-/* $Id: mandoc.c,v 1.38 2011/03/15 03:03:54 schwarze Exp $ */
+/* $Id: mandoc.c,v 1.119 2021/08/10 12:55:03 schwarze Exp $ */
/*
- * Copyright (c) 2008, 2009, 2010 Kristaps Dzonsons <kristaps@bsd.lv>
- * Copyright (c) 2011 Ingo Schwarze <schwarze@openbsd.org>
+ * Copyright (c) 2008-2011, 2014 Kristaps Dzonsons <kristaps@bsd.lv>
+ * Copyright (c) 2011-2015, 2017-2021 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
* 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_aux.h"
#include "mandoc.h"
+#include "roff.h"
#include "libmandoc.h"
-
-#define DATESIZE 32
+#include "roff_int.h"
static int a2time(time_t *, const char *, const char *);
static char *time2a(time_t);
-int
-mandoc_special(char *p)
+
+enum mandoc_esc
+mandoc_font(const char *cp, int sz)
{
- int len, i;
+ switch (sz) {
+ case 0:
+ return ESCAPE_FONTPREV;
+ case 1:
+ switch (cp[0]) {
+ case 'B':
+ case '3':
+ return ESCAPE_FONTBOLD;
+ case 'I':
+ case '2':
+ return ESCAPE_FONTITALIC;
+ case 'P':
+ return ESCAPE_FONTPREV;
+ case 'R':
+ case '1':
+ return ESCAPE_FONTROMAN;
+ case '4':
+ return ESCAPE_FONTBI;
+ default:
+ return ESCAPE_ERROR;
+ }
+ case 2:
+ switch (cp[0]) {
+ case 'B':
+ switch (cp[1]) {
+ case 'I':
+ return ESCAPE_FONTBI;
+ default:
+ return ESCAPE_ERROR;
+ }
+ case 'C':
+ switch (cp[1]) {
+ case 'B':
+ return ESCAPE_FONTCB;
+ case 'I':
+ return ESCAPE_FONTCI;
+ case 'R':
+ case 'W':
+ return ESCAPE_FONTCR;
+ default:
+ return ESCAPE_ERROR;
+ }
+ default:
+ return ESCAPE_ERROR;
+ }
+ default:
+ return ESCAPE_ERROR;
+ }
+}
+
+enum mandoc_esc
+mandoc_escape(const char **end, const char **start, int *sz)
+{
+ const char *local_start;
+ int local_sz, c, i;
char term;
- char *sv;
-
- len = 0;
- term = '\0';
- sv = p;
+ enum mandoc_esc gly;
- assert('\\' == *p);
- p++;
+ /*
+ * When the caller doesn't provide return storage,
+ * use local storage.
+ */
- switch (*p++) {
-#if 0
- case ('Z'):
- /* FALLTHROUGH */
- case ('X'):
- /* FALLTHROUGH */
- case ('x'):
- /* FALLTHROUGH */
- case ('S'):
- /* FALLTHROUGH */
- case ('R'):
- /* FALLTHROUGH */
- case ('N'):
- /* FALLTHROUGH */
- case ('l'):
- /* FALLTHROUGH */
- case ('L'):
- /* FALLTHROUGH */
- case ('H'):
- /* FALLTHROUGH */
- case ('h'):
- /* FALLTHROUGH */
- case ('D'):
- /* FALLTHROUGH */
- case ('C'):
- /* FALLTHROUGH */
- case ('b'):
- /* FALLTHROUGH */
- case ('B'):
- /* FALLTHROUGH */
- case ('a'):
- /* FALLTHROUGH */
- case ('A'):
- if (*p++ != '\'')
- return(0);
- term = '\'';
+ if (NULL == start)
+ start = &local_start;
+ if (NULL == sz)
+ sz = &local_sz;
+
+ /*
+ * Treat "\E" just like "\";
+ * it only makes a difference in copy mode.
+ */
+
+ if (**end == 'E')
+ ++*end;
+
+ /*
+ * 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.
