version 1.42, 2011/03/20 16:02:05 |
version 1.113, 2018/12/18 22:00:02 |
|
|
/* $Id$ */ |
/* $Id$ */ |
/* |
/* |
* Copyright (c) 2008, 2009, 2010 Kristaps Dzonsons <kristaps@bsd.lv> |
* Copyright (c) 2008-2011, 2014 Kristaps Dzonsons <kristaps@bsd.lv> |
* Copyright (c) 2011 Ingo Schwarze <schwarze@openbsd.org> |
* Copyright (c) 2011-2015, 2017, 2018 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 |
* purpose with or without fee is hereby granted, provided that the above |
* purpose with or without fee is hereby granted, provided that the above |
|
|
* ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING OUT OF |
* ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING OUT OF |
* OR IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE. |
* OR IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE. |
*/ |
*/ |
#ifdef HAVE_CONFIG_H |
|
#include "config.h" |
#include "config.h" |
#endif |
|
|
|
#include <sys/types.h> |
#include <sys/types.h> |
|
|
#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> |
#include <time.h> |
#include <time.h> |
|
|
|
#include "mandoc_aux.h" |
#include "mandoc.h" |
#include "mandoc.h" |
|
#include "roff.h" |
#include "libmandoc.h" |
#include "libmandoc.h" |
|
|
#define DATESIZE 32 |
|
|
|
static int a2time(time_t *, const char *, const char *); |
static int a2time(time_t *, const char *, const char *); |
static char *time2a(time_t); |
static char *time2a(time_t); |
|
|
int |
|
mandoc_special(char *p) |
enum mandoc_esc |
|
mandoc_font(const char *cp, int sz) |
{ |
{ |
int len, i; |
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: |
|
return ESCAPE_ERROR; |
|
} |
|
case 2: |
|
switch (cp[0]) { |
|
case 'B': |
|
switch (cp[1]) { |
|
case 'I': |
|
return ESCAPE_FONTBI; |
|
default: |
|
return ESCAPE_ERROR; |
|
} |
|
case 'C': |
|
switch (cp[1]) { |
|
case 'B': |
|
return ESCAPE_FONTBOLD; |
|
case 'I': |
|
return ESCAPE_FONTITALIC; |
|
case 'R': |
|
case 'W': |
|
return ESCAPE_FONTCW; |
|
default: |
|
return ESCAPE_ERROR; |
|
} |
|
default: |
|
return ESCAPE_ERROR; |
|
} |
|
default: |
|
return ESCAPE_ERROR; |
|
} |
|
} |
|
|
|
enum mandoc_esc |
|
mandoc_escape(const char **end, const char **start, int *sz) |
|
{ |
|
const char *local_start; |
|
int local_sz, c, i; |
char term; |
char term; |
char *sv; |
enum mandoc_esc gly; |
|
|
len = 0; |
|
term = '\0'; |
|
sv = p; |
|
|
|
assert('\\' == *p); |
/* |
p++; |
* When the caller doesn't provide return storage, |
|
* use local storage. |
|
*/ |
|
|
switch (*p++) { |
if (NULL == start) |
#if 0 |
start = &local_start; |
case ('Z'): |
if (NULL == sz) |
/* FALLTHROUGH */ |
sz = &local_sz; |
case ('X'): |
|
/* FALLTHROUGH */ |
/* |
case ('x'): |
* Treat "\E" just like "\"; |
/* FALLTHROUGH */ |
* it only makes a difference in copy mode. |
case ('S'): |
*/ |
/* FALLTHROUGH */ |
|
case ('R'): |
if (**end == 'E') |
/* FALLTHROUGH */ |
++*end; |
case ('N'): |
|
/* FALLTHROUGH */ |
/* |
case ('l'): |
* Beyond the backslash, at least one input character |
/* FALLTHROUGH */ |
* is part of the escape sequence. With one exception |
case ('L'): |
* (see below), that character won't be returned. |
/* FALLTHROUGH */ |
*/ |
case ('H'): |
|
/* FALLTHROUGH */ |
