(spec_norm(sv, 0));
- if (2 == terminator && *p != ']')
- return(spec_norm(sv, 0));
- ++p;
- ++c;
- }
+ /*
+ * The \z escape is supposed to output the following
+ * character without advancing the cursor position.
+ * Since we are mostly dealing with terminal mode,
+ * let us just skip the next character.
+ */
+ case 'z':
+ return ESCAPE_SKIPCHAR;
- return(spec_norm(sv, c));
- case ('f'):
- /* FALLTHROUGH */
- case ('F'):
+ /*
+ * Handle all triggers matching \X(xy, \Xx, and \X[xxxx], where
+ * 'X' is the trigger. These have opaque sub-strings.
+ */
+ case 'F':
+ case 'g':
+ case 'k':
+ case 'M':
+ case 'm':
+ case 'n':
+ case 'V':
+ case 'Y':
+ gly = ESCAPE_IGNORE;
/* FALLTHROUGH */
- case ('*'):
- if ('\0' == *++p || isspace((u_char)*p))
- return(spec_norm(sv, 0));
- switch (*p) {
- case ('('):
- if ('\0' == *++p || isspace((u_char)*p))
- return(spec_norm(sv, 0));
- return(spec_norm(sv, 4));
- case ('['):
- for (c = 3, p++; *p && ']' != *p; p++, c++)
- if (isspace((u_char)*p))
- break;
- return(spec_norm(sv, *p == ']' ? c : 0));
+ case 'f':
+ if (ESCAPE_ERROR == gly)
+ gly = ESCAPE_FONT;
+ switch (**start) {
+ case '(':
+ *start = ++*end;
+ *sz = 2;
+ break;
+ case '[':
+ *start = ++*end;
+ term = ']';
+ break;
default:
+ *sz = 1;
break;
}
- return(spec_norm(sv, 3));
- case ('('):
- if ('\0' == *++p || isspace((u_char)*p))
- return(spec_norm(sv, 0));
- if ('\0' == *++p || isspace((u_char)*p))
- return(spec_norm(sv, 0));
- return(spec_norm(sv, 4));
- case ('['):
break;
- default:
- return(spec_norm(sv, 0));
- }
-
- for (c = 3, p++; *p && ']' != *p; p++, c++)
- if (isspace((u_char)*p))
- break;
- return(spec_norm(sv, *p == ']' ? c : 0));
-}
+ /*
+ * These escapes are of the form \X'Y', where 'X' is the trigger
+ * and 'Y' is any string. These have opaque sub-strings.
+ * The \B and \w escapes are handled in roff.c, roff_res().
+ */
+ case 'A':
+ case 'b':
+ case 'D':
+ case 'R':
+ case 'X':
+ case 'Z':
+ gly = ESCAPE_IGNORE;
+ /* FALLTHROUGH */
+ case 'o':
+ if (**start == '\0')
+ return ESCAPE_ERROR;
+ if (gly == ESCAPE_ERROR)
+ gly = ESCAPE_OVERSTRIKE;
+ term = **start;
+ *start = ++*end;
+ break;
+ /*
+ * These escapes are of the form \X'N', where 'X' is the trigger
+ * and 'N' resolves to a numerical expression.
+ */
+ case 'h':
+ case 'H':
+ case 'L':
+ case 'l':
+ case 'S':
+ case 'v':
+ case 'x':
+ if (strchr(" %&()*+-./0123456789:<=>", **start)) {
+ if ('\0' != **start)
+ ++*end;
+ return ESCAPE_ERROR;
+ }
+ gly = ESCAPE_IGNORE;
+ term = **start;
+ *start = ++*end;
+ break;
-void *
-mandoc_calloc(size_t num, size_t size)
-{
- void *ptr;
+ /*
+ * Special handling for the numbered character escape.
+ * XXX Do any other escapes need similar handling?
+ */
+ case 'N':
+ if ('\0' == **start)
+ return ESCAPE_ERROR;
+ (*end)++;
+ if (isdigit((unsigned char)**start)) {
+ *sz = 1;
+ return ESCAPE_IGNORE;
+ }
+ (*start)++;
+ while (isdigit((unsigned char)**end))
+ (*end)++;
+ *sz = *end - *start;
+ if ('\0' != **end)
+ (*end)++;
+ return ESCAPE_NUMBERED;
- ptr = calloc(num, size);
- if (NULL == ptr) {
- perror(NULL);
- exit(EXIT_FAILURE);
- }
+ /*
+ * Sizes get a special category of their own.
