=================================================================== RCS file: /cvs/mandoc/mandoc.c,v retrieving revision 1.70 retrieving revision 1.82 diff -u -p -r1.70 -r1.82 --- mandoc/mandoc.c 2013/11/10 21:34:04 1.70 +++ mandoc/mandoc.c 2014/07/06 18:37:34 1.82 @@ -1,7 +1,7 @@ -/* $Id: mandoc.c,v 1.70 2013/11/10 21:34:04 schwarze Exp $ */ +/* $Id: mandoc.c,v 1.82 2014/07/06 18:37:34 schwarze Exp $ */ /* * Copyright (c) 2008, 2009, 2010, 2011 Kristaps Dzonsons - * Copyright (c) 2011, 2012, 2013 Ingo Schwarze + * Copyright (c) 2011, 2012, 2013, 2014 Ingo Schwarze * * Permission to use, copy, modify, and distribute this software for any * purpose with or without fee is hereby granted, provided that the above @@ -31,6 +31,7 @@ #include #include "mandoc.h" +#include "mandoc_aux.h" #include "libmandoc.h" #define DATESIZE 32 @@ -40,12 +41,12 @@ static char *time2a(time_t); enum mandoc_esc -mandoc_escape(const char const **end, const char const **start, int *sz) +mandoc_escape(const char **end, const char **start, int *sz) { const char *local_start; int local_sz; char term; - enum mandoc_esc gly; + enum mandoc_esc gly; /* * When the caller doesn't provide return storage, @@ -74,11 +75,11 @@ mandoc_escape(const char const **end, const char const * these, but each eventually returns a substring of the glyph * name. */ - case ('('): + case '(': gly = ESCAPE_SPECIAL; *sz = 2; break; - case ('['): + case '[': gly = ESCAPE_SPECIAL; /* * Unicode escapes are defined in groff as \[uXXXX] to @@ -90,7 +91,7 @@ mandoc_escape(const char const **end, const char const gly = ESCAPE_UNICODE; term = ']'; break; - case ('C'): + case 'C': if ('\'' != **start) return(ESCAPE_ERROR); *start = ++*end; @@ -102,44 +103,52 @@ mandoc_escape(const char const **end, const char const break; /* + * Escapes taking no arguments at all. + */ + case 'd': + /* FALLTHROUGH */ + case 'u': + return(ESCAPE_IGNORE); + + /* * The \z escape is supposed to output the following - * character without advancing the cursor position. + * character without advancing the cursor position. * Since we are mostly dealing with terminal mode, * let us just skip the next character. */ - case ('z'): + 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': /* FALLTHROUGH */ - case ('g'): + case 'g': /* FALLTHROUGH */ - case ('k'): + case 'k': /* FALLTHROUGH */ - case ('M'): + case 'M': /* FALLTHROUGH */ - case ('m'): + case 'm': /* FALLTHROUGH */ - case ('n'): + case 'n': /* FALLTHROUGH */ - case ('V'): + case 'V': /* FALLTHROUGH */ - case ('Y'): + case 'Y': gly = ESCAPE_IGNORE; /* FALLTHROUGH */ - case ('f'): + case 'f': if (ESCAPE_ERROR == gly) gly = ESCAPE_FONT; switch (**start) { - case ('('): + case '(': *start = ++*end; *sz = 2; break; - case ('['): + case '[': *start = ++*end; term = ']'; break; @@ -152,62 +161,59 @@ mandoc_escape(const char const **end, const char const /* * 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 'A': /* FALLTHROUGH */ - case ('b'): + case 'b': /* FALLTHROUGH */ - case ('D'): + case 'D': /* FALLTHROUGH */ - case ('o'): + case 'o': /* FALLTHROUGH */ - case ('R'): + case 'R': /* FALLTHROUGH */ - case ('X'): + case 'X': /* FALLTHROUGH */ - case ('Z'): - if ('\'' != **start) + case 'Z': + if ('\0' == **start) return(ESCAPE_ERROR); gly = ESCAPE_IGNORE; + term = **start; *start = ++*end; - term = '\''; break; /* * These escapes are of the form \X'N', where 'X' is the trigger * and 'N' resolves to a numerical expression. */ - case ('B'): + case 'h': /* FALLTHROUGH */ - case ('h'): + case 'H': /* FALLTHROUGH */ - case ('H'): + case 'L': /* FALLTHROUGH */ - case ('L'): + case 'l': /* FALLTHROUGH */ - case ('l'): - gly = ESCAPE_NUMBERED; + case 'S': /* FALLTHROUGH */ - case ('S'): + case 'v': /* FALLTHROUGH */ - case ('v'): - /* FALLTHROUGH */ - case ('w'): - /* FALLTHROUGH */ - case ('x'): - if ('\'' != **start) + case 'x': + if (strchr(" %&()*+-./0123456789:<=>", **start)) { + ++*end; return(ESCAPE_ERROR); - if (ESCAPE_ERROR == gly) - gly = ESCAPE_IGNORE; + } + gly = ESCAPE_IGNORE; + term = **start; *start = ++*end; - term = '\''; break; /* * Special handling for the numbered character escape. * XXX Do any other escapes need similar handling? */ - case ('N'): + case 'N': if ('\0' == **start) return(ESCAPE_ERROR); (*end)++; @@ -223,10 +229,10 @@ mandoc_escape(const char const **end, const char const (*end)++; return(ESCAPE_NUMBERED); - /* + /* * Sizes get a special category of their own. */ - case ('s'): + case 's': gly = ESCAPE_IGNORE; /* See +/- counts as a sign. */ @@ -234,15 +240,15 @@ mandoc_escape(const char const **end, const char const (*end)++; switch (**end) { - case ('('): + case '(': *start = ++*end; *sz = 2; break; - case ('['): + case '[': *start = ++*end; term = ']'; break; - case ('\''): + case '\'': *start = ++*end; term = '\''; break; @@ -274,9 +280,9 @@ mandoc_escape(const char const **end, const char const if ('\0' != term) { while (**end != term) { switch (**end) { - case ('\0'): + case '\0': return(ESCAPE_ERROR); - case ('\\'): + case '\\': (*end)++; if (ESCAPE_ERROR == mandoc_escape(end, NULL, NULL)) @@ -298,7 +304,7 @@ mandoc_escape(const char const **end, const char const /* Run post-processors. */ switch (gly) { - case (ESCAPE_FONT): + case ESCAPE_FONT: if (2 == *sz) { if ('C' == **start) { /* @@ -316,27 +322,27 @@ mandoc_escape(const char const **end, const char const break; switch (**start) { - case ('3'): + case '3': /* FALLTHROUGH */ - case ('B'): + case 'B': gly = ESCAPE_FONTBOLD; break; - case ('2'): + case '2': /* FALLTHROUGH */ - case ('I'): + case 'I': gly = ESCAPE_FONTITALIC; break; - case ('P'): + case 'P': gly = ESCAPE_FONTPREV; break; - case ('1'): + case '1': /* FALLTHROUGH */ - case ('R'): + case 'R': gly = ESCAPE_FONTROMAN; break; } break; - case (ESCAPE_SPECIAL): + case ESCAPE_SPECIAL: if (1 == *sz && 'c' == **start) gly = ESCAPE_NOSPACE; break; @@ -347,82 +353,14 @@ mandoc_escape(const char const **end, const char const return(gly); } -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; - - ptr = malloc(size); - if (NULL == ptr) { - perror(NULL); - exit((int)MANDOCLEVEL_SYSERR); - } - - return(ptr); -} - - -void * -mandoc_realloc(void *ptr, size_t size) -{ - - ptr = realloc(ptr, size); - if (NULL == ptr) { - perror(NULL); - exit((int)MANDOCLEVEL_SYSERR); - } - - 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) -{ - char *p; - - p = strdup(ptr); - if (NULL == p) { - perror(NULL); - exit((int)MANDOCLEVEL_SYSERR); - } - - return(p); -} - /* * 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. + * NUL-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. + * or to the NUL byte terminating the argument line. */ char * mandoc_getarg(struct mparse *parse, char **cpp, int ln, int *pos) @@ -436,7 +374,7 @@ mandoc_getarg(struct mparse *parse, char **cpp, int ln if ('"' == *start) { quoted = 1; start++; - } + } pairs = 0; white = 0; @@ -455,14 +393,14 @@ mandoc_getarg(struct mparse *parse, char **cpp, int ln * backslashes and backslash-t to literal tabs. */ switch (cp[1]) { - case ('t'): + case 't': cp[0] = '\t'; /* FALLTHROUGH */ - case ('\\'): + case '\\': pairs++; cp++; break; - case (' '): + case ' ': /* Skip escaped blanks. */ if (0 == quoted) cp++; @@ -493,7 +431,7 @@ mandoc_getarg(struct mparse *parse, char **cpp, int ln if (1 == quoted) mandoc_msg(MANDOCERR_BADQUOTE, parse, ln, *pos, NULL); - /* Null-terminate this argument and move to the next one. */ + /* NUL-terminate this argument and move to the next one. */ if (pairs) cp[-pairs] = '\0'; if ('\0' != *cp) { @@ -573,14 +511,14 @@ mandoc_normdate(struct mparse *parse, char *in, int ln if (NULL == in || '\0' == *in || 0 == strcmp(in, "$" "Mdocdate$")) { - mandoc_msg(MANDOCERR_NODATE, parse, ln, pos, NULL); + mandoc_msg(MANDOCERR_DATE_MISSING, parse, ln, pos, NULL); time(&t); } else if (a2time(&t, "%Y-%m-%d", in)) t = 0; else if (!a2time(&t, "$" "Mdocdate: %b %d %Y $", in) && !a2time(&t, "%b %d, %Y", in)) { - mandoc_msg(MANDOCERR_BADDATE, parse, ln, pos, NULL); + mandoc_msg(MANDOCERR_DATE_BAD, parse, ln, pos, in); t = 0; } out = t ? time2a(t) : NULL; @@ -588,10 +526,10 @@ mandoc_normdate(struct mparse *parse, char *in, int ln } 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); @@ -602,24 +540,24 @@ mandoc_eos(const char *p, size_t sz, int enclosed) * propagate outward. */ - found = 0; + enclosed = found = 0; for (q = p + (int)sz - 1; q >= p; q--) { switch (*q) { - case ('\"'): + case '\"': /* FALLTHROUGH */ - case ('\''): + case '\'': /* FALLTHROUGH */ - case (']'): + case ']': /* FALLTHROUGH */ - case (')'): + case ')': if (0 == found) enclosed = 1; break; - case ('.'): + case '.': /* FALLTHROUGH */ - case ('!'): + case '!': /* FALLTHROUGH */ - case ('?'): + case '?': found = 1; break; default: