Module bcder::guide::asn1_and_ber

A short introduction to ASN.1 and BER.

ASN.1 and Encoding Rules

The Abstract Syntax Notation One (ASN.1) is a formal language that can be used to describe the structure of data. Statements describe both data types and how they are composed from other types. In addition, values of these types can be specified. These rules do not, however, describe how the values are to be represented in files or on the network. This is done by accompanying standards known as encoding rules. Different such rules exist.

On set of such encoding rules is the Basic Encoding Rules (BER). They are a relatively simple type-length-values encoding that can be derived straightforwardly from the ASN.1 notation. BER gives the encoder a number of options to accommodate differing use cases. As this means there is not necessarily a canonical encoding of a type, two restrictive profiles on top of BER have been defined: The Canonical Encoding Rules (CER) cater for cases where data is streamed and the final size of values is not necessarily known up front while Distinguished Encoding Rules (DER) always use the most simple option but require the overall length to be availbale before starting to encode.

ASN.1 is currently defined in ITU-T recommendation X.680, the three encoding rules in X.690. Both these recommendations are freely available from the ITU.

ASN.1 Definitions

ASN.1 collects definitions into documents called modules. All definitions create named objects: data types if the names start with a capital letter or values if they start with a small letter.

As an example, here is how a certificate is defined by RFC 5280:

Certificate  ::=  SEQUENCE  {
     tbsCertificate       TBSCertificate,
     signatureAlgorithm   AlgorithmIdentifier,
     signature            BIT STRING  }

This snippet defines a data type Certificate in terms of a number of other types. The terms in all caps are universal types that are part of the standard. SEQUENCE is a sequence of well defined elements provided one after another within the braces. In this case there are three elements. The first part is the name of the element and the second its type. The two types TBSCertificate and AlgorithmIdentifer are in turn defined elsewhere in the module while BIT STRING is a universal type representing a sequence of bits.

Universal types are the fundamental building blocks that all ASN.1 structures eventually boil down to. Each of these types has an identifying code called their tag. The BIT STRING type, for instance, has the tag 3 in universal class.

Continues:

• implicit v. explicit tags,
• OPTIONAL, DEFAULT.

Encoding ASN.1 Values

The Basic Encoding Rules describe how to encode a values of data types described by ASN.1 defintions. Since these definitions are either fundamental types or composed from other types, BER distinguishes two categories of encodings: primitive encodings for those fundamental types and constructed encodings for types that are composed from other types.

Both categories consist of three parts: identifier octets, length octets, and content octets. The idenitifer octets describe the type of the value. They contain its tag and whether primitive or constructed encoding is used. The length octets determine the length of the value, more specifically the number of octets in the content octet. Those content octets finally contain the value’s content. For values in primitive encoding the interpretation of those octets depends on the type. For values in constructed encoding, the content octets are a sequence of values in BER encoding – i.e., the value is constructed from a sequence of more values.

Continues:

• details on encoding of tag,
• definite v. indefinite length.

Selected Universal Types

ASN.1 defines a wide range of types for all sorts of occassions. In practice, you are likely to encounter only a few. The following has some details on the types that the ber crate supports.

well, it will have, soon …