version 1.47, 2011/04/17 09:08:19 |
version 1.62, 2011/12/03 16:08:51 |
|
|
/* $Id$ */ |
/* $Id$ */ |
/* |
/* |
* Copyright (c) 2008, 2009, 2010 Kristaps Dzonsons <kristaps@bsd.lv> |
* Copyright (c) 2008, 2009, 2010, 2011 Kristaps Dzonsons <kristaps@bsd.lv> |
* Copyright (c) 2011 Ingo Schwarze <schwarze@openbsd.org> |
* Copyright (c) 2011 Ingo Schwarze <schwarze@openbsd.org> |
* |
* |
* Permission to use, copy, modify, and distribute this software for any |
* Permission to use, copy, modify, and distribute this software for any |
|
|
|
|
#include <assert.h> |
#include <assert.h> |
#include <ctype.h> |
#include <ctype.h> |
|
#include <errno.h> |
|
#include <limits.h> |
#include <stdlib.h> |
#include <stdlib.h> |
#include <stdio.h> |
#include <stdio.h> |
#include <string.h> |
#include <string.h> |
Line 95 numescape(const char *start) |
|
Line 97 numescape(const char *start) |
|
return(++i); |
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 |
enum mandoc_esc |
mandoc_escape(const char **end, const char **start, int *sz) |
mandoc_escape(const char **end, const char **start, int *sz) |
{ |
{ |
Line 132 mandoc_escape(const char **end, const char **start, in |
|
Line 125 mandoc_escape(const char **end, const char **start, in |
|
break; |
break; |
case ('['): |
case ('['): |
gly = ESCAPE_SPECIAL; |
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 = ']'; |
term = ']'; |
break; |
break; |
case ('C'): |
case ('C'): |
Line 160 mandoc_escape(const char **end, const char **start, in |
|
Line 161 mandoc_escape(const char **end, const char **start, in |
|
case ('V'): |
case ('V'): |
/* FALLTHROUGH */ |
/* FALLTHROUGH */ |
case ('Y'): |
case ('Y'): |
if (ESCAPE_ERROR == gly) |
gly = ESCAPE_IGNORE; |
gly = ESCAPE_IGNORE; |
|
/* FALLTHROUGH */ |
/* FALLTHROUGH */ |
case ('*'): |
|
if (ESCAPE_ERROR == gly) |
|
gly = ESCAPE_PREDEF; |
|
/* FALLTHROUGH */ |
|
case ('f'): |
case ('f'): |
if (ESCAPE_ERROR == gly) |
if (ESCAPE_ERROR == gly) |
gly = ESCAPE_FONT; |
gly = ESCAPE_FONT; |
Line 225 mandoc_escape(const char **end, const char **start, in |
|
Line 221 mandoc_escape(const char **end, const char **start, in |
|
case ('L'): |
case ('L'): |
/* FALLTHROUGH */ |
/* FALLTHROUGH */ |
case ('l'): |
case ('l'): |
|
gly = ESCAPE_NUMBERED; |
/* FALLTHROUGH */ |
/* FALLTHROUGH */ |
case ('N'): |
|
if (ESCAPE_ERROR == gly) |
|
gly = ESCAPE_NUMBERED; |
|
/* FALLTHROUGH */ |
|
case ('S'): |
case ('S'): |
/* FALLTHROUGH */ |
/* FALLTHROUGH */ |
case ('v'): |
case ('v'): |
Line 244 mandoc_escape(const char **end, const char **start, in |
|
Line 237 mandoc_escape(const char **end, const char **start, in |
|
term = numeric = '\''; |
term = numeric = '\''; |
break; |
break; |
|
|
|
/* |
|
* Special handling for the numbered character escape. |
|
* XXX Do any other escapes need similar handling? |
|
*/ |
|
case ('N'): |
|
if ('\0' == cp[i]) |
|
return(ESCAPE_ERROR); |
|
*end = &cp[++i]; |
|
if (isdigit((unsigned char)cp[i-1])) |
|
return(ESCAPE_IGNORE); |
|
while (isdigit((unsigned char)**end)) |
|
(*end)++; |
|
if (start) |
|
*start = &cp[i]; |
|
if (sz) |
|
*sz = *end - &cp[i]; |
|
if ('\0' != **end) |
|
(*end)++; |
|
return(ESCAPE_NUMBERED); |
|
|
/* |
/* |
* Sizes get a special category of their own. |
* Sizes get a special category of their own. |
*/ |
*/ |
|
|
|
|
switch (gly) { |
switch (gly) { |
case (ESCAPE_FONT): |
case (ESCAPE_FONT): |
if (1 != rlim) |
/* |
|
* Pretend that the constant-width font modes are the |
|
* same as the regular font modes. |
|
*/ |
|
if (2 == rlim && 'C' == *rstart) |
|
rstart++; |
|
else if (1 != rlim) |
break; |
break; |
|
|
switch (*rstart) { |
switch (*rstart) { |
case ('3'): |
case ('3'): |
/* FALLTHROUGH */ |
/* FALLTHROUGH */ |
Line 435 mandoc_realloc(void *ptr, size_t size) |
|
Line 455 mandoc_realloc(void *ptr, size_t size) |
|
return(ptr); |
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 * |
char * |
mandoc_strdup(const char *ptr) |
mandoc_strdup(const char *ptr) |
{ |
{ |
Line 460 mandoc_strdup(const char *ptr) |
|
Line 490 mandoc_strdup(const char *ptr) |
|
* or to the null byte terminating the argument line. |
* or to the null byte terminating the argument line. |
*/ |
*/ |
char * |
char * |
mandoc_getarg(struct mparse *parse, |