+ */
+
+ gly = ESCAPE_ERROR;
+ *start = ++*end;
+ *sz = 0;
+ term = '\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;
-#endif
- case ('h'):
- /* FALLTHROUGH */
- case ('v'):
- /* FALLTHROUGH */
- case ('s'):
- if (ASCII_HYPH == *p)
- *p = '-';
-
- i = 0;
- if ('+' == *p || '-' == *p) {
- p++;
- i = 1;
+ case '[':
+ if (**start == ' ') {
+ ++*end;
+ return ESCAPE_ERROR;
}
+ gly = ESCAPE_SPECIAL;
+ term = ']';
+ break;
+ case 'C':
+ if ('\'' != **start)
+ return ESCAPE_ERROR;
+ *start = ++*end;
+ gly = ESCAPE_SPECIAL;
+ term = '\'';
+ break;
- switch (*p++) {
- case ('('):
- len = 2;
- break;
- case ('['):
- term = ']';
+ /*
+ * Escapes taking no arguments at all.
+ */
+ case '!':
+ case '?':
+ return ESCAPE_UNSUPP;
+ case '%':
+ case '&':
+ case ')':
+ case ',':
+ case '/':
+ case '^':
+ case 'a':
+ case 'd':
+ case 'r':
+ case 't':
+ case 'u':
+ case '{':
+ case '|':
+ case '}':
+ return ESCAPE_IGNORE;
+ case 'c':
+ return ESCAPE_NOSPACE;
+ case 'p':
+ return ESCAPE_BREAK;
+
+ /*
+ * 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;
+
+ /*
+ * Handle all triggers matching \X(xy, \Xx, and \X[xxxx], where
+ * 'X' is the trigger. These have opaque sub-strings.
+ */
+ case 'F':
+ case 'f':
+ case 'g':
+ case 'k':
+ case 'M':
+ case 'm':
+ case 'n':
+ case 'O':
+ case 'V':
+ case 'Y':
+ case '*':
+ switch ((*start)[-1]) {
+ case 'f':
+ gly = ESCAPE_FONT;
break;
- case ('\''):
- term = '\'';
+ case '*':
+ gly = ESCAPE_DEVICE;
break;
- case ('0'):
- i = 1;
- /* FALLTHROUGH */
default:
- len = 1;
- p--;
+ gly = ESCAPE_IGNORE;
break;
}
-
- if (ASCII_HYPH == *p)
- *p = '-';
- if ('+' == *p || '-' == *p) {
- if (i)
- return(0);
- p++;
- }
-
- /* Handle embedded numerical subexp or escape. */
-
- if ('(' == *p) {
- while (*p && ')' != *p)
- if ('\\' == *p++) {
- i = mandoc_special(--p);
- if (0 == i)
- return(0);
- p += i;
- }
-
- if (')' == *p++)
- break;
-
- return(0);
- } else if ('\\' == *p) {
- if (0 == (i = mandoc_special(p)))
- return(0);
- p += i;
- }
-
- break;
-#if 0
- case ('Y'):
- /* FALLTHROUGH */
- case ('V'):
- /* FALLTHROUGH */
- case ('$'):
- /* FALLTHROUGH */
- case ('n'):
- /* FALLTHROUGH */
-#endif
- case ('k'):
- /* FALLTHROUGH */
- case ('M'):
- /* FALLTHROUGH */
- case ('m'):
- /* FALLTHROUGH */
- case ('f'):
- /* FALLTHROUGH */
- case ('F'):
- /* FALLTHROUGH */
- case ('*'):
- switch (*p++) {
- case ('('):
- len = 2;
+ switch (**start) {
+ case '(':
+ if ((*start)[-1] == 'O')
+ gly = ESCAPE_ERROR;
+ *start = ++*end;
+ *sz = 2;
break;
- case ('['):
+ case '[':
+ if ((*start)[-1] == 'O')
+ gly = (*start)[1] == '5' ?
