1 Asn1

1.1 Introduction

Features

The Asn1 application provides:

An ASN.1 compiler for Erlang, which generates encode and decode functions to be used by Erlang programs sending and receiving ASN.1 specified data.
Run-time functions used by the generated code.
The supported encoding rules are:
- Basic Encoding Rules (BER)
- Distinguished Encoding Rules (DER), a specialized form of BER that is used in security-conscious applications.
- Packed Encoding Rules (PER) both the aligned and unaligned variant.

Overview

ASN.1 (Abstract Syntax Notation 1) is a formal language for describing data structures to be exchanged between distributed computer systems. The purpose of ASN.1 is to have a platform and programming language independent notation to express types using a standardized set of rules for the transformation of values of a defined type, into a stream of bytes. This stream of bytes can then be sent on a communication channel set up by the lower layers in the stack of communication protocols e.g. TCP/IP or encapsulated within UDP packets. This way, two different applications written in two completely different programming languages running on different computers with different internal representation of data can exchange instances of structured data types (instead of exchanging bytes or bits). This makes programming faster and easier since no code has to be written to process the transport format of the data.

To write a network application which processes ASN.1 encoded messages, it is prudent and sometimes essential to have a set of off-line development tools such as an ASN.1 compiler which can generate the encode and decode logic for the specific ASN.1 data types. It is also necessary to combine this with some general language-specific runtime support for ASN.1 encoding and decoding.

The ASN.1 compiler must be directed towards a target language or a set of closely related languages. This manual describes a compiler which is directed towards the functional language Erlang. In order to use this compiler, familiarity with the language Erlang is essential. Therefore, the runtime support for ASN.1 is also closely related to the language Erlang and consist of a number of functions, which the compiler uses. The types in ASN.1 and how to represent values of those types in Erlang are described in this manual.

The following document is structured so that the first part describes how to use ASN.1 compiler, and then there are descriptions of all the primitive and constructed ASN.1 types and their representation in Erlang,

Prerequisites

It is assumed that the reader is familiar with the ASN.1 notation as documented in the standard definition [X.680] which is the primary text. It may also be helpful, but not necessary, to read the standard definitions [X.681] [X.682] [X.683] [X.690] [X.691].

A very good book explaining those reference texts is [DUBUISSON], free to download at http://www.oss.com/asn1/dubuisson.html .

Capability

This application covers all features of ASN.1 up to the 1997 edition of the specification. In the 2002 edition of ASN.1 a number of new features where introduced of which some are supported while others are not. For example the ECN (Encoding Control Notation) and XML notation are still unsupported. Though, the other features of the 2002 edition are fully or partly supported as shown below:

Decimal notation (e.g., "1.5e3") for REAL values. The NR1, NR2 and NR3 formats as explained in ISO6093 are supported.
The RELATIVE-OID type for relative object identifiers are fully supported.
The subtype constraint (CONTAINING/ENCODED BY) to constrain the content of an octet string or a bit string is parsed when compiling, but no further action is taken. This constraint is not a PER-visible constraint.
The subtype constraint by regular expressions (PATTERN) for character string types is parsed when compiling, but no further action is taken. This constraint is not a PER-visible constraint.
Multiple-line comments as in C, /* ... */, are supported.

It should also be added here that the encoding formats supported are BER, DER, PER aligned basic variant and PER unaligned basic variant.

1.2 Getting Started with Asn1

A First Example

The following example demonstrates the basic functionality used to run the Erlang ASN.1 compiler.

First, create a file called People.asn containing the following:

People DEFINITIONS IMPLICIT TAGS ::=

BEGIN
EXPORTS Person;

Person ::= [PRIVATE 19] SEQUENCE {
        name PrintableString,
        location INTEGER {home(0),field(1),roving(2)},
        age INTEGER OPTIONAL }
END

This file (people.asn) must be compiled before it can be used. The ASN.1 compiler checks that the syntax is correct and that the text represents proper ASN.1 code before generating an abstract syntax tree. The code-generator then uses the abstract syntax tree in order to generate code.

The generated Erlang files will be placed in the current directory or in the directory specified with the {outdir,Dir} option. The following shows how the compiler can be called from the Erlang shell:

1>asn1ct:compile("People", [ber]).
ok
2>

The verbose option can be given to have information about the generated files printed:

2>asn1ct:compile("People", [ber,verbose]).
Erlang ASN.1 compiling "People.asn" 
--{generated,"People.asn1db"}--
--{generated,"People.hrl"}--
--{generated,"People.erl"}--
ok
3>

The ASN.1 module People is now accepted and the abstract syntax tree is saved in the People.asn1db file, the generated Erlang code is compiled using the Erlang compiler and loaded into the Erlang runtime system. Now there is a user interface of encode/2 and decode/2 in the module People, which is invoked by:
'People':encode(<Type name>,<Value>),
or
'People':decode(<Type name>,<Value>),
Alternatively one can use the asn1rt:encode(<Module name> ,<Type name>,<Value>) and asn1rt:decode(< Module name>,<Type name>,<Value>) calls. However, they are not as efficient as the previous methods since they result in an additional apply/3 call.

Assume there is a network application which receives instances of the ASN.1 defined type Person, modifies and sends them back again:

receive
   {Port,{data,Bytes}} ->
       case 'People':decode('Person',Bytes) of
           {ok,P} ->
               {ok,Answer} = 'People':encode('Person',mk_answer(P)),
               Port ! {self(),{command,Answer}};
           {error,Reason} ->
               exit({error,Reason})
       end
    end,

In the example above, a series of bytes is received from an external source and the bytes are then decoded into a valid Erlang term. This was achieved with the call 'People':decode('Person',Bytes) which returned an Erlang value of the ASN.1 type Person. Then an answer was constructed and encoded using 'People':encode('Person',Answer) which takes an instance of a defined ASN.1 type and transforms it to a binary according to the BER or PER encoding-rules.
The encoder and the decoder can also be run from the shell. The following dialogue with the shell illustrates how the functions asn1rt:encode/3 and asn1rt:decode/3 are used.