gly = ESCAPE_ERROR; |
case ('h'): |
*start = ++*end; |
/* FALLTHROUGH */ |
*sz = 0; |
case ('D'): |
term = '\0'; |
/* FALLTHROUGH */ |
|
case ('C'): |
switch ((*start)[-1]) { |
/* FALLTHROUGH */ |
/* |
case ('b'): |
* First the glyphs. There are several different forms of |
/* FALLTHROUGH */ |
* these, but each eventually returns a substring of the glyph |
case ('B'): |
* name. |
/* FALLTHROUGH */ |
*/ |
case ('a'): |
case '(': |
/* FALLTHROUGH */ |
gly = ESCAPE_SPECIAL; |
case ('A'): |
*sz = 2; |
if (*p++ != '\'') |
break; |
return(0); |
case '[': |
|
if (**start == ' ') { |
|
++*end; |
|
return ESCAPE_ERROR; |
|
} |
|
gly = ESCAPE_SPECIAL; |
|
term = ']'; |
|
break; |
|
case 'C': |
|
if ('\'' != **start) |
|
return ESCAPE_ERROR; |
|
*start = ++*end; |
|
gly = ESCAPE_SPECIAL; |
term = '\''; |
term = '\''; |
break; |
break; |
#endif |
|
case ('h'): |
|
/* FALLTHROUGH */ |
|
case ('v'): |
|
/* FALLTHROUGH */ |
|
case ('s'): |
|
if (ASCII_HYPH == *p) |
|
*p = '-'; |
|
|
|
i = 0; |
/* |
if ('+' == *p || '-' == *p) { |
* Escapes taking no arguments at all. |
p++; |
*/ |
i = 1; |
case '!': |
} |
case '?': |
|
return ESCAPE_UNSUPP; |
|
case '%': |
|
case '&': |
|
case ')': |
|
case ',': |
|
case '/': |
|
case '^': |
|
case 'a': |
|
case 'd': |
|
case 'r': |
|
case 't': |
|
case 'u': |
|
case '{': |
|
case '|': |
|
case '}': |
|
return ESCAPE_IGNORE; |
|
case 'c': |
|
return ESCAPE_NOSPACE; |
|
case 'p': |
|
return ESCAPE_BREAK; |
|
|
switch (*p++) { |
/* |
case ('('): |
* The \z escape is supposed to output the following |
len = 2; |
* 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': |
|
case 'f': |
|
case 'g': |
|
case 'k': |
|
case 'M': |
|
case 'm': |
|
case 'n': |
|
case 'O': |
|
case 'V': |
|
case 'Y': |
|
gly = (*start)[-1] == 'f' ? ESCAPE_FONT : ESCAPE_IGNORE; |
|
switch (**start) { |
|
case '(': |
|
if ((*start)[-1] == 'O') |
|
gly = ESCAPE_ERROR; |
|
*start = ++*end; |
|
*sz = 2; |
break; |
break; |
case ('['): |
case '[': |
|
if ((*start)[-1] == 'O') |
|
gly = (*start)[1] == '5' ? |
|
ESCAPE_UNSUPP : ESCAPE_ERROR; |
|
*start = ++*end; |
term = ']'; |
term = ']'; |
break; |
break; |
case ('\''): |
|
term = '\''; |
|
break; |
|
case ('0'): |
|
i = 1; |
|
/* FALLTHROUGH */ |
|
default: |
default: |
len = 1; |
if ((*start)[-1] == 'O') { |
p--; |
switch (**start) { |
|
case '0': |
|
gly = ESCAPE_UNSUPP; |
|
break; |
|
case '1': |
|
case '2': |
|
case '3': |
|
case '4': |
|
break; |
|
default: |
|
gly = ESCAPE_ERROR; |
|
break; |
|
} |
|
} |
|
*sz = 1; |
break; |
break; |
} |
} |
|
break; |
|
case '*': |
|
if (strncmp(*start, "(.T", 3) != 0) |
|
abort(); |
|
gly = ESCAPE_DEVICE; |
|
*start = ++*end; |
|
*sz = 2; |
|
break; |
|
|
if (ASCII_HYPH == *p) |
/* |
*p = '-'; |
* These escapes are of the form \X'Y', where 'X' is the trigger |
if ('+' == *p || '-' == *p) { |
* and 'Y' is any string. These have opaque sub-strings. |
if (i) |
* The \B and \w escapes are handled in roff.c, roff_res(). |
return(0); |
*/ |
p++; |
case 'A': |
} |
case 'b': |
|
case 'D': |
/* Handle embedded numerical subexp or escape. */ |
case 'R': |
|
case 'X': |
|
case 'Z': |
|
gly = ESCAPE_IGNORE; |
|
/* FALLTHROUGH */ |
|
case 'o': |
|
if (**start == '\0') |
|