+ ESCAPE_UNSUPP : ESCAPE_ERROR;
+ *start = ++*end;
term = ']';
break;
default:
- len = 1;
- p--;
+ if ((*start)[-1] == 'O') {
+ switch (**start) {
+ case '0':
+ gly = ESCAPE_UNSUPP;
+ break;
+ case '1':
+ case '2':
+ case '3':
+ case '4':
+ break;
+ default:
+ gly = ESCAPE_ERROR;
+ break;
+ }
+ }
+ *sz = 1;
break;
}
break;
- case ('('):
- len = 2;
- break;
- case ('['):
- term = ']';
+
+ /*
+ * 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;
- case ('z'):
- len = 1;
- if ('\\' == *p) {
- if (0 == (i = mandoc_special(p)))
- return(0);
- p += i;
- return(*p ? (int)(p - sv) : 0);
+
+ /*
+ * 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;
}
- break;
- case ('o'):
- /* FALLTHROUGH */
- case ('w'):
- if ('\'' == *p++) {
- term = '\'';
+ switch ((*start)[-1]) {
+ case 'h':
+ gly = ESCAPE_HORIZ;
+ break;
+ case 'l':
+ gly = ESCAPE_HLINE;
+ break;
+ default:
+ gly = ESCAPE_IGNORE;
break;
}
- /* FALLTHROUGH */
- default:
- len = 1;
- p--;
+ term = **start;
+ *start = ++*end;
break;
- }
-
- if (term) {
- for ( ; *p && term != *p; p++)
- if (ASCII_HYPH == *p)
- *p = '-';
- return(*p ? (int)(p - sv) : 0);
- }
- for (i = 0; *p && i < len; i++, p++)
- if (ASCII_HYPH == *p)
- *p = '-';
- return(i == len ? (int)(p - sv) : 0);
-}
-
-
-void *
-mandoc_calloc(size_t num, size_t size)
-{
- void *ptr;
-
- ptr = calloc(num, size);
- if (NULL == ptr) {
- perror(NULL);
- exit((int)MANDOCLEVEL_SYSERR);
- }
-
- return(ptr);
-}
-
-
-void *
-mandoc_malloc(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 = malloc(size);
- if (NULL == ptr) {
- perror(NULL);
- exit((int)MANDOCLEVEL_SYSERR);
- }
+ /*
+ * 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_realloc(void *ptr, size_t size)
-{
+ break;
- ptr = realloc(ptr, size);
- if (NULL == ptr) {
- perror(NULL);
- exit((int)MANDOCLEVEL_SYSERR);
+ /*
+ * Several special characters can be encoded as
+ * one-byte escape sequences without using \[].
+ */
+ case ' ':
+ case '\'':
+ case '-':
+ case '.':
+ case '0':
+ case ':':
+ case '_':
+ case '`':
+ case 'e':
+ case '~':
+ gly = ESCAPE_SPECIAL;
+ /* FALLTHROUGH */
+ default:
+ if (gly == ESCAPE_ERROR)
+ gly = ESCAPE_UNDEF;
+ *start = --*end;
+ *sz = 1;
+ break;
}
- return(ptr);
-}
-
-
-char *
-mandoc_strdup(const char *ptr)
-{
- char *p;
+ /*
+ * Read up to the terminating character,
+ * paying attention to nested escapes.
+ */
- p = strdup(ptr);
- if (NULL == p) {
- perror(NULL);
- exit((int)MANDOCLEVEL_SYSERR);
+ 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;
+
+ /*
+ * The file chars.c only provides one common list
+ * of character names, but \[-] == \- is the only
+ * one of the characters with one-byte names that
+ * allows enclosing the name in brackets.
+ */
+ if (gly == ESCAPE_SPECIAL && *sz == 1 && **start != '-')
+ return ESCAPE_ERROR;
+ } else {
+ assert(*sz > 0);
+ if ((size_t)*sz > strlen(*start))
+ return ESCAPE_ERROR;
+ *end += *sz;
}
- return(p);
-}
+ /* Run post-processors. */
-/*
- * 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.
- * Null-terminate the argument in place.
- * Collapse pairs of quotes inside quoted arguments.
- * Advance the argument pointer to the next argument,
- * or to the null byte terminating the argument line.
