19972012 Ericsson AB. All Rights Reserved. The contents of this file are subject to the Erlang Public License, Version 1.1, (the "License"); you may not use this file except in compliance with the License. You should have received a copy of the Erlang Public License along with this software. If not, it can be retrieved online at http://www.erlang.org/. Software distributed under the License is distributed on an "AS IS" basis, WITHOUT WARRANTY OF ANY KIND, either express or implied. See the License for the specific language governing rights and limitations under the License. asn1ct Kenneth Lundin Kenneth Lundin 1 Kenneth Lundin 97-10-14 A asn1.sgml
asn1ct ASN.1 compiler and compile-time support functions

The ASN.1 compiler takes an ASN.1 module as input and generates a corresponding Erlang module which can encode and decode the data-types specified. Alternatively the compiler takes a specification module (se below) specifying all input modules and generates one module with encode/decode functions. There are also some generic functions which can be used in during development of applications which handles ASN.1 data (encoded as BER or PER).

 compile(Asn1module) -> ok | {error, Reason} compile(Asn1module, Options) -> ok | {error, Reason} Compile an ASN.1 module and generate encode/decode functions according to the encoding rules BER or PER. Asn1module = atom() | string() Options = [Option| OldOption] Option = ber_bin | per_bin | uper_bin | der | compact_bit_string | noobj | {n2n, EnumTypeName} |{outdir, Dir} | {i, IncludeDir} | optimize | nif | asn1config | undec_rest | {inline, OutputName} | inline | {macro_name_prefix, Prefix} | {record_name_prefix, Prefix} | verbose | warnings_as_errors OldOption = ber | per Reason = term() Prefix = string()

Compiles the ASN.1 module Asn1module and generates an Erlang module Asn1module.erl with encode and decode functions for the types defined in Asn1module. For each ASN.1 value defined in the module an Erlang function which returns the value in Erlang representation is generated.

If Asn1module is a filename without extension first ".asn1" is assumed, then ".asn" and finally ".py" (to be compatible with the old ASN.1 compiler). Of course Asn1module can be a full pathname (relative or absolute) including filename with (or without) extension.

If one wishes to compile a set of Asn1 modules into one Erlang file with encode/decode functions one has to list all involved files in a configuration file. This configuration file must have a double extension ".set.asn", (".asn" can alternatively be ".asn1" or ".py"). The input files' names must be listed, within quotation marks (""), one at each row in the file. If the input files are File1.asn, File2.asn and File3.asn the configuration file shall look like:

File1.asn
File2.asn
File3.asn

The output files will in this case get their names from the configuration file. If the configuration file has the name SetOfFiles.set.asn the name of the output files will be SetOfFiles.hrl, SetOfFiles.erl and SetOfFiles.asn1db.

Sometimes in a system of ASN.1 modules there are different default tag modes, e.g. AUTOMATIC, IMPLICIT or EXPLICIT. The multi file compilation resolves the default tagging as if the modules were compiled separately.

Another unwanted effect that may occur in multi file compilation is name collisions. The compiler solves this problem in two ways: If the definitions are identical then the output module keeps only one definition with the original name. But if definitions only have same name and differs in the definition, then they will be renamed. The new names will be the definition name and the original module name concatenated.

If any name collision have occurred the compiler reports a "NOTICE: ..." message that tells if a definition was renamed, and the new name that must be used to encode/decode data.

Options is a list with options specific for the asn1 compiler and options that are applied to the Erlang compiler. The latter are those that not is recognized as asn1 specific. For preferred option use see Preferred Option Use section in users guide. Available options are:

ber | ber_bin | per | per_bin | uper_bin

The encoding rule to be used. The supported encoding rules are BER (Basic Encoding Rules), PER aligned (Packed Encoding Rules) and PER unaligned. If the encoding rule option is omitted ber is the default. The per_bin option means the aligned variant. To use the unaligned variant the uper_bin option has to be used.

The generated Erlang module always gets the same name as the ASN.1 module and as a consequence of this only one encoding rule per ASN.1 module can be used at runtime.

The ber_bin and per_bin options are equivalent with the OldOptions ber and per with the difference that the generated encoding/decoding functions take advantage of the bit syntax, which in most cases increases the performance considerably. The result from encoding is a binary or an iolist.

der

By this option the Distinguished Encoding Rules (DER) is chosen. DER is regarded as a specialized variant of the BER encoding rule, therefore the der option only makes sense together with the ber or ber_bin option. This option sometimes adds sorting and value checks when encoding, which implies a slower encoding. The decoding routines are the same as for ber.

compact_bit_string

Makes it possible to use a compact notation for values of the BIT STRING type in Erlang. The notation: 

BitString = {Unused, Binary},
Unused = integer(),
Binary = binary()

Unused must be a number in the range 0 to 7. It tells how many bits in the least significant byte in Binary that is unused. For details see BIT STRING type section in users guide .

{n2n, EnumTypeName}

Tells the compiler to generate functions for conversion between names (as atoms) and numbers and vice versa for the EnumTypeName specified. There can be multiple occurrences of this option in order to specify several type names. The type names must be declared as ENUMERATIONS in the ASN.1 spec. If the EnumTypeName does not exist in the ASN.1 spec the compilation will stop with an error code. The generated conversion functions are named name2num_EnumTypeName/1 and num2name_EnumTypeName/1.

noobj

Do not compile (i.e do not produce object code) the generated .erl file. If this option is omitted the generated Erlang module will be compiled.

 {i, IncludeDir}

Adds IncludeDir to the search-path for .asn1db and asn1 source files. The compiler tries to open a .asn1db file when a module imports definitions from another ASN.1 module. If no .asn1db file is found the asn1 source file is parsed. Several {i, IncludeDir} can be given.

{outdir, Dir}

Specifies the directory Dir where all generated files shall be placed. If omitted the files are placed in the current directory.

 optimize

This option is only valid together with one of the per_bin or ber_bin option. It gives time optimized code generated and it uses another runtime module and in the per_bin case a nif. The result in the per_bin case from an encode when compiled with this option will be a binary.

 driver

As of R15B this means the same as the nif option. Kept for backwards compatability reasons.

 nif

Option valid together with ber_bin and optimize options. It enables the use of several nifs that gives faster encode and decode. Nifs are only enabled by the explicit use of the option nif

 asn1config

When one of the specialized decodes, exclusive or selective decode, is wanted one has to give instructions in a configuration file. The option asn1config enables specialized decodes and takes the configuration file, which has the same name as the ASN.1 spec but with extension .asn1config, in concern.

The instructions for exclusive decode must follow the instruction and grammar in the User's Guide.

You can also find the instructions for selective decode in the User's Guide.

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.

 {inline, OutputName}

Compiling with this option gives one output module containing all asn1 run-time functionality. The asn1 specs are provided in a target module Module.set.asn as described above. The name of the resulting module containing generated encode/decode functions and in-lined run-time functions will be OutputName.erl. The merging/in-lining of code is done by the igor module of syntax_tools. By default the functions generated from the first asn1 spec in the .set.asn are exported, unless a {export, [atom()]} or {export_all, true} option are provided. The list of atoms are names of chosen asn1 specs from the .set.asn file.

When used together with nif for ber_bin, the asn1 nifs will be used if the asn1rt_nif module is available. If it is not available, a slower erlang fallback will be used.

 inline

It is also possible to use the sole argument inline. It is as {inline, OutputName}, but the output file gets the default name of the source .set.asn file.

 {macro_name_prefix, Prefix}

All macro names generated by the compiler are prefixed with Prefix. This is useful when multiple protocols that contains macros with identical names are included in a single module.

 {record_name_prefix, Prefix}

All record names generated by the compiler are prefixed with Prefix. This is useful when multiple protocols that contains records with identical names are included in a single module.

 verbose

Causes more verbose information from the compiler describing what it is doing.

 warnings_as_errors

Causes warnings to be treated as errors.

Any additional option that is applied will be passed to the final step when the generated .erl file is compiled.

The compiler generates the following files:

Asn1module.hrl (if any SET or SEQUENCE is defined)

Asn1module.erl the Erlang module with encode, decode and value functions.

Asn1module.asn1db intermediate format used by the compiler when modules IMPORTS definitions from each other.

 encode(Module, Type, Value)-> {ok, Bytes} | {error, Reason} Encode an ASN.1 value. Module = Type = atom() Value = term() Bytes = [Int] when integer(Int), Int >= 0, Int =< 255 Reason = term()

Encodes Value of Type defined in the ASN.1 module Module. Returns a list of bytes if successful. To get as fast execution as possible the encode function only performs rudimentary tests that the input Value is a correct instance of Type. The length of strings is for example not always checked. Returns {ok, Bytes} if successful or {error, Reason} if an error occurred.

decode(Module, Type, Bytes) -> {ok, Value} | {error, Reason} Decode from Bytes into an ASN.1 value. Module = Type = atom() Value = Reason = term() Bytes = [Int] when integer(Int), Int >= 0, Int =< 255

Decodes Type from Module from the list of bytes Bytes. Returns {ok, Value} if successful.

 value(Module, Type) -> {ok, Value} | {error, Reason} Create an ASN.1 value for test purposes. Module = Type = atom() Value = term() Reason = term()

Returns an Erlang term which is an example of a valid Erlang representation of a value of the ASN.1 type Type. The value is a random value and subsequent calls to this function will for most types return different values.

 test(Module) -> ok | {error, Reason} test(Module, Type | Options) -> ok | {error, Reason} test(Module, Type, Value | Options) -> ok | {error, Reason} Perform a test of encode and decode for types in an ASN.1 module. Module = Type = atom() Value = term() Options = [{i, IncludeDir}] Reason = term()

Performs a test of encode and decode of types in Module. The generated functions are called by this function. This function is useful during test to secure that the generated encode and decode functions and the general runtime support work as expected.

test/1 iterates over all types in Module.

test/2 tests type Type with a random value.

test/3 tests type Type with Value.

Schematically the following happens for each type in the module: {ok, Value} = asn1ct:value(Module, Type), {ok, Bytes} = asn1ct:encode(Module, Type, Value), {ok, Value} = asn1ct:decode(Module, Type, Bytes).

The test functions utilizes the *.asn1db files for all included modules. If they are located in a different directory than the current working directory, use the include option to add paths. This is only needed when automatically generating values. For static values using Value no options are needed.