No longer use names that only occur in the SYNOPSIS section as names

[mandoc.git] / mandoc.c

diff --git a/mandoc.c b/mandoc.c
index 751eef6b3d9e536bd91f8826a8f60ff9d65b4e5f..3e16d2c64fbc616d6182274560d66a9e2fd85ccb 100644 (file)
--- a/mandoc.c
+++ b/mandoc.c
@@ -1,255 +1,458 @@
-/*     $Id: mandoc.c,v 1.9 2010/01/01 17:14:28 kristaps Exp $ */
+/*     $Id: mandoc.c,v 1.103 2017/07/03 13:40:19 schwarze Exp $ */

 /*
 /*

- * Copyright (c) 2008, 2009 Kristaps Dzonsons <kristaps@kth.se>
+ * Copyright (c) 2008-2011, 2014 Kristaps Dzonsons <kristaps@bsd.lv>
+ * Copyright (c) 2011-2015, 2017 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.
  *
  *
  * 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
  * 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.
  */
  * 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"
 #include "config.h"

-#endif

 
 #include <sys/types.h>
 
 #include <assert.h>
 #include <ctype.h>
 
 #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 <stdlib.h>
 #include <stdio.h>
 #include <string.h>
 #include <time.h>
 

+#include "mandoc_aux.h"
+#include "mandoc.h"
+#include "roff.h"

 #include "libmandoc.h"
 
 #include "libmandoc.h"
 

-static int      a2time(time_t *, const char *, const char *);
+static int      a2time(time_t *, const char *, const char *);
+static char    *time2a(time_t);

 
 
 
 

-int
-mandoc_special(const char *p)
+enum mandoc_esc
+mandoc_escape(const char **end, const char **start, int *sz)

 {
 {

-       int              terminator;    /* Terminator for \s. */
-       int              lim;           /* Limit for N in \s. */
-       int              c, i;
-       
-       if ('\\' != *p++)
-               return(0);
-
-       switch (*p) {
-       case ('\\'):
-               /* FALLTHROUGH */
-       case ('\''):
-               /* FALLTHROUGH */
-       case ('`'):
-               /* FALLTHROUGH */
-       case ('q'):
-               /* FALLTHROUGH */
-       case ('-'):
-               /* FALLTHROUGH */
-       case ('~'):
-               /* FALLTHROUGH */
-       case ('^'):
-               /* FALLTHROUGH */
-       case ('%'):
-               /* FALLTHROUGH */
-       case ('0'):
-               /* FALLTHROUGH */
-       case (' '):
-               /* FALLTHROUGH */
-       case ('|'):
-               /* FALLTHROUGH */
-       case ('&'):
-               /* FALLTHROUGH */
-       case ('.'):
-               /* FALLTHROUGH */
-       case (':'):
+       const char      *local_start;
+       int              local_sz;
+       char             term;
+       enum mandoc_esc  gly;
+
+       /*
+        * When the caller doesn't provide return storage,
+        * use local storage.
+        */
+
+       if (NULL == start)
+               start = &local_start;
+       if (NULL == sz)
+               sz = &local_sz;
+
+       /*
+        * 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;
+       case '[':
+               gly = ESCAPE_SPECIAL;
+               term = ']';
+               break;
+       case 'C':
+               if ('\'' != **start)
+                       return ESCAPE_ERROR;
+               *start = ++*end;
+               gly = ESCAPE_SPECIAL;
+               term = '\'';
+               break;
+
+       /*
+        * Escapes taking no arguments at all.
+        */
+       case 'd':
+       case 'u':
+       case ',':
+       case '/':
+               return ESCAPE_IGNORE;
+       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 'g':
+       case 'k':
+       case 'M':
+       case 'm':
+       case 'n':
+       case 'V':
+       case 'Y':
+               gly = ESCAPE_IGNORE;

                /* FALLTHROUGH */
                /* FALLTHROUGH */

-       case ('c'):
-               return(2);
-       case ('e'):
-               return(2);
-       case ('f'):
-               if ('\0' == *++p || ! isgraph((u_char)*p))
-                       return(0);
-               return(3);
-       case ('s'):
-               if ('\0' == *++p)
-                       return(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;
+       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;

                }
                }

+               break;
+
+       /*
+        * 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;

 
 

-               if (*p == '+' || *p == '-') {
-                       ++p;
-                       ++c;
+       /*
+        * 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;
+               }
+               switch ((*start)[-1]) {
+               case 'h':
+                       gly = ESCAPE_HORIZ;
+                       break;
+               case 'l':
+                       gly = ESCAPE_HLINE;
+                       break;
+               default:
+                       gly = ESCAPE_IGNORE;
+                       break;

                }
                }

+               term = **start;
+               *start = ++*end;
+               break;

 
 

-               if (*p == '\'') {
-                       if (terminator)
-                               return(0);
-                       lim = 0;
-                       terminator = 1;
-                       ++p;
-                       ++c;
-               } else if (*p == '[') {
-                       if (terminator)
-                               return(0);
-                       lim = 0;
-                       terminator = 2;
-                       ++p;
-                       ++c;
-               } else if (*p == '(') {
-                       if (terminator)
-                               return(0);
-                       lim = 2;
-                       terminator = 3;
-                       ++p;
-                       ++c;
+       /*
+        * 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;
+
+       /*
+        * Sizes get a special category of their own.
+        */
+       case 's':
+               gly = ESCAPE_IGNORE;
+
+               /* 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;

                }
 
                }
 

-               /* TODO: needs to handle floating point. */
+               break;

 
 

-               if ( ! isdigit((u_char)*p))
-                       return(0);
+       /*
+        * 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;
+       }

 
 

