ASN.1 TypeScript Library

• Author: Jonathan M. Wilbur <[email protected]>
• Copyright Year: 2020
• License: MIT License

This library was based off of my D ASN.1 Library.

What is ASN.1?

ASN.1 stands for Abstract Syntax Notation. ASN.1 was first specified in X.680 - Abstract Syntax Notation One (ASN.1), by the International Telecommunications Union. ASN.1 messages can be encoded in one of several encoding/decoding standards. It provides a system of types that are extensible, and can presumably describe every protocol. You can think of it as a protocol for describing other protocols as well as a family of standards for encoding and decoding said protocols. It is similar to Google's Protocol Buffers, or Sun Microsystems' External Data Representation (XDR).

For more information on what ASN.1 is, see documentation/asn1.md.

Why Use This Library?

I believe this library is the first complete implementation of ASN.1, meaning that it implements all data types defined in the specification, and encodes and decodes exactly as specified.

I have never seen any implementation of ASN.1 elsewhere that supports all data types; in fact, most just implement the ten or so most common types. Several implementations I have seen do not support things that are supported by the specification, such as constructed strings.

This library is fully compliant to the specification. If I am wrong, please write up an issue and I will correct it.

Why Not Use This Library?

This library is meant to be fully compliant to the specification, not lightweight. If you need a web application that encodes and decodes very simple ASN.1 data and loads lightning-fast, this may not be the library for you.

Building

You can build this library by running npm run build. The outputs will all be in dist.

• dist/index.js is the root for usage in NodeJS.
• dist/asn1.min.js is the entire ASN.1 library for the web browser, which is minified, and accessible under the variable asn1.

Library Usage

Data Types

In this library, all ASN.1 data types are represented as aliases to TypeScript / JavaScript types in source/macros.ts. For your convenience these aliases are copied here:

export type OPTIONAL<T> = T | undefined;
export type BOOLEAN = boolean;
export type INTEGER = number | bigint;
export type BIT_STRING = Uint8ClampedArray;
export type OCTET_STRING = Uint8Array;
export type NULL = null;
export type OBJECT_IDENTIFIER = ObjectIdentifier;
export type ObjectDescriptor = string;
export type EXTERNAL = External;
export type REAL = number;
export type INSTANCE_OF = External;
export type ENUMERATED = number;
export type EMBEDDED_PDV = EmbeddedPDV;
export type UTF8String = string;
export type RELATIVE_OID = number[];
export type SEQUENCE<T> = T[];
export type SEQUENCE_OF<T> = T[];
export type SET<T> = T[];
export type SET_OF<T> = T[];
export type GraphicString = string;
export type NumericString = string;
export type VisibleString = string;
export type PrintableString = string;
export type ISO646String = string;
export type TeletexString = Uint8Array;
export type GeneralString = string;
export type T61String = Uint8Array;
export type UniversalString = string;
export type VideotexString = Uint8Array;
export type BMPString = string;
export type IA5String = string;
// export type CharacterString = CharacterString;
export { default as CharacterString } from "./types/CharacterString";
export type UTCTime = Date;
export type GeneralizedTime = Date;
export type TIME = string;
export type DATE = Date;
export type TIME_OF_DAY = Date;
export type DATE_TIME = Date;
export type DURATION = DURATION_EQUIVALENT;
export type OID_IRI = string;
export type RELATIVE_OID_IRI = string;

Native ASN.1 types that don't have an obvious corollary to a native JavaScript type have an implementation in the types folder. This includes these types:

• OBJECT IDENTIFIER, implemented as ObjectIdentifier
• EXTERNAL, implemented as External
• EMBEDDED PDV, implemented as EmbeddedPDV
• CHARACTER STRING, implemented as CharacterString

The TYPE-IDENTIFIER object class is implemented as TypeIdentifier. There are also implementation of all of the structured time types specified in the ITU Recommendation X.696 in types/time.

Encoding and Decoding

For each codec in the library, usage entails instantiating the class, then using that class' properties to get and set the encoded value. For all classes, the empty constructor creates an END OF CONTENT element. The remaining constructors will be codec-specific.

Here is a TypeScript example of encoding with Basic Encoding Rules, using the BERElement class.

let el: BERElement = new BERElement();
el.tagClass = ASN1TagClass.universal; // Not technically necessary.
el.construction = ASN1Construction.construction; // Not technically necessary.
el.tagNumber = ASN1UniversalType.integer;
el.integer = 1433; // Now the data is encoded.
console.log(el.integer); // Logs '1433'

... and here is how you would decode that same element:

let encodedData: Uint8Array = el.toBytes();
let el2: BERElement = new BERElement();
el2.fromBytes(encodedData);
console.log(el2.integer); // Logs 1433

Tests under the test directory can also serve as examples.

