version 1.89, 2014/12/15 17:30:30 |
version 1.121, 2022/05/19 15:37:47 |
|
|
/* $Id$ */ |
/* $Id$ */ |
/* |
/* |
* Copyright (c) 2008, 2009, 2010, 2011 Kristaps Dzonsons <kristaps@bsd.lv> |
* Copyright (c) 2010, 2011, 2015, 2017, 2018, 2019, 2020, 2021 |
* Copyright (c) 2011, 2012, 2013, 2014 Ingo Schwarze <schwarze@openbsd.org> |
* Ingo Schwarze <schwarze@openbsd.org> |
|
* Copyright (c) 2009, 2010 Kristaps Dzonsons <kristaps@bsd.lv> |
* |
* |
* 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 |
|
|
* WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS, WHETHER IN AN |
* WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS, WHETHER IN AN |
* 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. |
|
* |
|
* 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. |
*/ |
*/ |
#include "config.h" |
#include "config.h" |
|
|
|
|
#include <string.h> |
#include <string.h> |
#include <time.h> |
#include <time.h> |
|
|
#include "mandoc.h" |
|
#include "mandoc_aux.h" |
#include "mandoc_aux.h" |
|
#include "mandoc.h" |
|
#include "roff.h" |
#include "libmandoc.h" |
#include "libmandoc.h" |
|
#include "roff_int.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); |
|
|
|
|
enum mandoc_esc |
enum mandoc_esc |
mandoc_escape(const char **end, const char **start, int *sz) |
mandoc_font(const char *cp, int sz) |
{ |
{ |
const char *local_start; |
switch (sz) { |
int local_sz; |
case 0: |
char term; |
return ESCAPE_FONTPREV; |
enum mandoc_esc gly; |
case 1: |
|
switch (cp[0]) { |
/* |
|
* 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; |
|
term = ']'; |
|
break; |
|
case 'C': |
|
if ('\'' != **start) |
|
return(ESCAPE_ERROR); |
|
*start = ++*end; |
|
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; |
|
default: |
|
*sz = 1; |
|
break; |
|
} |
|
break; |
|
|
|
/* |
|
* 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': |
|
/* FALLTHROUGH */ |
|
case 'b': |
|
/* FALLTHROUGH */ |
|
case 'D': |
|
/* FALLTHROUGH */ |
|
case 'o': |
|
/* FALLTHROUGH */ |
|
case 'R': |
|
/* FALLTHROUGH */ |
|
case 'X': |
|
/* FALLTHROUGH */ |
|
case 'Z': |
|
if ('\0' == **start) |
|
return(ESCAPE_ERROR); |
|
gly = ESCAPE_IGNORE; |
|
term = **start; |
|
*start = ++*end; |
|
break; |
|
|
|
/* |
|
* These escapes are of the form \X'N', where 'X' is the trigger |
|
* and 'N' resolves to a numerical expression. |
|
*/ |
|
case 'h': |
|
/* FALLTHROUGH */ |
|
case 'H': |
|
/* FALLTHROUGH */ |
|
case 'L': |
|
/* FALLTHROUGH */ |
|
case 'l': |
|
/* FALLTHROUGH */ |
|
case 'S': |
|
/* FALLTHROUGH */ |
|
case 'v': |
|
/* FALLTHROUGH */ |
|
case 'x': |
|
if (strchr(" %&()*+-./0123456789:<=>", **start)) { |
|
if ('\0' != **start) |
|
++*end; |
|
return(ESCAPE_ERROR); |
|
} |
|
gly = ESCAPE_IGNORE; |
|
term = **start; |
|
*start = ++*end; |
|
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; |
|
default: |
|
(*end)++; |
|
break; |
|
} |
|
} |
|
*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': |
case 'B': |
gly = ESCAPE_FONTBOLD; |
case '3': |
break; |
return ESCAPE_FONTBOLD; |
case '2': |
|
/* FALLTHROUGH */ |
|
case 'I': |
case 'I': |
gly = ESCAPE_FONTITALIC; |
case '2': |
break; |
return ESCAPE_FONTITALIC; |
case 'P': |
case 'P': |
gly = ESCAPE_FONTPREV; |
return ESCAPE_FONTPREV; |
break; |
|
case '1': |
|
/* FALLTHROUGH */ |
|
case 'R': |
case 'R': |
gly = ESCAPE_FONTROMAN; |
case '1': |
break; |
return ESCAPE_FONTROMAN; |
|
case '4': |
|
return ESCAPE_FONTBI; |
|
default: |
|
return ESCAPE_ERROR; |
} |
} |
break; |
case 2: |
case ESCAPE_SPECIAL: |
switch (cp[0]) { |
if (1 == *sz && 'c' == **start) |
case 'B': |
gly = ESCAPE_NOSPACE; |
|
/* |
|
* Unicode escapes are defined in groff as \[u0000] |
|
* to \[u10FFFF], where the contained value must be |
|
* a valid Unicode codepoint. Here, however, only |
|
* check the length and range. |
|
*/ |
|
if (**start != 'u' || *sz < 5 || *sz > 7) |
|
break; |
|
if (*sz == 7 && ((*start)[1] != '1' || (*start)[2] != '0')) |
|
break; |
|
if (*sz == 6 && (*start)[1] == '0') |
|
break; |
|
if ((int)strspn(*start + 1, "0123456789ABCDEFabcdef") |
|
+ 1 == *sz) |
|
gly = ESCAPE_UNICODE; |
|
break; |
|
default: |
|
break; |
|
} |
|
|
|
return(gly); |
|
} |
|
|
|
/* |
|
* 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. |
|
*/ |
|
switch (cp[1]) { |
switch (cp[1]) { |
case 't': |
case 'I': |