+ */
+ case 's':
+ gly = ESCAPE_IGNORE;
- return(ptr);
-}
+ /* See +/- counts as a sign. */
+ if ('+' == **end || '-' == **end || ASCII_HYPH == **end)
+ *start = ++*end;
+ switch (**end) {
+ case '(':
+ *start = ++*end;
+ *sz = 2;
+ break;
+ case '[':
+ *start = ++*end;
+ term = ']';
+ break;
+ case '\'':
+ *start = ++*end;
+ term = '\'';
+ break;
+ case '3':
+ case '2':
+ case '1':
+ *sz = (*end)[-1] == 's' &&
+ isdigit((unsigned char)(*end)[1]) ? 2 : 1;
+ break;
+ default:
+ *sz = 1;
+ break;
+ }
-void *
-mandoc_malloc(size_t size)
-{
- void *ptr;
+ break;
- ptr = malloc(size);
- if (NULL == ptr) {
- perror(NULL);
- exit(EXIT_FAILURE);
+ /*
+ * Anything else is assumed to be a glyph.
+ * In this case, pass back the character after the backslash.
+ */
+ default:
+ gly = ESCAPE_SPECIAL;
+ *start = --*end;
+ *sz = 1;
+ break;
}
- return(ptr);
-}
+ assert(ESCAPE_ERROR != gly);
+ /*
+ * Read up to the terminating character,
+ * paying attention to nested escapes.
+ */
-void *
-mandoc_realloc(void *ptr, size_t size)
-{
+ if ('\0' != term) {
+ while (**end != term) {
+ switch (**end) {
+ case '\0':
+ return ESCAPE_ERROR;
+ case '\\':
+ (*end)++;
+ if (ESCAPE_ERROR ==
+ mandoc_escape(end, NULL, NULL))
+ return ESCAPE_ERROR;
+ break;
+ default:
+ (*end)++;
+ break;
+ }
+ }
+ *sz = (*end)++ - *start;
+ } else {
+ assert(*sz > 0);
+ if ((size_t)*sz > strlen(*start))
+ return ESCAPE_ERROR;
+ *end += *sz;
+ }
+
+ /* Run post-processors. */
+
+ switch (gly) {
+ case ESCAPE_FONT:
+ if (2 == *sz) {
+ if ('C' == **start) {
+ /*
+ * Treat constant-width font modes
+ * just like regular font modes.
+ */
+ (*start)++;
+ (*sz)--;
+ } else {
+ if ('B' == (*start)[0] && 'I' == (*start)[1])
+ gly = ESCAPE_FONTBI;
+ break;
+ }
+ } else if (1 != *sz)
+ break;
- ptr = realloc(ptr, size);
- if (NULL == ptr) {
- perror(NULL);
- exit(EXIT_FAILURE);
+ switch (**start) {
+ case '3':
+ case 'B':
+ gly = ESCAPE_FONTBOLD;
+ break;
+ case '2':
+ case 'I':
+ gly = ESCAPE_FONTITALIC;
+ break;
+ case 'P':
+ gly = ESCAPE_FONTPREV;
+ break;
+ case '1':
+ case 'R':
+ gly = ESCAPE_FONTROMAN;
+ break;
+ }
+ break;
+ case ESCAPE_SPECIAL:
+ if (1 == *sz && 'c' == **start)
+ gly = ESCAPE_NOSPACE;
+ /*
+ * Unicode escapes are defined in groff as \[u0000]
+ * to \[u10FFFF], where the contained value must be
+ * a valid Unicode codepoint. Here, however, only
+ * check the length and range.
+ */
+ if (**start != 'u' || *sz < 5 || *sz > 7)
+ break;
+ if (*sz == 7 && ((*start)[1] != '1' || (*start)[2] != '0'))
+ break;
+ if (*sz == 6 && (*start)[1] == '0')
+ break;
+ if (*sz == 5 && (*start)[1] == 'D' &&
+ strchr("89ABCDEF", (*start)[2]) != NULL)
+ break;
+ if ((int)strspn(*start + 1, "0123456789ABCDEFabcdef")
+ + 1 == *sz)
+ gly = ESCAPE_UNICODE;
+ break;
+ default:
+ break;
}
- return(ptr);
+ return gly;
}
-
+/*
+ * Parse a quoted or unquoted roff-style request or macro argument.
+ * Return a pointer to the parsed argument, which is either the original
+ * pointer or advanced by one byte in case the argument is quoted.
+ * NUL-terminate the argument in place.
+ * Collapse pairs of quotes inside quoted arguments.
