[379] in Kerberos-V5-bugs
parse/unparse inconsistency
daemon@ATHENA.MIT.EDU (Jim Miller)
Mon Oct 18 20:25:02 1993
From: jim@bilbo.suite.com (Jim Miller)
Date: Mon, 18 Oct 93 19:11:13 -0500
To: krb5-bugs@MIT.EDU
Reply-To: Jim_Miller@suite.com
In Kerberos 5, pre-beta 3...
The functions "krb5_parse_name" and "krb5_unparse_name" do not handle quoted
characters the same.
In krb5_parse_name:
/*
* Pass 3. Now we go through the string a *third* time, this
* time filling in the krb5_principal structure which we just
* allocated.
*/
q = krb5_princ_component(principal, 0)->data;
for (i=0,cp = name; c = *cp; cp++) {
if (c == QUOTECHAR) {
cp++;
switch (c = *cp) {
case 'n':
*q++ = '\n';
break;
case 't':
*q++ = '\t';
break;
case 'b':
*q++ = '\b';
break;
case '0':
*q++ = '\0';
break;
default:
*q++ = c;
}
} else if ((c == COMPONENT_SEP) || (c == REALM_SEP)) {
i++;
*q++ = '\0';
if (c == COMPONENT_SEP)
q = krb5_princ_component(principal, i)->data;
else
q = krb5_princ_realm(principal)->data;
} else
*q++ = c;
}
In krb5_unparse_name:
for (i = 0; i < nelem; i++) {
cp = krb5_princ_component(principal, i)->data;
length = krb5_princ_component(principal, i)->length;
for (j=0; j < length; j++,cp++) {
switch (*cp) {
case COMPONENT_SEP:
case REALM_SEP:
case '\t':
case '\\':
*q++ = '\\';
*q++ = *cp;
break;
case '\0':
*q++ = '\\';
*q++ = '0';
break;
default:
*q++ = *cp;
}
}
*q++ = COMPONENT_SEP;
}
Krb5_unparse_name does not handle '\n' or '\b'. Or perhaps, krb5_parse_name
handles them unnecessarily.
Jim_Miller@suite.com