cp[0] = '\t'; |
return ESCAPE_FONTBI; |
/* FALLTHROUGH */ |
|
case '\\': |
|
pairs++; |
|
cp++; |
|
break; |
|
case ' ': |
|
/* Skip escaped blanks. */ |
|
if (0 == quoted) |
|
cp++; |
|
break; |
|
default: |
default: |
break; |
return ESCAPE_ERROR; |
} |
} |
} else if (0 == quoted) { |
case 'C': |
if (' ' == cp[0]) { |
switch (cp[1]) { |
/* Unescaped blanks end unquoted args. */ |
case 'B': |
white = 1; |
return ESCAPE_FONTCB; |
break; |
case 'I': |
|
return ESCAPE_FONTCI; |
|
case 'R': |
|
case 'W': |
|
return ESCAPE_FONTCR; |
|
default: |
|
return ESCAPE_ERROR; |
} |
} |
} else if ('"' == cp[0]) { |
default: |
if ('"' == cp[1]) { |
return ESCAPE_ERROR; |
/* Quoted quotes collapse. */ |
|
pairs++; |
|
cp++; |
|
} else { |
|
/* Unquoted quotes end quoted args. */ |
|
quoted = 2; |
|
break; |
|
} |
|
} |
} |
|
default: |
|
return ESCAPE_ERROR; |
} |
} |
|
|
/* Quoted argument without a closing quote. */ |
|
if (1 == quoted) |
|
mandoc_msg(MANDOCERR_ARG_QUOTE, 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_SPACE_EOL, parse, ln, *pos, NULL); |
|
|
|
return(start); |
|
} |
} |
|
|
static int |
static int |
Line 465 a2time(time_t *t, const char *fmt, const char *p) |
|
Line 111 a2time(time_t *t, const char *fmt, const char *p) |
|
#endif |
#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 * |
Line 479 time2a(time_t t) |
|
Line 125 time2a(time_t t) |
|
size_t ssz; |
size_t ssz; |
int isz; |
int isz; |
|
|
|
buf = NULL; |
tm = localtime(&t); |
tm = localtime(&t); |
if (tm == NULL) |
if (tm == NULL) |
return(NULL); |
goto fail; |
|
|
/* |
/* |
* Reserve space: |
* Reserve space: |
Line 489 time2a(time_t t) |
|
Line 136 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. |
|
*/ |
|
|
|
isz = snprintf(p, 4 + 1, "%d, ", tm->tm_mday); |
|
if (isz < 0 || isz > 4) |
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 mandoc_strdup(""); |
} |
} |
|
|
char * |
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; |
time_t t; |
|
|
if (NULL == in || '\0' == *in || |
/* No date specified. */ |
0 == strcmp(in, "$" "Mdocdate$")) { |
|
mandoc_msg(MANDOCERR_DATE_MISSING, parse, ln, pos, NULL); |
if (nch == NULL) { |
time(&t); |
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)) |
if (*nch->string == '\0') { |
t = 0; |
mandoc_msg(MANDOCERR_DATE_MISSING, nch->line, |
else if (!a2time(&t, "$" "Mdocdate: %b %d %Y $", in) && |
nch->pos, "%s", roff_name[nbl->tok]); |
!a2time(&t, "%b %d, %Y", in)) { |
return mandoc_strdup(""); |
mandoc_msg(MANDOCERR_DATE_BAD, parse, ln, pos, in); |
|
t = 0; |
|
} |
} |
out = t ? time2a(t) : NULL; |
if (strcmp(nch->string, "$" "Mdocdate$") == 0) |
return(out ? out : mandoc_strdup(in)); |
return time2a(time(NULL)); |
|
|
|
/* Valid mdoc(7) date format. */ |
|
|
|
if (a2time(&t, "$" "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 |
int |
Line 537 mandoc_eos(const char *p, size_t sz) |
|
Line 229 mandoc_eos(const char *p, size_t sz) |
|
int enclosed, 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 |
Line 549 mandoc_eos(const char *p, size_t sz) |
|
Line 241 mandoc_eos(const char *p, size_t sz) |
|
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 '\'': |
/* FALLTHROUGH */ |
|
case ']': |
case ']': |
/* FALLTHROUGH */ |
|
case ')': |
case ')': |
if (0 == found) |
if (0 == found) |
enclosed = 1; |
enclosed = 1; |
break; |
break; |
case '.': |
case '.': |
/* FALLTHROUGH */ |
|
case '!': |
case '!': |
/* FALLTHROUGH */ |
|
case '?': |
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; |
} |
} |
|
|
/* |
/* |
Line 585 mandoc_strntoi(const char *p, size_t sz, int base) |
|
Line 273 mandoc_strntoi(const char *p, size_t sz, int base) |
|
long v; |
long v; |
|
|
if (sz > 31) |
if (sz > 31) |
return(-1); |
return -1; |
|
|
memcpy(buf, p, sz); |
memcpy(buf, p, sz); |
buf[(int)sz] = '\0'; |
buf[(int)sz] = '\0'; |
Line 594 mandoc_strntoi(const char *p, size_t sz, int base) |
|
Line 282 mandoc_strntoi(const char *p, size_t sz, int base) |
|
v = strtol(buf, &ep, base); |
v = strtol(buf, &ep, base); |
|
|
if (buf[0] == '\0' || *ep != '\0') |
if (buf[0] == '\0' || *ep != '\0') |
return(-1); |
return -1; |
|
|
if (v > INT_MAX) |
if (v > INT_MAX) |
v = INT_MAX; |
v = INT_MAX; |
if (v < INT_MIN) |
if (v < INT_MIN) |
v = INT_MIN; |
v = INT_MIN; |
|
|
return((int)v); |
return (int)v; |
} |
} |