2> Rockstar = {'Person',"Some Name",roving,50}.
{'Person',"Some Name",roving,50}
3> {ok,Bin} = asn1rt:encode('People','Person',Rockstar).
{ok,<<243,17,19,9,83,111,109,101,32,78,97,109,101,2,1,2,
      2,1,50>>}
4> {ok,Person} = asn1rt:decode('People','Person',Bin).
{ok,{'Person',"Some Name",roving,50}}
5>

Module dependencies

It is common that asn1 modules import defined types, values and other entities from another asn1 module.

Earlier versions of the asn1 compiler required that modules that were imported from had to be compiled before the module that imported. This caused problems when asn1 modules had circular dependencies.

Now are referenced modules parsed when the compiler finds an entity that is imported. There will not be any code generated for the referenced module. However, the compiled module rely on that the referenced modules also will be compiled.

1.3 The Asn1 Application User Interface

The Asn1 application provides two separate user interfaces:

The module asn1ct which provides the compile-time functions (including the compiler).
The module asn1rt which provides the run-time functions. However, it is preferable to use the generated encode/2 and decode/2 functions in each module, ie. Module:encode(Type,Value), in favor of the asn1rt interface.

The reason for the division of the interface into compile-time and run-time is that only run-time modules (asn1rt*) need to be loaded in an embedded system.

Compile-time Functions

The ASN.1 compiler can be invoked directly from the command-line by means of the erlc program. This is convenient when compiling many ASN.1 files from the command-line or when using Makefiles. Here are some examples of how the erlc command can be used to invoke the ASN.1 compiler:

erlc Person.asn
erlc -bper Person.asn
erlc -bber ../Example.asn
erlc -o ../asnfiles -I ../asnfiles -I /usr/local/standards/asn1 Person.asn

The useful options for the ASN.1 compiler are:

-b[ber | per | uper]: Choice of encoding rules, if omitted ber is the default.
-o OutDirectory: Where to put the generated files, default is the current directory.
-I IncludeDir: Where to search for .asn1db files and asn1 source specs in order to resolve references to other modules. This option can be repeated many times if there are several places to search in. The compiler will always search the current directory first.
+der: DER encoding rule. Only when using -ber option.
+asn1config: This functionality works together with the flags ber. It enables the specialized decodes, see the Specialized Decode chapter.
+undec_rest: A buffer that holds a message, being decoded may also have some following bytes. Now it is possible to get those following bytes returned together with the decoded value. If an asn1 spec is compiled with this option a tuple {ok,Value,Rest} is returned. Rest may be a list or a binary. Earlier versions of the compiler ignored those following bytes.
+'Any Erlc Option': You may add any option to the Erlang compiler when compiling the generated Erlang files. Any option unrecognised by the asn1 compiler will be passed to the Erlang compiler.

For a complete description of erlc see Erts Reference Manual.

The compiler and other compile-time functions can also be invoked from the Erlang shell. Below follows a brief description of the primary functions, for a complete description of each function see the Asn1 Reference Manual, the asn1ct module.

The compiler is invoked by using asn1ct:compile/1 with default options, or asn1ct:compile/2 if explicit options are given. Example:

asn1ct:compile("H323-MESSAGES.asn1").

which equals:

asn1ct:compile("H323-MESSAGES.asn1",[ber]).

If one wants PER encoding:

asn1ct:compile("H323-MESSAGES.asn1",[per]).

The generic encode and decode functions can be invoked like this:

asn1ct:encode('H323-MESSAGES','SomeChoiceType',{call,"octetstring"}).
asn1ct:decode('H323-MESSAGES','SomeChoiceType',Bytes).

Or, preferable like:

'H323-MESSAGES':encode('SomeChoiceType',{call,"octetstring"}).
'H323-MESSAGES':decode('SomeChoiceType',Bytes).

Run-time Functions

A brief description of the major functions is given here. For a complete description of each function see the Asn1 Reference Manual, the asn1rt module.

The generic run-time encode and decode functions can be invoked as below:

asn1rt:encode('H323-MESSAGES','SomeChoiceType',{call,"octetstring"}).
asn1rt:decode('H323-MESSAGES','SomeChoiceType',Bytes).

Or, preferable like:

'H323-MESSAGES':encode('SomeChoiceType',{call,"octetstring"}).
'H323-MESSAGES':decode('SomeChoiceType',Bytes).

The asn1 nif is enabled in two occasions: encoding of asn1 values when the asn1 spec is compiled with per and or decode of encoded asn1 values when the asn1 spec is compiled with ber. In those cases the nif will be loaded automatically at the first call to encode/decode. If one doesn't want the performance overhead of the nif being loaded at the first call it is possible to load the nif separately by loading the asn1rt_nif module.

By invoking the function info/0 in a generated module, one gets information about which compiler options were used.

Errors

Errors detected at compile time appear on the screen together with a line number indicating where in the source file the error was detected. If no errors are found, an Erlang ASN.1 module will be created as default.

The run-time encoders and decoders (in the asn1rt module) do execute within a catch and returns {ok, Data} or {error, {asn1, Description}} where Description is an Erlang term describing the error.

1.4 Multi File Compilation

There are various reasons for using a multi file compilation:

You want to choose name for the generated module by any reason. Maybe you need to compile the same specs for different encoding/decoding standards.
You want only one resulting module.

You need to specify which asn1 specs you will compile in a module that must have the extension .set.asn. You chose name of the module and provide the names of the asn1 specs. For instance, if you have the specs File1.asn, File2.asn and File3.asn your module MyModule.set.asn will look like:

File1.asn
File2.asn
File3.asn

If you compile with:

~> erlc MyModule.set.asn

the result will be one merged module MyModule.erl with the generated code from the three asn1 specs.

1.5 The ASN.1 Types

This section describes the ASN.1 types including their functionality, purpose and how values are assigned in Erlang.