mandoc_getarg(struct mparse *parse, char **cpp, int ln, int *pos) |
char **cpp, int ln, int dowarn, int *pos) |
|
{ |
{ |
char *start, *cp; |
char *start, *cp; |
int quoted, pairs, white; |
int quoted, pairs, white; |
Line 508 mandoc_getarg(struct mparse *parse, |
|
Line 537 mandoc_getarg(struct mparse *parse, |
|
} |
} |
|
|
/* Quoted argument without a closing quote. */ |
/* Quoted argument without a closing quote. */ |
if (dowarn && 1 == quoted) |
if (1 == quoted) |
mandoc_msg(MANDOCERR_BADQUOTE, parse, ln, *pos, NULL); |
mandoc_msg(MANDOCERR_BADQUOTE, parse, ln, *pos, NULL); |
|
|
/* Null-terminate this argument and move to the next one. */ |
/* Null-terminate this argument and move to the next one. */ |
Line 522 mandoc_getarg(struct mparse *parse, |
|
Line 551 mandoc_getarg(struct mparse *parse, |
|
*pos += (int)(cp - start) + (quoted ? 1 : 0); |
*pos += (int)(cp - start) + (quoted ? 1 : 0); |
*cpp = cp; |
*cpp = cp; |
|
|
if (dowarn && '\0' == *cp && (white || ' ' == cp[-1])) |
if ('\0' == *cp && (white || ' ' == cp[-1])) |
mandoc_msg(MANDOCERR_EOLNSPACE, parse, ln, *pos, NULL); |
mandoc_msg(MANDOCERR_EOLNSPACE, parse, ln, *pos, NULL); |
|
|
return(start); |
return(start); |
Line 536 a2time(time_t *t, const char *fmt, const char *p) |
|
Line 565 a2time(time_t *t, const char *fmt, const char *p) |
|
|
|
memset(&tm, 0, sizeof(struct tm)); |
memset(&tm, 0, sizeof(struct tm)); |
|
|
|
pp = NULL; |
|
#ifdef HAVE_STRPTIME |
pp = strptime(p, fmt, &tm); |
pp = strptime(p, fmt, &tm); |
|
#endif |
if (NULL != pp && '\0' == *pp) { |
if (NULL != pp && '\0' == *pp) { |
*t = mktime(&tm); |
*t = mktime(&tm); |
return(1); |
return(1); |
Line 548 a2time(time_t *t, const char *fmt, const char *p) |
|
Line 580 a2time(time_t *t, const char *fmt, const char *p) |
|
static char * |
static char * |
time2a(time_t t) |
time2a(time_t t) |
{ |
{ |
struct tm tm; |
struct tm *tm; |
char *buf, *p; |
char *buf, *p; |
size_t ssz; |
size_t ssz; |
int isz; |
int isz; |
|
|
localtime_r(&t, &tm); |
tm = localtime(&t); |
|
|
/* |
/* |
* Reserve space: |
* Reserve space: |
Line 563 time2a(time_t t) |
|
Line 595 time2a(time_t t) |
|
*/ |
*/ |
p = buf = mandoc_malloc(10 + 4 + 4 + 1); |
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; |
goto fail; |
p += (int)ssz; |
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; |
goto fail; |
p += isz; |
p += isz; |
|
|
if (0 == strftime(p, 4 + 1, "%Y", &tm)) |
if (0 == strftime(p, 4 + 1, "%Y", tm)) |
goto fail; |
goto fail; |
return(buf); |
return(buf); |
|
|
Line 591 mandoc_normdate(struct mparse *parse, char *in, int ln |
|
Line 623 mandoc_normdate(struct mparse *parse, char *in, int ln |
|
mandoc_msg(MANDOCERR_NODATE, parse, ln, pos, NULL); |
mandoc_msg(MANDOCERR_NODATE, parse, ln, pos, NULL); |
time(&t); |
time(&t); |
} |
} |
|
else if (a2time(&t, "%Y-%m-%d", in)) |
|
t = 0; |
else if (!a2time(&t, "$" "Mdocdate: %b %d %Y $", in) && |
else if (!a2time(&t, "$" "Mdocdate: %b %d %Y $", in) && |
!a2time(&t, "%b %d, %Y", in) && |
!a2time(&t, "%b %d, %Y", in)) { |
!a2time(&t, "%Y-%m-%d", in)) { |
|
mandoc_msg(MANDOCERR_BADDATE, parse, ln, pos, NULL); |
mandoc_msg(MANDOCERR_BADDATE, parse, ln, pos, NULL); |
t = 0; |
t = 0; |
} |
} |
Line 613 mandoc_eos(const char *p, size_t sz, int enclosed) |
|
Line 646 mandoc_eos(const char *p, size_t sz, int enclosed) |
|
/* |
/* |
* End-of-sentence recognition must include situations where |
* End-of-sentence recognition must include situations where |
* some symbols, such as `)', allow prior EOS punctuation to |
* some symbols, such as `)', allow prior EOS punctuation to |
* propogate outward. |
* propagate outward. |
*/ |
*/ |
|
|
found = 0; |
found = 0; |
Line 644 mandoc_eos(const char *p, size_t sz, int enclosed) |
|
Line 677 mandoc_eos(const char *p, size_t sz, int enclosed) |
|
return(found && !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 |
* 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 |
* current position and return one; if it isn't, return zero and don't |
Line 695 mandoc_getcontrol(const char *cp, int *ppos) |
|
Line 701 mandoc_getcontrol(const char *cp, int *ppos) |
|
|
|
*ppos = pos; |
*ppos = pos; |
return(1); |
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); |
} |
} |