In this library, you can use the Basic Encoding Rules, Canonical Encoding Rules, and Distinguished Encoding Rules via the BERElement, CERElement, and DERElement classes respectively. You should use DERElement for anything that will be cryptographically signed or hashed.

The tag class can be read and written via the tagClass property using the ASN1TagClass enum. The construction (whether it is constructed or primitive) of the element can be read and written via the construction property using the ASN1Construction enum. The tag number can be set using the tagNumber property. For your convenience, the ASN1UniveralType enum contains the tag numbers of the UNIVERSAL tags by the data type.

Generic "bytes" are represented with the Uint8Array class. Note that Buffer is a subtype of Uint8Array, so you can use a Buffer anywhere you use a Uint8Array (but not the reverse). You can interact with the value bytes of the ASN.1 element through the value property. You can convert ASN.1 elements to and from bytes using toBytes() and fromBytes(). fromBytes() returns an integer indicating the number of bytes read from the Uint8Array. Here is an example of how you would decode multiple back-to-back encoded ASN.1 elements from a buffer:

const encodedElements: BERElement[] = [];
let i: number = 0;
while (i < value.length) {
    const next: BERElement = new BERElement();
    i += next.fromBytes(value.slice(i));
    encodedElements.push(next);
}

Here are the properties available for you to get and set ASN.1 values:

set boolean (value: BOOLEAN);
get boolean (): BOOLEAN;
set integer (value: INTEGER);
get integer (): INTEGER;
set bitString (value: BIT_STRING);
get bitString (): BIT_STRING;
set octetString (value: OCTET_STRING);
get octetString (): OCTET_STRING;
set objectIdentifier (value: OBJECT_IDENTIFIER);
get objectIdentifier (): OBJECT_IDENTIFIER;
set objectDescriptor (value: ObjectDescriptor);
get objectDescriptor (): ObjectDescriptor;
set external (value: EXTERNAL);
get external (): EXTERNAL;
set real (value: REAL);
get real (): REAL;
set enumerated (value: ENUMERATED);
get enumerated (): ENUMERATED;
set embeddedPDV (value: EMBEDDED_PDV);
get embeddedPDV (): EMBEDDED_PDV;
set utf8String (value: UTF8String);
get utf8String (): UTF8String;
set relativeObjectIdentifier (value: RELATIVE_OID);
get relativeObjectIdentifier (): RELATIVE_OID;
set time (value: TIME);
get time (): TIME;
set sequence (value: SEQUENCE<ASN1Element>);
get sequence (): SEQUENCE<ASN1Element>;
set set (value: SET<ASN1Element>);
get set (): SET<ASN1Element>;
set numericString (value: NumericString);
get numericString (): NumericString;
set printableString (value: PrintableString);
get printableString (): PrintableString;
set teletexString (value: TeletexString);
get teletexString (): TeletexString;
set videotexString (value: VideotexString);
get videotexString (): VideotexString;
set ia5String (value: IA5String);
get ia5String (): IA5String;
set utcTime (value: UTCTime);
get utcTime (): UTCTime;
set generalizedTime (value: GeneralizedTime);
get generalizedTime (): GeneralizedTime;
set graphicString (value: GraphicString);
get graphicString (): GraphicString;
set visibleString (value: VisibleString);
get visibleString (): VisibleString;
set generalString (value: GeneralString);
get generalString (): GeneralString;
set universalString (value: UniversalString);
get universalString (): UniversalString;
set characterString (value: CharacterString);
get characterString (): CharacterString;
set bmpString (value: BMPString);
get bmpString (): BMPString;
set date (value: DATE);
get date (): DATE;
set timeOfDay (value: TIME_OF_DAY);
get timeOfDay (): TIME_OF_DAY;
set dateTime (value: DATE_TIME);
get dateTime (): DATE_TIME;
set duration (value: DURATION);
get duration (): DURATION;
set oidIRI (value: OID_IRI);
get oidIRI (): OID_IRI;
set relativeOIDIRI (value: RELATIVE_OID_IRI);
get relativeOIDIRI (): RELATIVE_OID_IRI;

There are shorthand equivalents of the getters and setters above (created to make source files more concise), but these will be removed in a future version.

ASN.1 elements (BERElement, CERElement, and DERElement) support the toString() and toJSON() properties. In general, toString() encodes the elements according to how their values would be represented in an ASN.1 file. In general, toJSON() encodes the elements according to the JSON Encoding Rules (JER).