ASN.1 has both primitive and constructed types:

Primitive types	Constructed types
BOOLEAN	SEQUENCE
INTEGER	SET
REAL	CHOICE
NULL	SET OF and SEQUENCE OF
ENUMERATED	ANY
BIT STRING	ANY DEFINED BY
OCTET STRING	EXTERNAL
Character Strings	EMBEDDED PDV
OBJECT IDENTIFIER	CHARACTER STRING
Object Descriptor
The TIME types

Table 1.1: The supported ASN.1 types

Note

Values of each ASN.1 type has its own representation in Erlang described in the following subsections. Users shall provide these values for encoding according to the representation, as in the example below.

Operational ::= BOOLEAN --ASN.1 definition

In Erlang code it may look like:

Val = true,
{ok,Bytes}=asn1rt:encode(MyModule,'Operational',Val),

Below follows a description of how values of each type can be represented in Erlang.

BOOLEAN

Booleans in ASN.1 express values that can be either TRUE or FALSE. The meanings assigned to TRUE or FALSE is beyond the scope of this text.
In ASN.1 it is possible to have:

Operational ::= BOOLEAN

Assigning a value to the type Operational in Erlang is possible by using the following Erlang code:

Myvar1 = true,

Thus, in Erlang the atoms true and false are used to encode a boolean value.

INTEGER

ASN.1 itself specifies indefinitely large integers, and the Erlang systems with versions 4.3 and higher, support very large integers, in practice indefinitely large integers.

The concept of sub-typing can be applied to integers as well as to other ASN.1 types. The details of sub-typing are not explained here, for further info see [X.680]. A variety of syntaxes are allowed when defining a type as an integer:

T1 ::= INTEGER
T2 ::= INTEGER (-2..7)
T3 ::= INTEGER (0..MAX)
T4 ::= INTEGER (0<..MAX)
T5 ::= INTEGER (MIN<..-99)
T6 ::= INTEGER {red(0),blue(1),white(2)}

The Erlang representation of an ASN.1 INTEGER is an integer or an atom if a so called Named Number List (see T6 above) is specified.

Below is an example of Erlang code which assigns values for the above types:

T1value = 0,
T2value = 6,
T6value1 = blue,
T6value2 = 0,
T6value3 = white

The Erlang variables above are now bound to valid instances of ASN.1 defined types. This style of value can be passed directly to the encoder for transformation into a series of bytes.

The decoder will return an atom if the value corresponds to a symbol in the Named Number List.

REAL

In this version reals are not implemented. When they are, the following ASN.1 type is used:

R1 ::= REAL

Can be assigned a value in Erlang as:

R1value1 = 2.14,
R1value2 = {256,10,-2},

In the last line note that the tuple {256,10,-2} is the real number 2.56 in a special notation, which will encode faster than simply stating the number as 2.56. The arity three tuple is {Mantissa,Base,Exponent} i.e. Mantissa * Base^Exponent.

NULL

Null is suitable in cases where supply and recognition of a value is important but the actual value is not.

Notype ::= NULL

The NULL type can be assigned in Erlang:

N1 = 'NULL',

The actual value is the quoted atom 'NULL'.

ENUMERATED

The enumerated type can be used, when the value we wish to describe, may only take one of a set of predefined values.

DaysOfTheWeek ::= ENUMERATED { 
    sunday(1),monday(2),tuesday(3),
    wednesday(4),thursday(5),friday(6),saturday(7) }

For example to assign a weekday value in Erlang use the same atom as in the Enumerations of the type definition:

Day1 = saturday,

The enumerated type is very similar to an integer type, when defined with a set of predefined values. An enumerated type differs from an integer in that it may only have specified values, whereas an integer can also have any other value.

BIT STRING

The BIT STRING type can be used to model information which is made up of arbitrary length series of bits. It is intended to be used for a selection of flags, not for binary files.
In ASN.1 BIT STRING definitions may look like:

Bits1 ::= BIT STRING
Bits2 ::= BIT STRING {foo(0),bar(1),gnu(2),gnome(3),punk(14)}

There are five different notations available for representation of BIT STRING values in Erlang and as input to the encode functions.

A bitstring. By default, a BIT STRING with no symbolic names will be decoded to an Erlang bitstring.
A list of atoms corresponding to atoms in the NamedBitList in the BIT STRING definition. A BIT STRING with symbolic names will always be decoded to this format.
A list of binary digits (0 or 1). This format is always accepted as input to the encode functions. A BIT STRING will be decoded to this format if legacy_bit_string option has been given. This format may be withdrawn in a future release.
As {Unused,Binary} where Unused denotes how many trailing zero-bits 0 to 7 that are unused in the least significant byte in Binary. This format is always accepted as input to the encode functions. A BIT STRING will be decoded to this format if compact_bit_string has been given. This format may be withdrawn in a future release.
A hexadecimal number (or an integer). This format should be avoided, since it is easy to misinterpret a BIT STRING value in this format. This format may be withdrawn in a future release.

Note

It is recommended to either use the bitstring format (for BIT STRINGs with no symbolic names) or a list of symbolic names (for BIT STRINGs with symbolic names). The other formats should be avoided since they may be withdrawn in a future release.

Bits1Val1 = <<0:1,1:1,0:1,1:1,1:1>>,
Bits1Val2 = 16#1A,
Bits1Val3 = {3,<<0:1,1:1,0:1,1:1,1:1,0:3>>},
Bits1Val4 = [0,1,0,1,1]

Note that Bits1Val1, Bits1Val2, Bits1Val3, and Bits1Val1 denote the same value.

Bits2Val1 = [gnu,punk],
Bits2Val2 = <<2#1110:4>>,
Bits2Val3 = [bar,gnu,gnome],

Bits2Val2 and Bits2Val3 above denote the same value.

Bits2Val1 is assigned symbolic values. The assignment means that the bits corresponding to gnu and punk i.e. bits 2 and 14 are set to 1 and the rest set to 0. The symbolic values appear as a list of values. If a named value appears, which is not specified in the type definition, a run-time error will occur.