-               for (i = 0; isdigit((u_char)*p); i++) {
-                       if (lim && i >= lim)
+       assert(ESCAPE_ERROR != gly);
+
+       /*
+        * Read up to the terminating character,
+        * paying attention to nested escapes.
+        */
+
+       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;
                                break;

-                       ++p;
-                       ++c;
+                       }

                }
                }

+               *sz = (*end)++ - *start;
+       } else {
+               assert(*sz > 0);
+               if ((size_t)*sz > strlen(*start))
+                       return ESCAPE_ERROR;
+               *end += *sz;
+       }

 
 

-               if (terminator && terminator < 3) {
-                       if (1 == terminator && *p != '\'')
-                               return(0);
-                       if (2 == terminator && *p != ']')
-                               return(0);
-                       ++p;
-                       ++c;
-               }
+       /* 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;

 
 

-               return(c);
-       case ('*'):
-               if (0 == *++p || ! isgraph((u_char)*p))
-                       return(0);
-               switch (*p) {
-               case ('('):
-                       if (0 == *++p || ! isgraph((u_char)*p))
-                               return(0);
-                       return(4);
-               case ('['):
-                       for (c = 3, p++; *p && ']' != *p; p++, c++)
-                               if ( ! isgraph((u_char)*p))
-                                       break;
-                       return(*p == ']' ? c : 0);
-               default:
+               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;
                }

-               return(3);
-       case ('('):
-               if (0 == *++p || ! isgraph((u_char)*p))
-                       return(0);
-               if (0 == *++p || ! isgraph((u_char)*p))
-                       return(0);
-               return(4);
-       case ('['):
+               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;
        default:

-               return(0);
+               break;

        }
 
        }
 

-       for (c = 3, p++; *p && ']' != *p; p++, c++)
-               if ( ! isgraph((u_char)*p))
-                       break;
-
-       return(*p == ']' ? c : 0);
+       return gly;

 }
 
 }
 

-
-void *
-mandoc_calloc(size_t num, size_t size)
+/*
+ * 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_getarg(struct mparse *parse, char **cpp, int ln, int *pos)

 {
 {

-       void            *ptr;
-
-       ptr = calloc(num, size);
-       if (NULL == ptr) {
-               perror(NULL);
-               exit(EXIT_FAILURE);
+       char     *start, *cp;
+       int       quoted, pairs, white;
+
+       /* Quoting can only start with a new word. */
+       start = *cpp;
+       quoted = 0;
+       if ('"' == *start) {
+               quoted = 1;
+               start++;

        }
 
        }
 

-       return(ptr);
-}
-
-
-void *
-mandoc_malloc(size_t size)
-{
-       void            *ptr;
-
-       ptr = malloc(size);
-       if (NULL == ptr) {
-               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(ptr);
-}
-
-
-void *
-mandoc_realloc(void *ptr, size_t size)
-{
-
-       ptr = realloc(ptr, size);
-       if (NULL == ptr) {
-               perror(NULL);
-               exit(EXIT_FAILURE);
+       /* 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;

 
 

-       return(ptr);
-}
+       if ('\0' == *cp && (white || ' ' == cp[-1]))
+               mandoc_msg(MANDOCERR_SPACE_EOL, parse, ln, *pos, NULL);

 
 

-
-char *
-mandoc_strdup(const char *ptr)
-{
-       char            *p;
-
-       p = strdup(ptr);
-       if (NULL == p) {
-               perror(NULL);
-               exit(EXIT_FAILURE);
-       }
-
-       return(p);
+       return start;

 }
 
 }
 

-

 static int
 a2time(time_t *t, const char *fmt, const char *p)
 {
 static int
 a2time(time_t *t, const char *fmt, const char *p)
 {


@@ -258,47 +461,170 @@ a2time(time_t *t, const char *fmt, const char *p)
 
        memset(&tm, 0, sizeof(struct tm));
 
 
        memset(&tm, 0, sizeof(struct tm));
 

+       pp = NULL;
+#if HAVE_STRPTIME

        pp = strptime(p, fmt, &tm);
        pp = strptime(p, fmt, &tm);

+#endif

        if (NULL != pp && '\0' == *pp) {
                *t = mktime(&tm);
        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;
+
+       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;
+}

 
 

-/*
- * 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)
+char *
+mandoc_normdate(struct roff_man *man, char *in, int ln, int pos)

 {
 {

+       char            *cp;

        time_t           t;
 
        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, man->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)) {
+               cp = time2a(t);
+               if (t > time(NULL) + 86400)
+                       mandoc_msg(MANDOCERR_DATE_FUTURE, man->parse,
+                           ln, pos, cp);
+               return cp;
+       }

 
 

-       if (MTIME_REDUCED & flags) {
-               if (a2time(&t, "%d, %Y", p))
-                       return(t);
-               if (a2time(&t, "%Y", p))
-                       return(t);
+       /* In man(7), do not warn about the legacy format. */
+
+       if (a2time(&t, "%Y-%m-%d", in) == 0)
+               mandoc_msg(MANDOCERR_DATE_BAD, man->parse, ln, pos, in);
+       else if (t > time(NULL) + 86400)
+               mandoc_msg(MANDOCERR_DATE_FUTURE, man->parse, ln, pos, in);
+       else if (man->macroset == MACROSET_MDOC)
+               mandoc_vmsg(MANDOCERR_DATE_LEGACY, man->parse,
+                   ln, pos, "Dd %s", 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;
+
+       /*
+        * End-of-sentence recognition must include situations where
+        * some symbols, such as `)', allow prior EOS punctuation to
+        * propagate outward.
+        */
+
+       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;
+               default:
+                       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_strntoi(const char *p, size_t sz, int base)
+{
+       char             buf[32];
+       char            *ep;
+       long             v;
+
+       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;
+
+       return (int)v;
+}