static const char *
next_s(const char *s)
{
/* find next position in s, ignoring braces and ignoring TeX control
sequences and any space that follows them */
for (++s; (*s != '\0'); )
{
switch (*s)
{
case '\\': /* TeX control sequence */
++s; /* look at next character */
if (Isalpha((int)*s)) /* \<one-or-more-letters> */
{
while (Isalpha((int)*s))
++s;
}
else /* \<non-letter> */
++s;
while (Isspace((int)*s)) /* advance over trailing whitespace */
++s; /* since TeX does too */
break;
case '{':
case '}':
++s;
break;
case BIBTEX_HIDDEN_DELIMITER: /* ignore delimited inline comment */
for (++s; (*s != '\0'); ++s)
{
if (*s == BIBTEX_HIDDEN_DELIMITER)
{
++s;
break;
}
}
break;
default:
return (s);
} /* end switch */
} /* end for */
return (s);
}
YESorNO
match_pattern(const char *s, const char *pattern)
{
const char *org_s;
const char *org_pattern;
org_s = s;
org_pattern = pattern;
s = next_s(s-1);
for ( ; (*pattern != '\0'); ++pattern)
{
switch(*pattern)
{
case 'a': /* single letter */
if (!Isalpha((int)*s))
RETURN_MATCH_FAILURE("single letter");
s = next_s(s);
break;
case 'A': /* one or more letters */
if (!Isalpha((int)*s))
RETURN_MATCH_FAILURE("one or more letters");
while (Isalpha((int)*s))
s = next_s(s);
break;
case 'd':
if (!Isdigit((int)*s)) /* single digit */
RETURN_MATCH_FAILURE("single digit");
s = next_s(s);
break;
case 'D': /* one or more digits */
if (!Isdigit((int)*s))
RETURN_MATCH_FAILURE("one or more digits");
while (Isdigit((int)*s))
s = next_s(s);
break;
case 'r': /* single roman numeral */
if (!is_roman((int)*s))
RETURN_MATCH_FAILURE("single roman numeral");
s = next_s(s);
break;
case 'R': /* one or more roman numerals */
if (!is_roman((int)*s))
RETURN_MATCH_FAILURE("one or more roman numerals");
while (is_roman((int)*s))
s = next_s(s);
break;
case 'w': /* one word (letters and digits) */
if (!Isalnum((int)*s))
RETURN_MATCH_FAILURE("one word (letters and digits)");
while (Isalnum((int)*s))
s = next_s(s);
break;
case 'W': /* one or more space-separated words */
if (!Isalnum((int)*s))
RETURN_MATCH_FAILURE("one or more space-separated words");
while (Isalnum((int)*s)) /* parse first word */
s = next_s(s);
for (;;)
{
if (!Isspace((int)*s))
break;
while (Isspace((int)*s)) /* parse separators */
s = next_s(s);
while (Isalnum((int)*s)) /* parse another word */
s = next_s(s);
}
break;
case 'X': /* one or more special-separated words */
if (!Isalnum((int)*s))
RETURN_MATCH_FAILURE("one or more special-separated words");
while (Isalnum((int)*s)) /* parse first word */
s = next_s(s);
for (;;)
{
if (!is_special(*s))
break;
while (is_special(*s)) /* parse separators */
s = next_s(s);
while (Isalnum((int)*s)) /* parse another word */
s = next_s(s);
}
break;
case ' ': /* one or more whitespace characters */
if (!Isspace((int)*s))
RETURN_MATCH_FAILURE("one or more whitespace characters");
while (Isspace((int)*s))
s = next_s(s);
break;
case '.': /* exactly one special character */
if (!is_special(*s))
RETURN_MATCH_FAILURE("exactly one special character");
s = next_s(s); /* [07-Mar-1999] bug fix: missing before bibclean 2.12 */
break;
case ':': /* one or more special characters */
if (!is_special(*s))
RETURN_MATCH_FAILURE("one or more special characters");
while (is_special(*s))
s = next_s(s);
break;
case '\\': /* literal next character */
pattern++;
/* fall through to exact match test */
/*@fallthrough@*/ /*FALLTHROUGH*/
default: /* anything else: exact match */
if (*pattern != *s)
RETURN_MATCH_FAILURE("anything else: exact match");
s = next_s(s);
} /* end switch */
} /* end for (; ;) */
if (*s == '\0')
return (YES);
else
RETURN_MATCH_FAILURE("end of string");
}
eChar
parseescape(const Char **ptr)
{
const Char *p;
Char c;
p = *ptr;
if ((p[1] & CHAR) == 0) {
xprintf(CGETS(9, 8, "Something must follow: %c\n"), (char)*p);
return CHAR_ERR;
}
if ((*p & CHAR) == '\\') {
p++;
switch (*p & CHAR) {
case 'a':
c = CTL_ESC('\007'); /* Bell */
break;
case 'b':
c = CTL_ESC('\010'); /* Backspace */
break;
case 'e':
c = CTL_ESC('\033'); /* Escape */
break;
case 'f':
c = CTL_ESC('\014'); /* Form Feed */
break;
case 'n':
c = CTL_ESC('\012'); /* New Line */
break;
case 'r':
c = CTL_ESC('\015'); /* Carriage Return */
break;
case 't':
c = CTL_ESC('\011'); /* Horizontal Tab */
break;
case 'v':
c = CTL_ESC('\013'); /* Vertical Tab */
break;
case '\\':
c = '\\';
break;
case '0':
case '1':
case '2':
case '3':
case '4':
case '5':
case '6':
case '7':
{
int cnt, val;
Char ch;
for (cnt = 0, val = 0; cnt < 3; cnt++) {
ch = *p++ & CHAR;
if (ch < '0' || ch > '7') {
p--;
break;
}
val = (val << 3) | (ch - '0');
}
if ((val & ~0xff) != 0) {
xprintf("%s", CGETS(9, 9,
"Octal constant does not fit in a char.\n"));
return 0;
}
#ifndef IS_ASCII
if (CTL_ESC(val) != val && adrof(STRwarnebcdic))
xprintf(/*CGETS(9, 9, no NLS-String yet!*/
"Warning: Octal constant \\%3.3o is interpreted as EBCDIC value.\n", val/*)*/);
#endif
c = (Char) val;
--p;
}
break;
default:
c = *p;
break;
}
}
else if ((*p & CHAR) == '^' && (Isalpha(p[1] & CHAR) ||
strchr("@^_?\\|[{]}", p[1] & CHAR))) {
p++;
#ifdef IS_ASCII
c = ((*p & CHAR) == '?') ? CTL_ESC('\177') : ((*p & CHAR) & 0237);
#else
c = ((*p & CHAR) == '?') ? CTL_ESC('\177') : _toebcdic[_toascii[*p & CHAR] & 0237];
if (adrof(STRwarnebcdic))
xprintf(/*CGETS(9, 9, no NLS-String yet!*/
"Warning: Control character ^%c may be interpreted differently in EBCDIC.\n", *p & CHAR /*)*/);
#endif
}
else
c = *p;
*ptr = p;
return (c);
}