The compact notation equivalent to the empty BIT STRING is {0,<<>>}, which in the other notations is <<>>, [], or 0.

BIT STRINGS may also be sub-typed with, for example, a SIZE specification:

Bits3 ::= BIT STRING (SIZE(0..31))

This means that no bit higher than 31 can ever be set.

OCTET STRING

The OCTET STRING is the simplest of all ASN.1 types The OCTET STRING only moves or transfers e.g. binary files or other unstructured information complying to two rules. Firstly, the bytes consist of octets and secondly, encoding is not required.

It is possible to have the following ASN.1 type definitions:

O1 ::= OCTET STRING
O2 ::= OCTET STRING (SIZE(28))

With the following example assignments in Erlang:

O1Val = [17,13,19,20,0,0,255,254],
O2Val = "must be exactly 28 chars....",

Observe that O1Val is assigned a series of numbers between 0 and 255 i.e. octets. O2Val is assigned using the string notation.

Character Strings

ASN.1 supports a wide variety of character sets. The main difference between OCTET STRINGS and the Character strings is that OCTET STRINGS have no imposed semantics on the bytes delivered.

However, when using for instance the IA5String (which closely resembles ASCII) the byte 65 (in decimal notation) means the character 'A'.

For example, if a defined type is to be a VideotexString and an octet is received with the unsigned integer value X, then the octet should be interpreted as specified in the standard ITU-T T.100,T.101.

The ASN.1 to Erlang compiler will not determine the correct interpretation of each BER (Basic Encoding Rules) string octet value with different Character strings. Interpretation of octets is the responsibility of the application. Therefore, from the BER string point of view, octets appear to be very similar to character strings and are compiled in the same way.

It should be noted that when PER (Packed Encoding Rules) is used, there is a significant difference in the encoding scheme between OCTET STRINGS and other strings. The constraints specified for a type are especially important for PER, where they affect the encoding.

Please note that all the Character strings are supported and it is possible to use the following ASN.1 type definitions:

Digs ::= NumericString (SIZE(1..3))
TextFile ::= IA5String (SIZE(0..64000))

and the following Erlang assignments:

DigsVal1 = "456",
DigsVal2 = "123",
TextFileVal1 = "abc...xyz...",
TextFileVal2 = [88,76,55,44,99,121 .......... a lot of characters here ....]

The Erlang representation for "BMPString" and "UniversalString" is either a list of ASCII values or a list of quadruples. The quadruple representation associates to the Unicode standard representation of characters. The ASCII characters are all represented by quadruples beginning with three zeros like {0,0,0,65} for the 'A' character. When decoding a value for these strings the result is a list of quadruples, or integers when the value is an ASCII character. The following example shows how it works:

In a file PrimStrings.asn1 the type BMP is defined as
BMP ::= BMPString then using BER encoding (ber option)the input/output format will be:

1> {ok,Bytes1} = asn1rt:encode('PrimStrings','BMP',[{0,0,53,53},{0,0,45,56}]).
{ok,[30,4,"55-8"]}
2> asn1rt:decode('PrimStrings','BMP',list_to_binary(Bytes1)).
{ok,[{0,0,53,53},{0,0,45,56}]}
3> {ok,Bytes2} = asn1rt:encode('PrimStrings','BMP',[{0,0,53,53},{0,0,0,65}]).
{ok,[30,4,[53,53,0,65]]}
4> asn1rt:decode('PrimStrings','BMP',list_to_binary(Bytes2)).
{ok,[{0,0,53,53},65]}
5> {ok,Bytes3} = asn1rt:encode('PrimStrings','BMP',"BMP string").
{ok,[30,20,[0,66,0,77,0,80,0,32,0,115,0,116,0,114,0,105,0,110,0,103]]}
6> asn1rt:decode('PrimStrings','BMP',list_to_binary(Bytes3)).
{ok,"BMP string"}

The UTF8String is represented in Erlang as a list of integers, where each integer represents the unicode value of one character. When a value shall be encoded one first has to transform it to a UTF8 encoded binary, then it can be encoded by asn1. When decoding the result is a UTF8 encoded binary, which may be transformed to an integer list. The transformation functions, utf8_binary_to_list and utf8_list_to_binary, are in the asn1rt module. In the example below we assume an asn1 definition UTF ::= UTF8String in a module UTF.asn:

1> asn1ct:compile('UTF',[ber]).
Erlang ASN.1 version "1.4.3.3" compiling "UTF.asn" 
Compiler Options: [ber]
--{generated,"UTF.asn1db"}--
--{generated,"UTF.erl"}--
ok
2> UTF8Val1 = "hello".
"hello"
3> {ok,UTF8bin1} = asn1rt:utf8_list_to_binary(UTF8Val1).
{ok,<<104,101,108,108,111>>}
4> {ok,B}='UTF':encode('UTF',UTF8bin1).
{ok,[12,
     5,
     <<104,101,108,108,111>>]}
5> Bin = list_to_binary(B).
<<12,5,104,101,108,108,111>>
6> {ok,UTF8bin1}='UTF':decode('UTF',Bin).
{ok,<<104,101,108,108,111>>}
7> asn1rt:utf8_binary_to_list(UTF8bin1).
{ok,"hello"}           
8> UTF8Val2 = [16#00,16#100,16#ffff,16#ffffff].
[0,256,65535,16777215]
9> {ok,UTF8bin2} = asn1rt:utf8_list_to_binary(UTF8Val2).
{ok,<<0,196,128,239,191,191,248,191,191,191,191>>}
10> {ok,B2} = 'UTF':encode('UTF',UTF8bin2).
{ok,[12,
     11,
     <<0,196,128,239,191,191,248,191,191,191,191>>]}
11> Bin2 = list_to_binary(B2).
<<12,11,0,196,128,239,191,191,248,191,191,191,191>>
12> {ok,UTF8bin2} = 'UTF':decode('UTF',Bin2).
{ok,<<0,196,128,239,191,191,248,191,191,191,191>>}
13> asn1rt:utf8_binary_to_list(UTF8bin2).
{ok,[0,256,65535,16777215]}
14>

OBJECT IDENTIFIER

The OBJECT IDENTIFIER is used whenever a unique identity is required. An ASN.1 module, a transfer syntax, etc. is identified with an OBJECT IDENTIFIER. Assume the example below:

Oid ::= OBJECT IDENTIFIER

Therefore, the example below is a valid Erlang instance of the type 'Oid'.

