=================================================================== RCS file: /cvs/mandoc/mandoc.c,v retrieving revision 1.45 retrieving revision 1.59 diff -u -p -r1.45 -r1.59 --- mandoc/mandoc.c 2011/04/09 15:29:40 1.45 +++ mandoc/mandoc.c 2011/09/18 14:14:15 1.59 @@ -1,6 +1,6 @@ -/* $Id: mandoc.c,v 1.45 2011/04/09 15:29:40 kristaps Exp $ */ +/* $Id: mandoc.c,v 1.59 2011/09/18 14:14:15 schwarze Exp $ */ /* - * Copyright (c) 2008, 2009, 2010 Kristaps Dzonsons + * Copyright (c) 2008, 2009, 2010, 2011 Kristaps Dzonsons * Copyright (c) 2011 Ingo Schwarze * * Permission to use, copy, modify, and distribute this software for any @@ -23,6 +23,8 @@ #include #include +#include +#include #include #include #include @@ -95,15 +97,6 @@ numescape(const char *start) return(++i); } -/* - * Handle an escaped sequeence. This should be called with any - * string subsequent a `\'. Pass a pointer to this substring as "end"; - * it will be set to the supremum of the parsed escape sequence. If - * this returns ESCAPE_ERROR, the string is bogus and should be thrown - * away. If not ESCAPE_ERROR or ESCAPE_IGNORE, "start" is set to the - * first relevant character of the substring (font, glyph, whatever) of - * length sz. Both "start" and "sz" may be NULL. - */ enum mandoc_esc mandoc_escape(const char **end, const char **start, int *sz) { @@ -116,10 +109,9 @@ mandoc_escape(const char **end, const char **start, in rstart = cp; if (start) *start = rstart; - i = 0; + i = lim = 0; gly = ESCAPE_ERROR; - term = '\0'; - numeric = 0; + term = numeric = '\0'; switch ((c = cp[i++])) { /* @@ -133,6 +125,14 @@ mandoc_escape(const char **end, const char **start, in break; case ('['): gly = ESCAPE_SPECIAL; + /* + * Unicode escapes are defined in groff as \[uXXXX] to + * \[u10FFFF], where the contained value must be a valid + * Unicode codepoint. Here, however, only check whether + * it's not a zero-width escape. + */ + if ('u' == cp[i] && ']' != cp[i + 1]) + gly = ESCAPE_UNICODE; term = ']'; break; case ('C'): @@ -164,10 +164,6 @@ mandoc_escape(const char **end, const char **start, in if (ESCAPE_ERROR == gly) gly = ESCAPE_IGNORE; /* FALLTHROUGH */ - case ('*'): - if (ESCAPE_ERROR == gly) - gly = ESCAPE_PREDEF; - /* FALLTHROUGH */ case ('f'): if (ESCAPE_ERROR == gly) gly = ESCAPE_FONT; @@ -379,6 +375,7 @@ out: gly = ESCAPE_FONTROMAN; break; } + break; case (ESCAPE_SPECIAL): if (1 != rlim) break; @@ -435,7 +432,17 @@ mandoc_realloc(void *ptr, size_t size) return(ptr); } +char * +mandoc_strndup(const char *ptr, size_t sz) +{ + char *p; + p = mandoc_malloc(sz + 1); + memcpy(p, ptr, sz); + p[(int)sz] = '\0'; + return(p); +} + char * mandoc_strdup(const char *ptr) { @@ -467,11 +474,11 @@ mandoc_getarg(struct mparse *parse, char **cpp, int ln /* Quoting can only start with a new word. */ start = *cpp; + quoted = 0; if ('"' == *start) { quoted = 1; start++; - } else - quoted = 0; + } pairs = 0; white = 0; @@ -535,7 +542,10 @@ a2time(time_t *t, const char *fmt, const char *p) memset(&tm, 0, sizeof(struct tm)); + pp = NULL; +#ifdef HAVE_STRPTIME pp = strptime(p, fmt, &tm); +#endif if (NULL != pp && '\0' == *pp) { *t = mktime(&tm); return(1); @@ -547,12 +557,12 @@ a2time(time_t *t, const char *fmt, const char *p) static char * time2a(time_t t) { - struct tm tm; + struct tm *tm; char *buf, *p; size_t ssz; int isz; - localtime_r(&t, &tm); + tm = localtime(&t); /* * Reserve space: @@ -562,15 +572,15 @@ time2a(time_t t) */ p = buf = mandoc_malloc(10 + 4 + 4 + 1); - if (0 == (ssz = strftime(p, 10 + 1, "%B ", &tm))) + if (0 == (ssz = strftime(p, 10 + 1, "%B ", tm))) goto fail; p += (int)ssz; - if (-1 == (isz = snprintf(p, 4 + 1, "%d, ", tm.tm_mday))) + if (-1 == (isz = snprintf(p, 4 + 1, "%d, ", tm->tm_mday))) goto fail; p += isz; - if (0 == strftime(p, 4 + 1, "%Y", &tm)) + if (0 == strftime(p, 4 + 1, "%Y", tm)) goto fail; return(buf); @@ -612,7 +622,7 @@ mandoc_eos(const char *p, size_t sz, int enclosed) /* * End-of-sentence recognition must include situations where * some symbols, such as `)', allow prior EOS punctuation to - * propogate outward. + * propagate outward. */ found = 0; @@ -643,33 +653,6 @@ mandoc_eos(const char *p, size_t sz, int enclosed) return(found && !enclosed); } -int -mandoc_hyph(const char *start, const char *c) -{ - - /* - * Choose whether to break at a hyphenated character. We only - * do this if it's free-standing within a word. - */ - - /* 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); -} - /* * Find out whether a line is a macro line or not. If it is, adjust the * current position and return one; if it isn't, return zero and don't @@ -694,4 +677,35 @@ mandoc_getcontrol(const char *cp, int *ppos) *ppos = pos; return(1); +} + +/* + * 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); }