[59] in Kerberos-V5-bugs
Bug in pepy or posy, ISODE 6.0 and 7.0-beta
daemon@ATHENA.MIT.EDU (John T Kohl)
Wed Nov 7 11:15:36 1990
Date: Wed, 7 Nov 90 11:08:45 -0500
From: John T Kohl <jtkohl@ATHENA.MIT.EDU>
To: Bug-ISODE@NISC.NYSER.NET
Cc: krb5-testers@ATHENA.MIT.EDU, krbdev@ATHENA.MIT.EDU
I've uncovered a bug in the decoding routines generated by ISODE 6.0
(identical code is generated by ISODE 7.0 beta) pepy and/or posy.
The attached ASN.1 description generates code which improperly decodes
the optional sequence 'third' in the 'Frob' type. It seems that the
decoder is counting up the elements in the TOP-LEVEL sequence in order
to discover the number of elements in the optional sequence. The
elements of the optional sequence do appear to get properly decoded,
however. The suspect code is around the first occurrence of prim2seq()
in the generated function decode_FOO_Frob().
posy arguments: -f -h2 -a -s -o FOO-types.py FOO-asn.py
pepy arguments: -a PY_advise -s -r FOO-types.py
FOO DEFINITIONS ::=
BEGIN
Foo ::= SEQUENCE {
a[0] INTEGER,
b[1] INTEGER
}
Frob ::= SEQUENCE {
first[0] INTEGER,
second[1] GeneralizedTime,
third[2] SEQUENCE OF Foo OPTIONAL
}
END