OidVal1 = {1,2,55},

The OBJECT IDENTIFIER value is simply a tuple with the consecutive values which must be integers.

The first value is limited to the values 0, 1 or 2 and the second value must be in the range 0..39 when the first value is 0 or 1.

The OBJECT IDENTIFIER is a very important type and it is widely used within different standards to uniquely identify various objects. In [DUBUISSON], there is an easy-to-understand description of the usage of OBJECT IDENTIFIER.

Object Descriptor

Values of this type can be assigned a value as an ordinary string i.e.
"This is the value of an Object descriptor"

The TIME Types

Two different time types are defined within ASN.1, Generalized Time and UTC (Universal Time Coordinated), both are assigned a value as an ordinary string within double quotes i.e. "19820102070533.8".

In case of DER encoding the compiler does not check the validity of the time values. The DER requirements upon those strings is regarded as a matter for the application to fulfill.

SEQUENCE

The structured types of ASN.1 are constructed from other types in a manner similar to the concepts of array and struct in C.
A SEQUENCE in ASN.1 is comparable with a struct in C and a record in Erlang. A SEQUENCE may be defined as:

Pdu ::= SEQUENCE {
   a INTEGER,
   b REAL,
   c OBJECT IDENTIFIER,
   d NULL }

This is a 4-component structure called 'Pdu'. The major format for representation of SEQUENCE in Erlang is the record format. For each SEQUENCE and SET in an ASN.1 module an Erlang record declaration is generated. For Pdu above, a record like this is defined:

-record('Pdu',{a, b, c, d}).

The record declarations for a module M are placed in a separate M.hrl file.

Values can be assigned in Erlang as shown below:

MyPdu = #'Pdu'{a=22,b=77.99,c={0,1,2,3,4},d='NULL'}.

The decode functions will return a record as result when decoding a SEQUENCE or a SET.

A SEQUENCE and a SET may contain a component with a DEFAULT key word followed by the actual value that is the default value. In case of BER encoding it is optional to encode the value if it equals the default value. If the application uses the atom asn1_DEFAULT as value or if the value is a primitive value that equals the default value the encoding omits the bytes for this value, which is more efficient and it results in fever bytes to send to the receiving application.

For instance, if the following types exists in a file "File.asn":

Seq1 ::= SEQUENCE {
    a INTEGER DEFAULT 1,
    b Seq2 DEFAULT {aa TRUE, bb 15}
}

Seq2 ::= SEQUENCE {
    aa BOOLEAN,
    bb INTEGER
}

Some values and the corresponding encoding in an Erlang terminal is shown below:

1> asn1ct:compile('File').
Erlang ASN.1 version "1.3.2" compiling "File.asn1" 
Compiler Options: []
--{generated,"File.asn1db"}--
--{generated,"File.hrl"}--
--{generated,"File.erl"}--
ok
2> 'File':encode('Seq1',{'Seq1',asn1_DEFAULT,asn1_DEFAULT}).
{ok,["0",[0],[[],[]]]}
3> lists:flatten(["0",[0],[[],[]]]).
[48,0]
4> 'File':encode('Seq1',{'Seq1',1,{'Seq2',true,15}}).
{ok,["0","\\b",[[],["\\241",[6],[[[128],[1],"\\377"],[[129],[1],[15]]]]]]}
5> lists:flatten(["0","\\b",[[],["\\241",[6],[[[128],[1],"\\377"],[[129],[1],[15]]]]]]).
[48,8,161,6,128,1,255,129,1,15]
6>

The result after command line 3, in the example above,shows that the encoder omits the encoding of default values when they are specific by asn1_DEFAULT. Line 5 shows that even primitive values that equals the default value are detected and not encoded. But the constructed value of component b in Seq1 is not recognized as the default value. Checking of default values in BER is not done in case of complex values, because it would be to expensive.

But, the DER encoding format has stronger requirements regarding default values both for SET and SEQUENCE. A more elaborate and time expensive check of default values will take place. The following is an example with the same types and values as above but with der encoding format.

1> asn1ct:compile('File',[der]).
Erlang ASN.1 version "1.3.2" compiling "File.asn1" 
Compiler Options: [der]
--{generated,"File.asn1db"}--
--{generated,"File.hrl"}--
--{generated,"File.erl"}--
ok
2> 'File':encode('Seq1',{'Seq1',asn1_DEFAULT,asn1_DEFAULT}).
{ok,["0",[0],[[],[]]]}
3> lists:flatten(["0",[0],[[],[]]]).
[48,0]
4> 'File':encode('Seq1',{'Seq1',1,{'Seq2',true,15}}).
{ok,["0",[0],[[],[]]]}
5> lists:flatten(["0",[0],[[],[]]]).
[48,0]
6>

Line 5 shows that even values of constructed types is checked and if it equals the default value it will not be encoded.

SET

The SET type is an unusual construct and normally the SEQUENCE type is more appropriate to use. Set is also inefficient compared with SEQUENCE, as the components can be in any order. Hence, it must be possible to distinguish every component in 'SET', both when encoding and decoding a value of a type defined to be a SET. The tags of all components must be different from each other in order to be easily recognizable.

A SET may be defined as:

Pdu2 ::= SET {
    a INTEGER,
    b BOOLEAN,
    c ENUMERATED {on(0),off(1)} }

A SET is represented as an Erlang record. For each SEQUENCE and SET in an ASN.1 module an Erlang record declaration is generated. For Pdu2 above a record is defined like this:

-record('Pdu2',{a, b, c}).

The record declarations for a module M are placed in a separate M.hrl file.