Finally, there are functional equivalents of the codecs above in functional.ts such as _decodeUTF8String. These will not be documented (for now). They were implemented to support Wildboar Software's ASN.1 Compiler. You can look at the source code for the NPM package @wildboar/x500 for an example for how it works.

Error Handling

This library throws a few subtypes of ASN1Error, which is a subtype of Error. These are:

• ASN1NotImplementedError, which is thrown when some functionality of this library is not implemented.
• ASN1RecursionError, which is thrown when an ASN.1 element is too deeply constructed (e.g. a string is encoded on a construction of constructions of constructions, etc.).
• ASN1TruncationError, which is thrown when the ASN.1 element was too short. This error may be thrown when you are receiving encoded ASN.1 elements over a network and you simply have not received the complete encoding yet. For this reason, if you are reading ASN.1 elements from a network buffer, you may have to catch this error and just wait for more data to come in.
• ASN1OverflowError, which is thrown when an encoded data type, such as an INTEGER or an arc of an OBJECT IDENTIFIER encodes such a large value that this library cannot encode or decode it.
• ASN1SizeError, which is thrown when a value is encoded on an incorrect number of bytes. This can happen with a UNIVERSAL type is encoded on a wrong number of bytes (such as a BOOLEAN being encoded on more than one byte), or it can be thrown by third-party libraries when a SIZE constraint is violated.
• ASN1PaddingError, which is thrown when an encoded value has prohibited padding bytes. For instance, in the Basic Encoding Rules encoding of an INTEGER, the integer must be encoded on the minimum number of octets possible; any encoding containing more than the minimum would throw this error.
• ASN1UndefinedError, which is thrown when some encoding is not defined, such as an unrecognized encoding for a REAL value.
• ASN1CharactersError, which is thrown when a string contains a character that is prohibited for that string type.
• ASN1ConstructionError, which is thrown when the construction of an ASN.1 element is incorrect, such as an INTEGER being "constructed." This may also be thrown by third-party libraries when a SEQUENCE is constructed of an invalid sequence of elements.

BIT STRING

In this library, BIT STRING is represented by a Uint8ClampedArray. Set bits are represented by a value of 1 and unset bits are represented by a value of 0. To "pack" these bytes into bytes, you can use packBits, and to reverse this, you can use unpackBits. Note that this does not include the "unused bits" prefix that the BIT STRING requires; packBits and unpackBits are only responsible for packing eight bits into a byte, and the reverse, respectively.

OBJECT IDENTIFIER

OBJECT IDENTIFIER is represented as objects of the ObjectIdentifier class.

You can encode and decode object identifiers to and from their dot-delimited string representations (e.g. 2.5.4.3) using toString() and ObjectIdentifier.fromString() (the latter is a static method.)

Future Development

• Implement a function for ITU X.690, Section 11.6.
• Implement these codecs:
  • Octet Encoding Rules (This is used by Simple Transportation Management Protocol (STMP) and DATEX-ASN.)
  • Canonical Octet Encoding Rules
  • NTCIP Encoding Rules (This is used by Simple Transportation Management Protocol (STMP) and DATEX-ASN.)
  • JSON Encoding Rules (May require changes to ASN1Element, or a separate element.)
  • Lightweight Encoding Rules (I cannot find a standard anywhere for this.)
  • BACNet Encoding Rules
  • Packed Encoding Rules
    • Basic Aligned Packed Encoding Rules (PER) (This is used by MCS / T.125, which is used by RDP. I believe it is also used by J2735 / DSRC.)
    • Make a separate RDPER (Remoted Desktop Protocol Encoding Rules)
    • Basic Unaligned Packed Encoding Rules (UPER) (May require changes to ASN1Element) (Used by 3GPP RRC)
    • Canonical Aligned Packed Encoding Rules (CPER)
    • Canonical Unaligned Packed Encoding Rules (CUPER) (May require changes to ASN1Element)
• Internationalized strings
• Serverless Functions (May be done in a different repo.)

See Also

• X.680 - Abstract Syntax Notation One (ASN.1)
• X.690 - BER, CER, and DER
• X.691 - Packed Encoding Rules
• X.693 - XML Encoding Rules
• X.696 - Octet Encoding Rules
• X.697 - JSON Encoding Rules
• ISO 16484-5:2017 - BACNet Encoding Rules
• NTCIP 1102:2004 - Octet Encoding Rules (OER) Base Protocol
• ASN.1: Communication Between Heterogeneous Systems by Olivier Dubuisson