On 09/21/2012 10:15 AM, Ivan Shmakov wrote: > I'm writing an application which has to support the full range > of Unicode characters being stored to and retrieved from its > (structured) data files, which are to be binary for both space > and performance reasons. > Although my experience with ASN.1 is limited, it seems like a > suitable base format for such an application. As per [1], my > guess is that I should use either UniversalString or UTF8String > to store Unicode text strings (though I don't understand the > difference.) However, as per [2], the only character sequence > type currently supported by GNU Libtasn1 is GeneralString.
We use UTF8String and UniversalString with libtasn1 in gnutls by defining it as: UTF8String ::= [UNIVERSAL 12] IMPLICIT OCTET STRING UniversalString ::= [UNIVERSAL 28] IMPLICIT OCTET STRING Of course there is no checking on whether the contents of these fields conform to any standards. They are treated as opaque. regards, Nikos