Values can be assigned in Erlang as demonstrated below:

V = #'Pdu2'{a=44,b=false,c=off}.

The decode functions will return a record as result when decoding a SET.

The difference between SET and SEQUENCE is that the order of the components (in the BER encoded format) is undefined for SET and defined as the lexical order from the ASN.1 definition for SEQUENCE. The ASN.1 compiler for Erlang will always encode a SET in the lexical order. The decode routines can handle SET components encoded in any order but will always return the result as a record. Since all components of the SET must be distinguishable both in the encoding phase as well as the decoding phase the following type is not allowed in a module with EXPLICIT or IMPLICIT as tag-default :

Bad ::= SET {i INTEGER,
             j INTEGER }

The ASN.1 to Erlang compiler rejects the above type. We shall not explain the concept of tag further here, we refer to [X.680].

Encoding of a SET with components with DEFAULT values behaves similar as a SEQUENCE, see above. The DER encoding format restrictions on DEFAULT values is the same for SET as for SEQUENCE, and is supported by the compiler, see above.

Moreover, in DER the elements of a SET will be sorted. If a component is an un-tagged choice the sorting have to take place in run-time. This fact emphasizes the following recommendation if DER encoding format is used.

The concept of SET is an unusual construct and one cannot think of one single application where the set type is essential. (Imagine if someone "invented'' the shuffled array in 'C') People tend to think that 'SET' sounds nicer and more mathematical than 'SEQUENCE' and hence use it when 'SEQUENCE' would have been more appropriate. It is also most inefficient, since every correct implementation of SET must always be prepared to accept the components in any order. So, if possible use SEQUENCE instead of SET.

Notes about Extend-ability for SEQUENCE and SET

When a SEQUENCE or SET contains an extension marker and extension components like this:

SExt ::= SEQUENCE {
           a INTEGER,
           ...,
           b BOOLEAN }

It means that the type may get more components in newer versions of the ASN.1 spec. In this case it has got a new component b. Thus, incoming messages that will be decoded may have more or fever components than this one.

The component b will be treated as an original component when encoding a message. In this case, as it is not an optional element, it must be encoded.

During decoding the b field of the record will get the decoded value of the b component if present and otherwise the value asn1_NOVALUE.

CHOICE

The CHOICE type is a space saver and is similar to the concept of a 'union' in the C-language. As with the previous SET-type, the tags of all components of a CHOICE need to be distinct. If AUTOMATIC TAGS are defined for the module (which is preferable) the tags can be omitted completely in the ASN.1 specification of a CHOICE.

Assume:

T ::= CHOICE {
        x [0] REAL,
        y [1] INTEGER,
        z [2] OBJECT IDENTIFIER }

It is then possible to assign values:

TVal1 = {y,17},
TVal2 = {z,{0,1,2}},

A CHOICE value is always represented as the tuple {ChoiceAlternative, Val} where ChoiceAlternative is an atom denoting the selected choice alternative.

It is also allowed to have a CHOICE type tagged as follow:

C ::= [PRIVATE 111] CHOICE {
        C1,
        C2 }

C1 ::= CHOICE { 
         a [0] INTEGER,
         b [1] BOOLEAN }

C2 ::= CHOICE {
         c [2] INTEGER,
         d [3] OCTET STRING }

In this case, the top type C appears to have no tags at all in its components, however, both C1 and C2 are also defined as CHOICE types and they have distinct tags among themselves. Hence, the above type C is both legal and allowed.

Extendable CHOICE

When a CHOICE contains an extension marker and the decoder detects an unknown alternative of the CHOICE the value is represented as:

{asn1_ExtAlt, BytesForOpenType}

Where BytesForOpenType is a list of bytes constituting the encoding of the "unknown" CHOICE alternative.

SET OF and SEQUENCE OF

The SET OF and SEQUENCE OF types correspond to the concept of an array found in several programming languages. The Erlang syntax for both of these types is straight forward. For example:

Arr1 ::= SET SIZE (5) OF INTEGER (4..9) 
Arr2 ::= SEQUENCE OF OCTET STRING

We may have the following in Erlang:

Arr1Val = [4,5,6,7,8],
Arr2Val = ["abc",[14,34,54],"Octets"],

Please note that the definition of the SET OF type implies that the order of the components is undefined, but in practice there is no difference between SET OF and SEQUENCE OF. The ASN.1 compiler for Erlang does not randomize the order of the SET OF components before encoding.

However, in case of a value of the type SET OF, the DER encoding format requires the elements to be sent in ascending order of their encoding, which implies an expensive sorting procedure in run-time. Therefore it is strongly recommended to use SEQUENCE OF instead of SET OF if it is possible.

ANY and ANY DEFINED BY

The types ANY and ANY DEFINED BY have been removed from the standard since 1994. It is recommended not to use these types any more. They may, however, exist in some old ASN.1 modules. The idea with this type was to leave a "hole" in a definition where one could put unspecified data of any kind, even non ASN.1 data.

A value of this type is encoded as an open type.

Instead of ANY/ANY DEFINED BY one should use information object class, table constraints and parameterization. In particular the construct TYPE-IDENTIFIER.@Type accomplish the same as the deprecated ANY.

EXTERNAL, EMBEDDED PDV and CHARACTER STRING

These types are used in presentation layer negotiation. They are encoded according to their associated type, see [X.680].

The EXTERNAL type had a slightly different associated type before 1994. [X.691] states that encoding shall follow the older associate type. Therefore does generated encode/decode functions convert values of the newer format to the older format before encoding. This implies that it is allowed to use EXTERNAL type values of either format for encoding. Decoded values are always returned on the newer format.

Embedded Named Types

The structured types previously described may very well have other named types as their components. The general syntax to assign a value to the component C of a named ASN.1 type T in Erlang is the record syntax #'T'{'C'=Value}. Where Value may be a value of yet another type T2.

