/*
* Parse a roff node's type from the input buffer. This must be in the
* form of ".foo xxx" in the usual way.
*/
static enum rofft
roff_parse(struct roff *r, const char *buf, int *pos)
{
const char *mac;
size_t maclen;
enum rofft t;
if ('\0' == buf[*pos] || '"' == buf[*pos] ||
'\t' == buf[*pos] || ' ' == buf[*pos])
return(ROFF_MAX);
/*
* We stop the macro parse at an escape, tab, space, or nil.
* However, `\}' is also a valid macro, so make sure we don't
* clobber it by seeing the `\' as the end of token.
*/
mac = buf + *pos;
maclen = strcspn(mac + 1, " \\\t\0") + 1;
t = (r->current_string = roff_getstrn(r, mac, maclen))
? ROFF_USERDEF : roffhash_find(mac, maclen);
*pos += (int)maclen;
while (buf[*pos] && ' ' == buf[*pos])
(*pos)++;
return(t);
}
/*
* Pre-filter each and every line for reserved words (one beginning with
* `\*', e.g., `\*(ab'). These must be handled before the actual line
* is processed.
* This also checks the syntax of regular escapes.
*/
static enum rofferr
roff_res(struct roff *r, char **bufp, size_t *szp, int ln, int pos)
{
enum mandoc_esc esc;
const char *stesc; /* start of an escape sequence ('\\') */
const char *stnam; /* start of the name, after "[(*" */
const char *cp; /* end of the name, e.g. before ']' */
const char *res; /* the string to be substituted */
int i, maxl, expand_count;
size_t nsz;
char *n;
expand_count = 0;
again:
cp = *bufp + pos;
while (NULL != (cp = strchr(cp, '\\'))) {
stesc = cp++;
/*
* The second character must be an asterisk.
* If it isn't, skip it anyway: It is escaped,
* so it can't start another escape sequence.
*/
if ('\0' == *cp)
return(ROFF_CONT);
if ('*' != *cp) {
res = cp;
esc = mandoc_escape(&cp, NULL, NULL);
if (ESCAPE_ERROR != esc)
continue;
cp = res;
mandoc_msg
(MANDOCERR_BADESCAPE, r->parse,
ln, (int)(stesc - *bufp), NULL);
return(ROFF_CONT);
}
cp++;
/*
* The third character decides the length
* of the name of the string.
* Save a pointer to the name.
*/
switch (*cp) {
case ('\0'):
return(ROFF_CONT);
case ('('):
cp++;
maxl = 2;
break;
case ('['):
cp++;
maxl = 0;
break;
default:
maxl = 1;
break;
}
stnam = cp;
/* Advance to the end of the name. */
for (i = 0; 0 == maxl || i < maxl; i++, cp++) {
if ('\0' == *cp) {
mandoc_msg
(MANDOCERR_BADESCAPE,
r->parse, ln,
(int)(stesc - *bufp), NULL);
return(ROFF_CONT);
}
if (0 == maxl && ']' == *cp)
break;
}
/*
* Retrieve the replacement string; if it is
* undefined, resume searching for escapes.
*/
res = roff_getstrn(r, stnam, (size_t)i);
if (NULL == res) {
mandoc_msg
(MANDOCERR_BADESCAPE, r->parse,
ln, (int)(stesc - *bufp), NULL);
res = "";
}
/* Replace the escape sequence by the string. */
pos = stesc - *bufp;
nsz = *szp + strlen(res) + 1;
n = mandoc_malloc(nsz);
strlcpy(n, *bufp, (size_t)(stesc - *bufp + 1));
strlcat(n, res, nsz);
strlcat(n, cp + (maxl ? 0 : 1), nsz);
free(*bufp);
*bufp = n;
*szp = nsz;
if (EXPAND_LIMIT >= ++expand_count)
goto again;
/* Just leave the string unexpanded. */
mandoc_msg(MANDOCERR_ROFFLOOP, r->parse, ln, pos, NULL);
return(ROFF_IGN);
}
return(ROFF_CONT);
}
/*
* In the current line, expand user-defined strings ("\*")
* and references to number registers ("\n").
* Also check the syntax of other escape sequences.
*/
static enum rofferr
roff_res(struct roff *r, char **bufp, size_t *szp, int ln, int pos)
{
char ubuf[12]; /* buffer to print the number */
const char *stesc; /* start of an escape sequence ('\\') */
const char *stnam; /* start of the name, after "[(*" */
const char *cp; /* end of the name, e.g. before ']' */
const char *res; /* the string to be substituted */
char *nbuf; /* new buffer to copy bufp to */
size_t nsz; /* size of the new buffer */
size_t maxl; /* expected length of the escape name */
size_t naml; /* actual length of the escape name */
int expand_count; /* to avoid infinite loops */
expand_count = 0;
again:
cp = *bufp + pos;
while (NULL != (cp = strchr(cp, '\\'))) {
stesc = cp++;
/*
* The second character must be an asterisk or an n.
* If it isn't, skip it anyway: It is escaped,
* so it can't start another escape sequence.
*/
if ('\0' == *cp)
return(ROFF_CONT);
switch (*cp) {
case ('*'):
res = NULL;
break;
case ('n'):
res = ubuf;
break;
default:
if (ESCAPE_ERROR != mandoc_escape(&cp, NULL, NULL))
continue;
mandoc_msg
(MANDOCERR_BADESCAPE, r->parse,
ln, (int)(stesc - *bufp), NULL);
return(ROFF_CONT);
}
cp++;
/*
* The third character decides the length
* of the name of the string or register.
* Save a pointer to the name.
*/
switch (*cp) {
case ('\0'):
return(ROFF_CONT);
case ('('):
cp++;
maxl = 2;
break;
case ('['):
cp++;
maxl = 0;
break;
default:
maxl = 1;
break;
}
stnam = cp;
/* Advance to the end of the name. */
for (naml = 0; 0 == maxl || naml < maxl; naml++, cp++) {
if ('\0' == *cp) {
mandoc_msg
(MANDOCERR_BADESCAPE,
r->parse, ln,
(int)(stesc - *bufp), NULL);
return(ROFF_CONT);
}
if (0 == maxl && ']' == *cp)
break;
}
/*
* Retrieve the replacement string; if it is
* undefined, resume searching for escapes.
*/
if (NULL == res)
res = roff_getstrn(r, stnam, naml);
else
snprintf(ubuf, sizeof(ubuf), "%d",
roff_getregn(r, stnam, naml));
if (NULL == res) {
mandoc_msg
(MANDOCERR_BADESCAPE, r->parse,
ln, (int)(stesc - *bufp), NULL);
res = "";
}
/* Replace the escape sequence by the string. */
pos = stesc - *bufp;
nsz = *szp + strlen(res) + 1;
nbuf = mandoc_malloc(nsz);
strlcpy(nbuf, *bufp, (size_t)(stesc - *bufp + 1));
strlcat(nbuf, res, nsz);
strlcat(nbuf, cp + (maxl ? 0 : 1), nsz);
free(*bufp);
*bufp = nbuf;
*szp = nsz;
if (EXPAND_LIMIT >= ++expand_count)
goto again;
/* Just leave the string unexpanded. */
mandoc_msg(MANDOCERR_ROFFLOOP, r->parse, ln, pos, NULL);
return(ROFF_IGN);
}
return(ROFF_CONT);
}
