=================================================================== RCS file: /cvs/mandoc/mandoc.c,v retrieving revision 1.7 retrieving revision 1.15 diff -u -p -r1.7 -r1.15 --- mandoc/mandoc.c 2009/11/02 06:22:45 1.7 +++ mandoc/mandoc.c 2010/05/15 07:01:51 1.15 @@ -1,4 +1,4 @@ -/* $Id: mandoc.c,v 1.7 2009/11/02 06:22:45 kristaps Exp $ */ +/* $Id: mandoc.c,v 1.15 2010/05/15 07:01:51 kristaps Exp $ */ /* * Copyright (c) 2008, 2009 Kristaps Dzonsons * @@ -14,8 +14,8 @@ * ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING OUT OF * OR IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE. */ -#if defined(__linux__) || defined(__MINT__) -# define _GNU_SOURCE /* strptime() */ +#ifdef HAVE_CONFIG_H +#include "config.h" #endif #include @@ -35,14 +35,14 @@ static int a2time(time_t *, const char *, const char int mandoc_special(const char *p) { - int c; + 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 ('`'): @@ -73,10 +73,85 @@ mandoc_special(const char *p) return(2); case ('e'): return(2); - case ('f'): - if (0 == *++p || ! isgraph((u_char)*p)) + 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; + } + + if (*p == '+' || *p == '-') { + ++p; + ++c; + } + + 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; + } + + /* TODO: needs to handle floating point. */ + + if ( ! isdigit((u_char)*p)) return(0); - return(3); + + for (i = 0; isdigit((u_char)*p); i++) { + if (lim && i >= lim) + break; + ++p; + ++c; + } + + if (terminator && terminator < 3) { + if (1 == terminator && *p != '\'') + return(0); + if (2 == terminator && *p != ']') + return(0); + ++p; + ++c; + } + + return(c); + case ('f'): + /* FALLTHROUGH */ + case ('F'): + /* FALLTHROUGH */ case ('*'): if (0 == *++p || ! isgraph((u_char)*p)) return(0); @@ -225,3 +300,43 @@ mandoc_a2time(int flags, const char *p) return(0); } + +int +mandoc_eos(const char *p, size_t sz) +{ + + if (0 == sz) + return(0); + + /* + * End-of-sentence recognition must include situations where + * some symbols, such as `)', allow prior EOS punctuation to + * propogate outward. + */ + + for ( ; sz; sz--) { + switch (p[(int)sz - 1]) { + case ('\"'): + /* FALLTHROUGH */ + case ('\''): + /* FALLTHROUGH */ + case (']'): + /* FALLTHROUGH */ + case (')'): + break; + case ('.'): + /* Escaped periods. */ + if (sz > 1 && '\\' == p[(int)sz - 2]) + return(0); + /* FALLTHROUGH */ + case ('!'): + /* FALLTHROUGH */ + case ('?'): + return(1); + default: + return(0); + } + } + + return(0); +}