For example:

B ::= SEQUENCE {
        a Arr1,
        b [0] T }

Arr1 ::= SET SIZE (5) OF INTEGER (4..9) 

T ::= CHOICE {
        x [0] REAL,
        y [1] INTEGER,
        z [2] OBJECT IDENTIFIER }

The above example can be assigned like this in Erlang:

V2 = #'B'{a=[4,5,6,7,8], b={x,7.77}}.

1.6 Naming of Records in .hrl Files

When an asn1 specification is compiled all defined types of type SET or SEQUENCE will result in a corresponding record in the generated hrl file. This is because the values for SET/SEQUENCE as mentioned in sections above are represented as records.

Though there are some special cases of this functionality that are presented below.

Embedded Structured Types

It is also possible in ASN.1 to have components that are themselves structured types. For example, it is possible to have:

Emb ::= SEQUENCE {
    a SEQUENCE OF OCTET STRING,
    b SET {
       a [0] INTEGER,
       b [1] INTEGER DEFAULT 66},
    c CHOICE {
       a INTEGER,
       b FooType } }

FooType ::= [3] VisibleString

The following records are generated because of the type Emb:

-record('Emb,{a, b, c}).
-record('Emb_b',{a, b = asn1_DEFAULT}). % the embedded SET type

Values of the Emb type can be assigned like this:

V = #'Emb'{a=["qqqq",[1,2,255]], 
           b = #'Emb_b'{a=99}, 
           c ={b,"Can you see this"}}.

For an embedded type of type SEQUENCE/SET in a SEQUENCE/SET the record name is extended with an underscore and the component name. If the embedded structure is deeper with SEQUENCE, SET or CHOICE types in the line, each component-/alternative-name will be added to the record-name.

For example:

Seq ::= SEQUENCE{
    a CHOICE{
        b SEQUENCE {
           c  INTEGER
        }
    }
}

will result in the following record:

-record('Seq_a_b',{c}).

If the structured type has a component with an embedded SEQUENCE OF/SET OF which embedded type in turn is a SEQUENCE/SET it will give a record with the SEQOF/SETOF addition as in the following example:

Seq ::= SEQUENCE {
    a SEQUENCE OF SEQUENCE {
           b
               }
    c SET OF SEQUENCE {
           d
               }
}

This results in the records:

-record('Seq_a_SEQOF'{b}).
-record('Seq_c_SETOF'{d}).

A parameterized type should be considered as an embedded type. Each time a such type is referenced an instance of it is defined. Thus in the following example a record with name 'Seq_b' is generated in the .hrl file and used to hold values.

Seq ::= SEQUENCE {
    b PType{INTEGER}
}

PType{T} ::= SEQUENCE{
    id T
}

Recursive Types

Types may refer to themselves. Suppose:

Rec ::= CHOICE {
     nothing [0] NULL,
     something SEQUENCE {
          a INTEGER,
          b OCTET STRING,
          c Rec }}

This type is recursive; that is, it refers to itself. This is allowed in ASN.1 and the ASN.1-to-Erlang compiler supports this recursive type. A value for this type is assigned in Erlang as shown below:

V = {something,#'Rec_something'{a = 77, 
                                b = "some octets here", 
                                c = {nothing,'NULL'}}}.

1.7 ASN.1 Values

Values can be assigned to ASN.1 type within the ASN.1 code itself, as opposed to the actions taken in the previous chapter where a value was assigned to an ASN.1 type in Erlang. The full value syntax of ASN.1 is supported and [X.680] describes in detail how to assign values in ASN.1. Below is a short example:

TT ::= SEQUENCE {
   a INTEGER,
   b SET OF OCTET STRING }

tt TT ::= {a 77,b {"kalle","kula"}}

The value defined here could be used in several ways. Firstly, it could be used as the value in some DEFAULT component:

SS ::= SET {
    s [0] OBJECT IDENTIFIER,
    val TT DEFAULT tt }

It could also be used from inside an Erlang program. If the above ASN.1 code was defined in ASN.1 module Values, then the ASN.1 value tt can be reached from Erlang as a function call to 'Values':tt() as in the example below.

1> Val = 'Values':tt().
{'TT',77,["kalle","kula"]}
2> {ok,Bytes} = 'Values':encode('TT',Val).
{ok,<<48,18,128,1,77,161,13,4,5,107,97,108,108,101,4,4,
      107,117,108,97>>}
4> 'Values':decode('TT',Bytes).
{ok,{'TT',77,["kalle","kula"]}}
5>

The above example shows that a function is generated by the compiler that returns a valid Erlang representation of the value, even though the value is of a complex type.

Furthermore, there is a macro generated for each value in the .hrl file. So, the defined value tt can also be extracted by ?tt in application code.

1.8 Macros

MACRO is not supported as the the type is no longer part of the ASN.1 standard.

1.9 ASN.1 Information Objects (X.681)

Information Object Classes, Information Objects and Information Object Sets, (in the following called classes, objects and object sets respectively), are defined in the standard definition [X.681]. In the following only a brief explanation is given.

These constructs makes it possible to define open types, i.e. values of that type can be of any ASN.1 type. It is also possible to define relationships between different types and values, since classes can hold types, values, objects, object sets and other classes in its fields. An Information Object Class may be defined in ASN.1 as:

GENERAL-PROCEDURE ::= CLASS {
      &Message,
      &Reply               OPTIONAL,
      &Error               OPTIONAL,
      &id          PrintableString UNIQUE
}
WITH SYNTAX {
      NEW MESSAGE     &Message
      [REPLY           &Reply]
      [ERROR           &Error]
      ADDRESS          &id
}

An object is an instance of a class and an object set is a set containing objects of one specified class. A definition may look like below.

The object object1 is an instance of the CLASS GENERAL-PROCEDURE and has one type field and one fixed type value field. The object object2 also has an OPTIONAL field ERROR, which is a type field.

object1 GENERAL-PROCEDURE ::= {
    NEW MESSAGE      PrintableString
    ADDRESS          "home"
}

object2 GENERAL-PROCEDURE ::= {
    NEW MESSAGE INTEGER
    ERROR INTEGER
    ADDRESS "remote"
}

The field ADDRESS is a UNIQUE field. Objects in an object set must have unique values in their UNIQUE field, as in GENERAL-PROCEDURES:

GENERAL-PROCEDURES GENERAL-PROCEDURE ::= {
    object1 | object2}

One can not encode a class, object or object set, only referring to it when defining other ASN.1 entities. Typically one refers to a class and to object sets by table constraints and component relation constraints [X.682] in ASN.1 types, as in:

StartMessage  ::= SEQUENCE {
    msgId  GENERAL-PROCEDURE.&id  ({GENERAL-PROCEDURES}),
    content GENERAL-PROCEDURE.&Message ({GENERAL-PROCEDURES}{@msgId}),
    }

In the type StartMessage the constraint following the content field tells that in a value of type StartMessage the value in the content field must come from the same object that is chosen by the msgId field.

So, the value #'StartMessage'{msgId="home",content="Any Printable String"} is legal to encode as a StartMessage value, while the value #'StartMessage'{msgId="remote", content="Some String"} is illegal since the constraint in StartMessage tells that when you have chosen a value from a specific object in the object set GENERAL-PROCEDURES in the msgId field you have to choose a value from that same object in the content field too. In this second case it should have been any INTEGER value.

StartMessage can in the content field be encoded with a value of any type that an object in the GENERAL-PROCEDURES object set has in its NEW MESSAGE field. This field refers to a type field &Message in the class. The msgId field is always encoded as a PrintableString, since the field refers to a fixed type in the class.

1.10 Parameterization (X.683)

Parameterization, which is defined in the standard [X.683], can be used when defining types, values, value sets, information object classes, information objects or information object sets. A part of a definition can be supplied as a parameter. For instance, if a Type is used in a definition with certain purpose, one want the type-name to express the intention. This can be done with parameterization.

When many types (or an other ASN.1 entity) only differs in some minor cases, but the structure of the types are similar, only one general type can be defined and the differences may be supplied through parameters.

One example of use of parameterization is:

General{Type} ::= SEQUENCE
{
     number     INTEGER,
     string     Type
}
      
T1 ::= General{PrintableString}

T2 ::= General{BIT STRING}

An example of a value that can be encoded as type T1 is {12,"hello"}.

Observe that the compiler not generates encode/decode functions for parameterized types, only for the instances of the parameterized types. So, if a file contains the types General{}, T1 and T2 above, encode/decode functions will only be generated for T1 and T2.

1.11 Tags

Every built-in ASN.1 type, except CHOICE and ANY have a universal tag. This is a unique number that clearly identifies the type.
It is essential for all users of ASN.1 to understand all the details about tags.

Tags are implicitly encoded in the BER encoding as shown below, but are hardly not accounted for in the PER encoding. In PER tags are used for instance to sort the components of a SET.

There are four different types of tags.

universal: For types whose meaning is the same in all applications. Such as integers, sequences and so on; that is, all the built in types.
application: For application specific types for example, the types in X.400 Message handling service have this sort of tag.
private: For your own private types.
context: This is used to distinguish otherwise indistinguishable types in a specific context. For example, if we have two components of a CHOICE type that are both INTEGER values, there is no way for the decoder to decipher which component was actually chosen, since both components will be tagged as INTEGER. When this or similar situations occur, one or both of the components should be given a context specific to resolve the ambiguity.

The tag in the case of the 'Apdu' type [PRIVATE 1] is encoded to a sequence of bytes making it possible for a decoder to look at the (initial) bytes that arrive and determine whether the rest of the bytes must be of the type associated with that particular sequence of bytes. This means that each tag must be uniquely associated with only one ASN.1 type.

Immediately following the tag is a sequence of bytes informing the decoder of the length of the instance. This is sometimes referred to as TLV (Tag length value) encoding. Hence, the structure of a BER encoded series of bytes is as shown in the table below.

Tag

Len

Value

Table 1.2: Structure of a BER encoded series of bytes

1.12 Encoding Rules

When the first recommendation on ASN.1 was released 1988 it was accompanied with the Basic Encoding Rules, BER, as the only alternative for encoding. BER is a somewhat verbose protocol. It adopts a so-called TLV (type, length, value) approach to encoding in which every element of the encoding carries some type information, some length information and then the value of that element. Where the element is itself structured, then the Value part of the element is itself a series of embedded TLV components, to whatever depth is necessary. In summary, BER is not a compact encoding but is relatively fast and easy to produce.

The DER (Distinguished Encoding Rule) encoding format was included in the standard in 1994. It is a specialized form of BER, which gives the encoder the option to encode some entities differently. For instance, is the value for TRUE any octet with any bit set to one. But, DER does not leave any such choices. The value for TRUE in the DER case is encoded as the octet 11111111. So, the same value encoded by two different DER encoders must result in the same bit stream.

A more compact encoding is achieved with the Packed Encoding Rules PER which was introduced together with the revised recommendation in 1994. PER takes a rather different approach from that taken by BER. The first difference is that the tag part in the TLV is omitted from the encodings, and any tags in the notation are not encoded. The potential ambiguities are resolved as follows:

A CHOICE is encoded by first encoding a choice index which identifies the chosen alternative by its position in the notation.
The elements of a SEQUENCE are transmitted in textual order. OPTIONAL or DEFAULT elements are preceded by a bit map to identify which elements are present. After sorting the elements of a SET in the "canonical tag order" as defined in X.680 8.6 they are treated as a SEQUENCE regarding OPTIONAL and DEFAULT elements. A SET is transferred in the sorted order.

A second difference is that PER takes full account of the sub-typing information in that the encoded bytes are affected by the constraints. The BER encoded bytes are unaffected by the constraints. PER uses the sub-typing information to for example omit length fields whenever possible.

The run-time functions, sometimes take the constraints into account both for BER and PER. For instance are SIZE constrained strings checked.

There are two variants of PER, aligned and unaligned. In summary, PER results in compact encodings which require much more computation to produce than BER.