- */
-char *
-mandoc_getarg(char **cpp, mandocmsg msg, void *data, int ln, int *pos)
-{
- char *start, *cp;
- int quoted, pairs, white;
-
- /* Quoting can only start with a new word. */
- start = *cpp;
- if ('"' == *start) {
- quoted = 1;
- start++;
- } else
- quoted = 0;
-
- pairs = 0;
- white = 0;
- for (cp = start; '\0' != *cp; cp++) {
- /* Move left after quoted quotes and escaped backslashes. */
- if (pairs)
- cp[-pairs] = cp[0];
- if ('\\' == cp[0]) {
- if ('\\' == cp[1]) {
- /* Poor man's copy mode. */
- pairs++;
- cp++;
- } else if (0 == quoted && ' ' == cp[1])
- /* Skip escaped blanks. */
- cp++;
- } else if (0 == quoted) {
- if (' ' == cp[0]) {
- /* Unescaped blanks end unquoted args. */
- white = 1;
+ switch (gly) {
+ case ESCAPE_FONT:
+ gly = mandoc_font(*start, *sz);
+ break;
+ case ESCAPE_SPECIAL:
+ if (**start == 'c') {
+ if (*sz < 6 || *sz > 7 ||
+ strncmp(*start, "char", 4) != 0 ||
+ (int)strspn(*start + 4, "0123456789") + 4 < *sz)
break;
- }
- } else if ('"' == cp[0]) {
- if ('"' == cp[1]) {
- /* Quoted quotes collapse. */
- pairs++;
- cp++;
- } else {
- /* Unquoted quotes end quoted args. */
- quoted = 2;
+ c = 0;
+ for (i = 4; i < *sz; i++)
+ c = 10 * c + ((*start)[i] - '0');
+ if (c < 0x21 || (c > 0x7e && c < 0xa0) || c > 0xff)
break;
- }
+ *start += 4;
+ *sz -= 4;
+ gly = ESCAPE_NUMBERED;
+ break;
}
- }
- /* Quoted argument without a closing quote. */
- if (1 == quoted && msg)
- (*msg)(MANDOCERR_BADQUOTE, data, ln, *pos, NULL);
-
- /* Null-terminate this argument and move to the next one. */
- if (pairs)
- cp[-pairs] = '\0';
- if ('\0' != *cp) {
- *cp++ = '\0';
- while (' ' == *cp)
- cp++;
+ /*
+ * 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;
+ case ESCAPE_DEVICE:
+ assert(*sz == 2 && (*start)[0] == '.' && (*start)[1] == 'T');
+ break;
+ default:
+ break;
}
- *pos += (cp - start) + (quoted ? 1 : 0);
- *cpp = cp;
- if ('\0' == *cp && msg && (white || ' ' == cp[-1]))
- (*msg)(MANDOCERR_EOLNSPACE, data, ln, *pos, NULL);
-
- return(start);
+ return gly;
}
-
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;
+ struct tm *tm;
char *buf, *p;
size_t ssz;
int isz;
- localtime_r(&t, &tm);
+ buf = NULL;
+ tm = localtime(&t);
+ if (tm == NULL)
+ goto fail;
/*
* Reserve space:
* 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 (0 == (ssz = strftime(p, 10 + 1, "%B ", &tm)))
+ if ((ssz = strftime(p, 10 + 1, "%B ", tm)) == 0)
goto fail;
p += (int)ssz;
- if (-1 == (isz = snprintf(p, 4 + 1, "%d, ", tm.tm_mday)))
+ /*
+ * 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.