return ESCAPE_ERROR; |
|
if (gly == ESCAPE_ERROR) |
|
gly = ESCAPE_OVERSTRIKE; |
|
term = **start; |
|
*start = ++*end; |
|
break; |
|
|
if ('(' == *p) { |
/* |
while (*p && ')' != *p) |
* These escapes are of the form \X'N', where 'X' is the trigger |
if ('\\' == *p++) { |
* and 'N' resolves to a numerical expression. |
i = mandoc_special(--p); |
*/ |
if (0 == i) |
case 'h': |
return(0); |
case 'H': |
p += i; |
case 'L': |
} |
case 'l': |
|
case 'S': |
if (')' == *p++) |
case 'v': |
break; |
case 'x': |
|
if (strchr(" %&()*+-./0123456789:<=>", **start)) { |
return(0); |
if ('\0' != **start) |
} else if ('\\' == *p) { |
++*end; |
if (0 == (i = mandoc_special(p))) |
return ESCAPE_ERROR; |
return(0); |
|
p += i; |
|
} |
} |
|
switch ((*start)[-1]) { |
break; |
case 'h': |
#if 0 |
gly = ESCAPE_HORIZ; |
case ('Y'): |
|
/* FALLTHROUGH */ |
|
case ('V'): |
|
/* FALLTHROUGH */ |
|
case ('$'): |
|
/* FALLTHROUGH */ |
|
case ('n'): |
|
/* FALLTHROUGH */ |
|
#endif |
|
case ('k'): |
|
/* FALLTHROUGH */ |
|
case ('M'): |
|
/* FALLTHROUGH */ |
|
case ('m'): |
|
/* FALLTHROUGH */ |
|
case ('f'): |
|
/* FALLTHROUGH */ |
|
case ('F'): |
|
/* FALLTHROUGH */ |
|
case ('*'): |
|
switch (*p++) { |
|
case ('('): |
|
len = 2; |
|
break; |
break; |
case ('['): |
case 'l': |
term = ']'; |
gly = ESCAPE_HLINE; |
break; |
break; |
default: |
default: |
len = 1; |
gly = ESCAPE_IGNORE; |
p--; |
|
break; |
break; |
} |
} |
|
term = **start; |
|
*start = ++*end; |
break; |
break; |
case ('('): |
|
len = 2; |
/* |
break; |
* Special handling for the numbered character escape. |
case ('['): |
* XXX Do any other escapes need similar handling? |
term = ']'; |
*/ |
break; |
case 'N': |
case ('z'): |
if ('\0' == **start) |
len = 1; |
return ESCAPE_ERROR; |
if ('\\' == *p) { |
(*end)++; |
if (0 == (i = mandoc_special(p))) |
if (isdigit((unsigned char)**start)) { |
return(0); |
*sz = 1; |
p += i; |
return ESCAPE_IGNORE; |
return(*p ? (int)(p - sv) : 0); |
|
} |
} |
break; |
(*start)++; |
case ('o'): |
while (isdigit((unsigned char)**end)) |
/* FALLTHROUGH */ |
(*end)++; |
case ('w'): |
*sz = *end - *start; |
if ('\'' == *p++) { |
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) |
|
*start = ++*end; |
|
|
|
switch (**end) { |
|
case '(': |
|
*start = ++*end; |
|
*sz = 2; |
|
break; |
|
case '[': |
|
*start = ++*end; |
|
term = ']'; |
|
break; |
|
case '\'': |
|
*start = ++*end; |
term = '\''; |
term = '\''; |
break; |
break; |
|
case '3': |
|
case '2': |
|
case '1': |
|
*sz = (*end)[-1] == 's' && |
|
isdigit((unsigned char)(*end)[1]) ? 2 : 1; |
|
break; |
|
default: |
|
*sz = 1; |
|
break; |
} |
} |
|
|
|
break; |
|
|
|
/* |
|
* Several special characters can be encoded as |
|
* one-byte escape sequences without using \[]. |
|
*/ |
|
case ' ': |
|
case '\'': |
|
case '-': |
|
case '.': |
|
case '0': |
|
case ':': |
|
case '_': |
|
case '`': |
|
case 'e': |
|
case '~': |
|
gly = ESCAPE_SPECIAL; |
/* FALLTHROUGH */ |
/* FALLTHROUGH */ |
default: |
default: |
len = 1; |
if (gly == ESCAPE_ERROR) |
p--; |
gly = ESCAPE_UNDEF; |
|
*start = --*end; |
|
*sz = 1; |
break; |
break; |
} |
} |
|
|
if (term) { |
/* |
for ( ; *p && term != *p; p++) |
* Read up to the terminating character, |