+ * Advance the argument pointer to the next argument,
+ * or to the NUL byte terminating the argument line.
+ */
char *
-mandoc_strdup(const char *ptr)
+mandoc_getarg(struct mparse *parse, char **cpp, int ln, int *pos)
{
- char *p;
+ char *start, *cp;
+ int quoted, pairs, white;
+
+ /* Quoting can only start with a new word. */
+ start = *cpp;
+ quoted = 0;
+ if ('"' == *start) {
+ quoted = 1;
+ start++;
+ }
- p = strdup(ptr);
- if (NULL == p) {
- perror(NULL);
- exit(EXIT_FAILURE);
+ pairs = 0;
+ white = 0;
+ for (cp = start; '\0' != *cp; cp++) {
+
+ /*
+ * Move the following text left
+ * after quoted quotes and after "\\" and "\t".
+ */
+ if (pairs)
+ cp[-pairs] = cp[0];
+
+ if ('\\' == cp[0]) {
+ /*
+ * In copy mode, translate double to single
+ * backslashes and backslash-t to literal tabs.
+ */
+ switch (cp[1]) {
+ case 't':
+ cp[0] = '\t';
+ /* FALLTHROUGH */
+ case '\\':
+ pairs++;
+ cp++;
+ break;
+ case ' ':
+ /* Skip escaped blanks. */
+ if (0 == quoted)
+ cp++;
+ break;
+ default:
+ break;
+ }
+ } else if (0 == quoted) {
+ if (' ' == cp[0]) {
+ /* Unescaped blanks end unquoted args. */
+ white = 1;
+ break;
+ }
+ } else if ('"' == cp[0]) {
+ if ('"' == cp[1]) {
+ /* Quoted quotes collapse. */
+ pairs++;
+ cp++;
+ } else {
+ /* Unquoted quotes end quoted args. */
+ quoted = 2;
+ break;
+ }
+ }
}
- return(p);
-}
+ /* Quoted argument without a closing quote. */
+ if (1 == quoted)
+ mandoc_msg(MANDOCERR_ARG_QUOTE, parse, ln, *pos, NULL);
+
+ /* NUL-terminate this argument and move to the next one. */
+ if (pairs)
+ cp[-pairs] = '\0';
+ if ('\0' != *cp) {
+ *cp++ = '\0';
+ while (' ' == *cp)
+ cp++;
+ }
+ *pos += (int)(cp - start) + (quoted ? 1 : 0);
+ *cpp = cp;
+ if ('\0' == *cp && (white || ' ' == cp[-1]))
+ mandoc_msg(MANDOCERR_SPACE_EOL, parse, ln, *pos, NULL);
+
+ return start;
+}
static int
a2time(time_t *t, const char *fmt, const char *p)
memset(&tm, 0, sizeof(struct tm));
+ pp = NULL;
+#if HAVE_STRPTIME
pp = strptime(p, fmt, &tm);
+#endif
if (NULL != pp && '\0' == *pp) {
*t = mktime(&tm);
- return(1);
+ return 1;
}
- return(0);
+ return 0;
}
+static char *
+time2a(time_t t)
+{
+ struct tm *tm;
+ char *buf, *p;
+ size_t ssz;
+ int isz;
-/*
- * Convert from a manual date string (see mdoc(7) and man(7)) into a
- * date according to the stipulated date type.
- */
-time_t
-mandoc_a2time(int flags, const char *p)
+ tm = localtime(&t);
+ if (tm == NULL)
+ return NULL;
+
+ /*
+ * Reserve space:
+ * up to 9 characters for the month (September) + blank
+ * up to 2 characters for the day + comma + blank
+ * 4 characters for the year and a terminating '\0'
+ */
+
+ p = buf = mandoc_malloc(10 + 4 + 4 + 1);
+
+ if ((ssz = strftime(p, 10 + 1, "%B ", tm)) == 0)
+ goto fail;
+ p += (int)ssz;
+
+ /*
+ * The output format is just "%d" here, not "%2d" or "%02d".
+ * That's also the reason why we can't just format the
+ * date as a whole with "%B %e, %Y" or "%B %d, %Y".
+ * Besides, the present approach is less prone to buffer
+ * overflows, in case anybody should ever introduce the bug
+ * of looking at LC_TIME.
+ */
+
+ if ((isz = snprintf(p, 4 + 1, "%d, ", tm->tm_mday)) == -1)
+ goto fail;
+ p += isz;
+
+ if (strftime(p, 4 + 1, "%Y", tm) == 0)
+ goto fail;
+ return buf;
+
+fail:
+ free(buf);
+ return NULL;
+}
+
+char *
+mandoc_normdate(struct mparse *parse, char *in, int ln, int pos)
{
time_t t;
- if (MTIME_MDOCDATE & flags) {
- if (0 == strcmp(p, "$" "Mdocdate$"))
- return(time(NULL));
- if (a2time(&t, "$" "Mdocdate: %b %d %Y $", p))
- return(t);
+ /* No date specified: use today's date. */
+
+ if (in == NULL || *in == '\0' || strcmp(in, "$" "Mdocdate$") == 0) {
+ mandoc_msg(MANDOCERR_DATE_MISSING, parse, ln, pos, NULL);
+ return time2a(time(NULL));
}
- if (MTIME_CANONICAL & flags || MTIME_REDUCED & flags)
- if (a2time(&t, "%b %d, %Y", p))
- return(t);
+ /* Valid mdoc(7) date format. */
- if (MTIME_ISO_8601 & flags)
- if (a2time(&t, "%Y-%m-%d", p))
- return(t);
+ if (a2time(&t, "$" "Mdocdate: %b %d %Y $", in) ||
+ a2time(&t, "%b %d, %Y", in))
+ return time2a(t);
- if (MTIME_REDUCED & flags) {
- if (a2time(&t, "%d, %Y", p))
- return(t);
- if (a2time(&t, "%Y", p))
- return(t);
- }
+ /* Do not warn about the legacy man(7) format. */
- return(0);
-}
+ if ( ! a2time(&t, "%Y-%m-%d", in))
+ mandoc_msg(MANDOCERR_DATE_BAD, parse, ln, pos, in);
+ /* Use any non-mdoc(7) date verbatim. */
+
+ return mandoc_strdup(in);
+}
int
mandoc_eos(const char *p, size_t sz)
{
+ const char *q;
+ int enclosed, found;
if (0 == sz)
- return(0);
+ return 0;
/*
* End-of-sentence recognition must include situations where
* some symbols, such as `)', allow prior EOS punctuation to
- * propogate outward.
+ * propagate outward.
*/
- for ( ; sz; sz--) {
- switch (p[(int)sz - 1]) {
- case ('\"'):
- /* FALLTHROUGH */
- case ('\''):
- /* FALLTHROUGH */
- case (']'):
- /* FALLTHROUGH */
- case (')'):
+ enclosed = found = 0;
+ for (q = p + (int)sz - 1; q >= p; q--) {
+ switch (*q) {
+ case '\"':
+ case '\'':
+ case ']':
+ case ')':
+ if (0 == found)
+ enclosed = 1;
+ break;
+ case '.':
+ case '!':
+ case '?':
+ found = 1;
break;
- case ('.'):
- /* Escaped periods. */
- if (sz > 1 && '\\' == p[(int)sz - 2])
- return(0);
- /* FALLTHROUGH */
- case ('!'):
- /* FALLTHROUGH */
- case ('?'):
- return(1);
default:
- return(0);
+ return found &&
+ (!enclosed || isalnum((unsigned char)*q));
}
}
- return(0);
+ return found && !enclosed;
}
-
+/*
+ * Convert a string to a long that may not be <0.
+ * If the string is invalid, or is less than 0, return -1.
+ */
int
-mandoc_hyph(const char *start, const char *c)
+mandoc_strntoi(const char *p, size_t sz, int base)
{
+ char buf[32];
+ char *ep;
+ long v;
- /*
- * Choose whether to break at a hyphenated character. We only
- * do this if it's free-standing within a word.
- */
+ if (sz > 31)
+ return -1;
+
+ memcpy(buf, p, sz);
+ buf[(int)sz] = '\0';
+
+ errno = 0;
+ v = strtol(buf, &ep, base);
+
+ if (buf[0] == '\0' || *ep != '\0')
+ return -1;
+
+ if (v > INT_MAX)
+ v = INT_MAX;
+ if (v < INT_MIN)
+ v = INT_MIN;
- /* Skip first/last character of buffer. */
- if (c == start || '\0' == *(c + 1))
- return(0);
- /* Skip first/last character of word. */
- if ('\t' == *(c + 1) || '\t' == *(c - 1))
- return(0);
- if (' ' == *(c + 1) || ' ' == *(c - 1))
- return(0);
- /* Skip double invocations. */
- if ('-' == *(c + 1) || '-' == *(c - 1))
- return(0);
- /* Skip escapes. */
- if ('\\' == *(c - 1))
- return(0);
-
- return(1);
+ return (int)v;
}
git.ckatri.com / mandoc.git / blobdiff