=================================================================== RCS file: /cvs/mandoc/mandoc.c,v retrieving revision 1.72 retrieving revision 1.121 diff -u -p -r1.72 -r1.121 --- mandoc/mandoc.c 2013/12/25 22:45:33 1.72 +++ mandoc/mandoc.c 2022/05/19 15:37:47 1.121 @@ -1,7 +1,8 @@ -/* $Id: mandoc.c,v 1.72 2013/12/25 22:45:33 schwarze Exp $ */ +/* $Id: mandoc.c,v 1.121 2022/05/19 15:37:47 schwarze Exp $ */ /* - * Copyright (c) 2008, 2009, 2010, 2011 Kristaps Dzonsons - * Copyright (c) 2011, 2012, 2013 Ingo Schwarze + * Copyright (c) 2010, 2011, 2015, 2017, 2018, 2019, 2020, 2021 + * Ingo Schwarze + * Copyright (c) 2009, 2010 Kristaps Dzonsons * * Permission to use, copy, modify, and distribute this software for any * purpose with or without fee is hereby granted, provided that the above @@ -14,10 +15,13 @@ * 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. + * + * Utility functions to handle end of sentence punctuation + * and dates and times, for use by mdoc(7) and man(7) parsers. + * Utility functions to handle fonts and numbers, + * for use by mandoc(1) parsers and formatters. */ -#ifdef HAVE_CONFIG_H #include "config.h" -#endif #include @@ -30,492 +34,67 @@ #include #include +#include "mandoc_aux.h" #include "mandoc.h" +#include "roff.h" #include "libmandoc.h" +#include "roff_int.h" -#define DATESIZE 32 - static int a2time(time_t *, const char *, const char *); static char *time2a(time_t); enum mandoc_esc -mandoc_escape(const char const **end, const char const **start, int *sz) +mandoc_font(const char *cp, int sz) { - 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; - /* - * 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' == (*start)[0] && ']' != (*start)[1]) - gly = ESCAPE_UNICODE; - term = ']'; - break; - case ('C'): - if ('\'' != **start) - return(ESCAPE_ERROR); - *start = ++*end; - if ('u' == (*start)[0] && '\'' != (*start)[1]) - gly = ESCAPE_UNICODE; - else - gly = ESCAPE_SPECIAL; - term = '\''; - 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. - * 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'): - /* FALLTHROUGH */ - case ('g'): - /* FALLTHROUGH */ - case ('k'): - /* FALLTHROUGH */ - case ('M'): - /* FALLTHROUGH */ - case ('m'): - /* FALLTHROUGH */ - case ('n'): - /* FALLTHROUGH */ - case ('V'): - /* FALLTHROUGH */ - case ('Y'): - gly = ESCAPE_IGNORE; - /* FALLTHROUGH */ - case ('f'): - if (ESCAPE_ERROR == gly) - gly = ESCAPE_FONT; - switch (**start) { - case ('('): - *start = ++*end; - *sz = 2; - break; - case ('['): - *start = ++*end; - term = ']'; - break; + 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: - *sz = 1; - break; + return ESCAPE_ERROR; } - break; - - /* - * These escapes are of the form \X'Y', where 'X' is the trigger - * and 'Y' is any string. These have opaque sub-strings. - */ - case ('A'): - /* FALLTHROUGH */ - case ('b'): - /* FALLTHROUGH */ - case ('D'): - /* FALLTHROUGH */ - case ('o'): - /* FALLTHROUGH */ - case ('R'): - /* FALLTHROUGH */ - case ('X'): - /* FALLTHROUGH */ - case ('Z'): - if ('\'' != **start) - return(ESCAPE_ERROR); - gly = ESCAPE_IGNORE; - *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'): - /* FALLTHROUGH */ - case ('h'): - /* FALLTHROUGH */ - case ('H'): - /* FALLTHROUGH */ - case ('L'): - /* FALLTHROUGH */ - case ('l'): - gly = ESCAPE_NUMBERED; - /* FALLTHROUGH */ - case ('S'): - /* FALLTHROUGH */ - case ('v'): - /* FALLTHROUGH */ - case ('w'): - /* FALLTHROUGH */ - case ('x'): - if ('\'' != **start) - return(ESCAPE_ERROR); - if (ESCAPE_ERROR == gly) - gly = ESCAPE_IGNORE; - *start = ++*end; - term = '\''; - break; - - /* - * 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) - (*end)++; - - switch (**end) { - case ('('): - *start = ++*end; - *sz = 2; - break; - case ('['): - *start = ++*end; - term = ']'; - break; - case ('\''): - *start = ++*end; - term = '\''; - break; - default: - *sz = 1; - break; - } - - break; - - /* - * 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; - } - - 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; + case 2: + switch (cp[0]) { + case 'B': + switch (cp[1]) { + case 'I': + return ESCAPE_FONTBI; default: - (*end)++; - break; + return ESCAPE_ERROR; } - } - *sz = (*end)++ - *start; - } else { - assert(*sz > 0); - if ((size_t)*sz > strlen(*start)) - return(ESCAPE_ERROR); - *end += *sz; - } - - /* 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; - - switch (**start) { - case ('3'): - /* FALLTHROUGH */ - case ('B'): - gly = ESCAPE_FONTBOLD; - break; - case ('2'): - /* FALLTHROUGH */ - case ('I'): - gly = ESCAPE_FONTITALIC; - break; - case ('P'): - gly = ESCAPE_FONTPREV; - break; - case ('1'): - /* FALLTHROUGH */ - case ('R'): - gly = ESCAPE_FONTROMAN; - break; - } - break; - case (ESCAPE_SPECIAL): - if (1 == *sz && 'c' == **start) - gly = ESCAPE_NOSPACE; - break; - default: - break; - } - - 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. - * 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) -{ - char *start, *cp; - int quoted, pairs, white; - - /* Quoting can only start with a new word. */ - start = *cpp; - quoted = 0; - if ('"' == *start) { - quoted = 1; - start++; - } - - 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. - */ + case 'C': switch (cp[1]) { - case ('t'): - cp[0] = '\t'; - /* FALLTHROUGH */ - case ('\\'): - pairs++; - cp++; - break; - case (' '): - /* Skip escaped blanks. */ - if (0 == quoted) - cp++; - break; + case 'B': + return ESCAPE_FONTCB; + case 'I': + return ESCAPE_FONTCI; + case 'R': + case 'W': + return ESCAPE_FONTCR; default: - break; + return ESCAPE_ERROR; } - } 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; - } + default: + return ESCAPE_ERROR; } + default: + return ESCAPE_ERROR; } - - /* Quoted argument without a closing quote. */ - if (1 == quoted) - mandoc_msg(MANDOCERR_BADQUOTE, 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; - - if ('\0' == *cp && (white || ' ' == cp[-1])) - mandoc_msg(MANDOCERR_EOLNSPACE, parse, ln, *pos, NULL); - - return(start); } static int @@ -527,15 +106,15 @@ a2time(time_t *t, const char *fmt, const char *p) memset(&tm, 0, sizeof(struct tm)); pp = NULL; -#ifdef HAVE_STRPTIME +#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 * @@ -546,7 +125,10 @@ time2a(time_t t) size_t ssz; int isz; + buf = NULL; tm = localtime(&t); + if (tm == NULL) + goto fail; /* * Reserve space: @@ -554,55 +136,100 @@ time2a(time_t t) * 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(struct mparse *parse, char *in, 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, "$" "Mdocdate$")) { - mandoc_msg(MANDOCERR_NODATE, parse, 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, "%Y-%m-%d", in)) - t = 0; - else if (!a2time(&t, "$" "Mdocdate: %b %d %Y $", in) && - !a2time(&t, "%b %d, %Y", in)) { - mandoc_msg(MANDOCERR_BADDATE, parse, ln, pos, NULL); - t = 0; + if (*nch->string == '\0') { + mandoc_msg(MANDOCERR_DATE_MISSING, nch->line, + nch->pos, "%s", roff_name[nbl->tok]); + return mandoc_strdup(""); } - out = t ? time2a(t) : NULL; - return(out ? out : mandoc_strdup(in)); + if (strcmp(nch->string, "$" "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 @@ -610,32 +237,28 @@ 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 ('\"'): - /* 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; } /* @@ -650,7 +273,7 @@ mandoc_strntoi(const char *p, size_t sz, int base) long v; if (sz > 31) - return(-1); + return -1; memcpy(buf, p, sz); buf[(int)sz] = '\0'; @@ -659,12 +282,12 @@ mandoc_strntoi(const char *p, size_t sz, int base) v = strtol(buf, &ep, base); if (buf[0] == '\0' || *ep != '\0') - return(-1); + return -1; if (v > INT_MAX) v = INT_MAX; if (v < INT_MIN) v = INT_MIN; - return((int)v); + return (int)v; }