if (ASCII_HYPH == *p) |
* paying attention to nested escapes. |
*p = '-'; |
*/ |
return(*p ? (int)(p - sv) : 0); |
|
} |
|
|
|
for (i = 0; *p && i < len; i++, p++) |
if ('\0' != term) { |
if (ASCII_HYPH == *p) |
while (**end != term) { |
*p = '-'; |
switch (**end) { |
return(i == len ? (int)(p - sv) : 0); |
case '\0': |
} |
return ESCAPE_ERROR; |
|
case '\\': |
|
(*end)++; |
|
if (ESCAPE_ERROR == |
|
mandoc_escape(end, NULL, NULL)) |
|
return ESCAPE_ERROR; |
|
break; |
|
default: |
|
(*end)++; |
|
break; |
|
} |
|
} |
|
*sz = (*end)++ - *start; |
|
|
|
/* |
void * |
* The file chars.c only provides one common list |
mandoc_calloc(size_t num, size_t size) |
* of character names, but \[-] == \- is the only |
{ |
* one of the characters with one-byte names that |
void *ptr; |
* allows enclosing the name in brackets. |
|
*/ |
ptr = calloc(num, size); |
if (gly == ESCAPE_SPECIAL && *sz == 1 && **start != '-') |
if (NULL == ptr) { |
return ESCAPE_ERROR; |
perror(NULL); |
} else { |
exit((int)MANDOCLEVEL_SYSERR); |
assert(*sz > 0); |
|
if ((size_t)*sz > strlen(*start)) |
|
return ESCAPE_ERROR; |
|
*end += *sz; |
} |
} |
|
|
return(ptr); |
/* Run post-processors. */ |
} |
|
|
|
|
switch (gly) { |
void * |
case ESCAPE_FONT: |
mandoc_malloc(size_t size) |
gly = mandoc_font(*start, *sz); |
{ |
break; |
void *ptr; |
case ESCAPE_SPECIAL: |
|
if (**start == 'c') { |
ptr = malloc(size); |
if (*sz < 6 || *sz > 7 || |
if (NULL == ptr) { |
strncmp(*start, "char", 4) != 0 || |
perror(NULL); |
(int)strspn(*start + 4, "0123456789") + 4 < *sz) |
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_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. |
|
* Collapse pairs of quotes inside quoted arguments. |
|
* Advance the argument pointer to the next argument, |
|
* or to the null 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; |
|
if ('"' == *start) { |
|
quoted = 1; |
|
start++; |
|
} else |
|
quoted = 0; |
|
|
|
pairs = 0; |
|
white = 0; |
|
for (cp = start; '\0' != *cp; cp++) { |
|
/* Move left after quoted quotes and escaped backslashes. */ |
|
if (pairs) |
|
cp[-pairs] = cp[0]; |
|
if ('\\' == cp[0]) { |
|
if ('\\' == cp[1]) { |
|
/* Poor man's copy mode. */ |
|
pairs++; |
|
cp++; |
|
} else if (0 == quoted && ' ' == cp[1]) |
|
/* Skip escaped blanks. */ |
|
cp++; |
|
} else if (0 == quoted) { |
|
if (' ' == cp[0]) { |
|
/* Unescaped blanks end unquoted args. */ |
|
white = 1; |
|
break; |
break; |
} |
c = 0; |
} else if ('"' == cp[0]) { |
for (i = 4; i < *sz; i++) |
if ('"' == cp[1]) { |
c = 10 * c + ((*start)[i] - '0'); |
/* Quoted quotes collapse. */ |
if (c < 0x21 || (c > 0x7e && c < 0xa0) || c > 0xff) |
pairs++; |
|
cp++; |
|
} else { |
|
/* Unquoted quotes end quoted args. */ |
|
quoted = 2; |
|
break; |
break; |
} |
*start += 4; |
|
*sz -= 4; |
|
gly = ESCAPE_NUMBERED; |
|
break; |
} |
} |
} |
|
|
|
/* Quoted argument without a closing quote. */ |
/* |
if (1 == quoted) |
* Unicode escapes are defined in groff as \[u0000] |
mandoc_msg(MANDOCERR_BADQUOTE, parse, ln, *pos, NULL); |
* to \[u10FFFF], where the contained value must be |
|
* a valid Unicode codepoint. Here, however, only |
/* Null-terminate this argument and move to the next one. */ |