+ */
+
+ isz = snprintf(p, 4 + 1, "%d, ", tm->tm_mday);
+ if (isz < 0 || isz > 4)
goto fail;
p += isz;
- if (0 == strftime(p, 4 + 1, "%Y", &tm))
+ if (strftime(p, 4 + 1, "%Y", tm) == 0)
goto fail;
- return(buf);
+ return buf;
fail:
free(buf);
- return(NULL);
+ return mandoc_strdup("");
}
-
char *
-mandoc_normdate(char *in, mandocmsg msg, void *data, int ln, int pos)
+mandoc_normdate(struct roff_node *nch, struct roff_node *nbl)
{
- char *out;
+ char *cp;
time_t t;
- if (NULL == in || '\0' == *in ||
- 0 == strcmp(in, "$" "Mdocdate$")) {
- (*msg)(MANDOCERR_NODATE, data, ln, pos, NULL);
- time(&t);
+ /* No date specified. */
+
+ if (nch == NULL) {
+ if (nbl == NULL)
+ mandoc_msg(MANDOCERR_DATE_MISSING, 0, 0, NULL);
+ else
+ mandoc_msg(MANDOCERR_DATE_MISSING, nbl->line,
+ nbl->pos, "%s", roff_name[nbl->tok]);
+ return mandoc_strdup("");
}
- else if (!a2time(&t, "$" "Mdocdate: %b %d %Y $", in) &&
- !a2time(&t, "%b %d, %Y", in) &&
- !a2time(&t, "%Y-%m-%d", in)) {
- (*msg)(MANDOCERR_BADDATE, data, ln, pos, NULL);
- t = 0;
+ if (*nch->string == '\0') {
+ mandoc_msg(MANDOCERR_DATE_MISSING, nch->line,
+ nch->pos, "%s", roff_name[nbl->tok]);
+ return mandoc_strdup("");
}
- out = t ? time2a(t) : NULL;
- return(out ? out : mandoc_strdup(in));
-}
+ if (strcmp(nch->string, "$" "Mdocdate$") == 0)
+ return time2a(time(NULL));
+
+ /* Valid mdoc(7) date format. */
+
+ if (a2time(&t, "$" "Mdocdate: %b %d %Y $", nch->string) ||
+ a2time(&t, "%b %d, %Y", nch->string)) {
+ cp = time2a(t);
+ if (t > time(NULL) + 86400)
+ mandoc_msg(MANDOCERR_DATE_FUTURE, nch->line,
+ nch->pos, "%s %s", roff_name[nbl->tok], cp);
+ else if (*nch->string != '$' &&
+ strcmp(nch->string, cp) != 0)
+ mandoc_msg(MANDOCERR_DATE_NORM, nch->line,
+ nch->pos, "%s %s", roff_name[nbl->tok], cp);
+ return cp;
+ }
+
+ /* In man(7), do not warn about the legacy format. */
+ if (a2time(&t, "%Y-%m-%d", nch->string) == 0)
+ mandoc_msg(MANDOCERR_DATE_BAD, nch->line, nch->pos,
+ "%s %s", roff_name[nbl->tok], nch->string);
+ else if (t > time(NULL) + 86400)
+ mandoc_msg(MANDOCERR_DATE_FUTURE, nch->line, nch->pos,
+ "%s %s", roff_name[nbl->tok], nch->string);
+ else if (nbl->tok == MDOC_Dd)
+ mandoc_msg(MANDOCERR_DATE_LEGACY, nch->line, nch->pos,
+ "Dd %s", nch->string);
+
+ /* Use any non-mdoc(7) date verbatim. */
+
+ return mandoc_strdup(nch->string);
+}
int
-mandoc_eos(const char *p, size_t sz, int enclosed)
+mandoc_eos(const char *p, size_t sz)
{
- const char *q;
- int found;
+ 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.
*/
- found = 0;
+ enclosed = found = 0;
for (q = p + (int)sz - 1; q >= p; q--) {
switch (*q) {
- case ('\"'):
- /* FALLTHROUGH */
- case ('\''):
- /* FALLTHROUGH */
- case (']'):
- /* FALLTHROUGH */
- case (')'):
+ case '\"':
+ case '\'':
+ case ']':
+ case ')':
if (0 == found)
enclosed = 1;
break;
- case ('.'):
- /* FALLTHROUGH */
- case ('!'):
- /* FALLTHROUGH */
- case ('?'):
+ case '.':
+ case '!':
+ case '?':
found = 1;
break;
default:
- return(found && (!enclosed || isalnum((unsigned char)*q)));
+ return found &&
+ (!enclosed || isalnum((unsigned char)*q));
}
}
- return(found && !enclosed);
+ 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