* check the length and range. |
if (pairs) |
*/ |
cp[-pairs] = '\0'; |
if (**start != 'u' || *sz < 5 || *sz > 7) |
if ('\0' != *cp) { |
break; |
*cp++ = '\0'; |
if (*sz == 7 && ((*start)[1] != '1' || (*start)[2] != '0')) |
while (' ' == *cp) |
break; |
cp++; |
if (*sz == 6 && (*start)[1] == '0') |
|
break; |
|
if (*sz == 5 && (*start)[1] == 'D' && |
|
strchr("89ABCDEF", (*start)[2]) != NULL) |
|
break; |
|
if ((int)strspn(*start + 1, "0123456789ABCDEFabcdef") |
|
+ 1 == *sz) |
|
gly = ESCAPE_UNICODE; |
|
break; |
|
default: |
|
break; |
} |
} |
*pos += (int)(cp - start) + (quoted ? 1 : 0); |
|
*cpp = cp; |
|
|
|
if ('\0' == *cp && (white || ' ' == cp[-1])) |
return gly; |
mandoc_msg(MANDOCERR_EOLNSPACE, parse, ln, *pos, NULL); |
|
|
|
return(start); |
|
} |
} |
|
|
static int |
static int |
Line 371 a2time(time_t *t, const char *fmt, const char *p) |
|
Line 473 a2time(time_t *t, const char *fmt, const char *p) |
|
|
|
memset(&tm, 0, sizeof(struct tm)); |
memset(&tm, 0, sizeof(struct tm)); |
|
|
|
pp = NULL; |
|
#if 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; |
} |
} |
|
|
return(0); |
return 0; |
} |
} |
|
|
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); |
|
if (tm == NULL) |
|
return NULL; |
|
|
/* |
/* |
* Reserve space: |
* Reserve space: |
Line 396 time2a(time_t t) |
|
Line 503 time2a(time_t t) |
|
* up to 2 characters for the day + comma + blank |
* up to 2 characters for the day + comma + blank |
* 4 characters for the year and a terminating '\0' |
* 4 characters for the year and a terminating '\0' |
*/ |
*/ |
|
|
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 ((ssz = strftime(p, 10 + 1, "%B ", tm)) == 0) |
goto fail; |
goto fail; |
p += (int)ssz; |
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. |
|
*/ |
|
|
|
if ((isz = snprintf(p, 4 + 1, "%d, ", tm->tm_mday)) == -1) |
goto fail; |
goto fail; |
p += isz; |
p += isz; |
|
|
if (0 == strftime(p, 4 + 1, "%Y", &tm)) |
if (strftime(p, 4 + 1, "%Y", tm) == 0) |
goto fail; |
goto fail; |
return(buf); |
return buf; |
|
|
fail: |
fail: |
free(buf); |
free(buf); |
return(NULL); |
return NULL; |
} |
} |
|
|
char * |
char * |
mandoc_normdate(struct mparse *parse, char *in, int ln, int pos) |
mandoc_normdate(struct roff_man *man, char *in, int ln, int pos) |
{ |
{ |
char *out; |
char *cp; |
time_t t; |
time_t t; |
|
|
if (NULL == in || '\0' == *in || |
/* No date specified: use today's date. */ |
0 == strcmp(in, "$" "Mdocdate$")) { |
|
mandoc_msg(MANDOCERR_NODATE, parse, ln, pos, NULL); |
if (in == NULL || *in == '\0' || strcmp(in, "$" "Mdocdate$") == 0) { |
time(&t); |
mandoc_msg(MANDOCERR_DATE_MISSING, ln, pos, NULL); |
|
return time2a(time(NULL)); |
} |
} |
else if (!a2time(&t, "$" "Mdocdate: %b %d %Y $", in) && |
|
!a2time(&t, "%b %d, %Y", in) && |
/* Valid mdoc(7) date format. */ |
!a2time(&t, "%Y-%m-%d", in)) { |
|
mandoc_msg(MANDOCERR_BADDATE, parse, ln, pos, NULL); |
if (a2time(&t, "$" "Mdocdate: %b %d %Y $", in) || |
t = 0; |
a2time(&t, "%b %d, %Y", in)) { |
|
cp = time2a(t); |
|
if (t > time(NULL) + 86400) |
|
mandoc_msg(MANDOCERR_DATE_FUTURE, ln, pos, "%s", cp); |
|
else if (*in != '$' && strcmp(in, cp) != 0) |
|
mandoc_msg(MANDOCERR_DATE_NORM, ln, pos, "%s", cp); |
|
return cp; |
} |
} |
out = t ? time2a(t) : NULL; |
|
return(out ? out : mandoc_strdup(in)); |
/* In man(7), do not warn about the legacy format. */ |
|
|
|
if (a2time(&t, "%Y-%m-%d", in) == 0) |
|
mandoc_msg(MANDOCERR_DATE_BAD, ln, pos, "%s", in); |
|
else if (t > time(NULL) + 86400) |
|
mandoc_msg(MANDOCERR_DATE_FUTURE, ln, pos, "%s", in); |
|
else if (man->macroset == MACROSET_MDOC) |
|
mandoc_msg(MANDOCERR_DATE_LEGACY, ln, pos, "Dd %s", in); |
|
|
|
/* Use any non-mdoc(7) date verbatim. */ |
|
|
|
return mandoc_strdup(in); |
} |
} |
|
|
int |
int |
mandoc_eos(const char *p, size_t sz, int enclosed) |
mandoc_eos(const char *p, size_t sz) |
{ |
{ |
const char *q; |
const char *q; |
int found; |
int enclosed, found; |
|
|
if (0 == sz) |
if (0 == sz) |
return(0); |
return 0; |
|
|
/* |
/* |
* 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; |
enclosed = found = 0; |
for (q = p + (int)sz - 1; q >= p; q--) { |
for (q = p + (int)sz - 1; q >= p; q--) { |
switch (*q) { |
switch (*q) { |
case ('\"'): |
case '\"': |
/* FALLTHROUGH */ |
case '\'': |
case ('\''): |
case ']': |
/* FALLTHROUGH */ |
case ')': |
case (']'): |
|
/* FALLTHROUGH */ |
|
case (')'): |
|
if (0 == found) |
if (0 == found) |
enclosed = 1; |
enclosed = 1; |
break; |
break; |
case ('.'): |
case '.': |
/* FALLTHROUGH */ |
case '!': |
case ('!'): |
case '?': |
/* FALLTHROUGH */ |
|
case ('?'): |
|
found = 1; |
found = 1; |
break; |
break; |
default: |
default: |
return(found && (!enclosed || isalnum((unsigned char)*q))); |
return found && |
|
(!enclosed || isalnum((unsigned char)*q)); |
} |
} |
} |
} |
|
|
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); |
|
} |
|
|
|
/* |
/* |
* Check if a string is a punctuation delimiter. This only applies to |
* Convert a string to a long that may not be <0. |
* mdoc(7) documents, but as it's used in both front-ends and back-ends, |
* If the string is invalid, or is less than 0, return -1. |
* it needs to go here (instead of, say, in libmdoc.h). |
|
*/ |
*/ |
enum mdelim |
int |
mandoc_isdelim(const char *p) |
mandoc_strntoi(const char *p, size_t sz, int base) |
{ |
{ |
|
char buf[32]; |
|
char *ep; |
|
long v; |
|
|
if ('\0' == p[0]) |
if (sz > 31) |
return(DELIM_NONE); |
return -1; |
|
|
if ('\0' == p[1]) |
memcpy(buf, p, sz); |
switch (p[0]) { |
buf[(int)sz] = '\0'; |
case('('): |
|
/* FALLTHROUGH */ |
|
case('['): |
|
return(DELIM_OPEN); |
|
case('|'): |
|
return(DELIM_MIDDLE); |
|
case('.'): |
|
/* FALLTHROUGH */ |
|
case(','): |
|
/* FALLTHROUGH */ |
|
case(';'): |
|
/* FALLTHROUGH */ |
|
case(':'): |
|
/* FALLTHROUGH */ |
|
case('?'): |
|
/* FALLTHROUGH */ |
|
case('!'): |
|
/* FALLTHROUGH */ |
|
case(')'): |
|
/* FALLTHROUGH */ |
|
case(']'): |
|
return(DELIM_CLOSE); |
|
default: |
|
return(DELIM_NONE); |
|
} |
|
|
|
if ('\\' != p[0]) |
errno = 0; |
return(DELIM_NONE); |
v = strtol(buf, &ep, base); |
|
|
if (0 == strcmp(p + 1, ".")) |
if (buf[0] == '\0' || *ep != '\0') |
return(DELIM_CLOSE); |
return -1; |
if (0 == strcmp(p + 1, "*(Ba")) |
|
return(DELIM_MIDDLE); |
|
|
|
return(DELIM_NONE); |
if (v > INT_MAX) |
|
v = INT_MAX; |
|
if (v < INT_MIN) |
|
v = INT_MIN; |
|
|
|
return (int)v; |
} |
} |