yangson manual page

Synopsis

yangson [<options>] [<operation>] <in_file>

yangson -h

Description

yangson is a tool for performing selected high-level operations with a YANG data model and JSON-encoded instance objects from the command line.

<in_file> is the name of a file containing either

  • a data model specification in JSON representation conforming to the original YANG library format of [RFC7895], or

  • (with the --pickled option) a serialized data model object generated using the Python pickle module. See also --dump option below.

If no operation is specified, the program just parses the data model and exits.

Operations

-d, --digest

Print the schema digest of the data model in JSON format. See also DataModel.schema_digest().

-D <out_file>, --dump <out_file>

Dump the serialized (pickled) data model object to <out_file>.

-h, --help

Show an overview of the command syntax and exit.

-i, --id

Print the unique module set identifier that can be used, for example, as the value of the module-set-id leaf in YANG library data. See also DataModel.module_set_id().

-t, --tree

Print the schema tree of the complete data model as ASCII art. See also DataModel.ascii_tree().

-v <instance>, --validate <instance>

Validate an instance object against the data model. The instance argument is the name of a file containing an instance object in JSON representation.

Validation can be controlled by means of --scope and --ctype options.

See also InstanceNode.validate().

Options

-c <content_type>, --ctype <content_type>

This option specifies the content type of the instance object, and is only relevant when used with the --validate operation. The content_type arguments can be one of config (configuration data), nonconfig (non-configuration data) and all (all data, which is the default). See also InstanceNode.validate().

-n, --no_types

This option is used to suppress data type information in ASCII tree output. It is relevant only for the --tree operation.

-p <module_path>, --path <module_path>

This option specifies a list of directories to search for YANG modules. It is only applicable if the --pickled option is not used. The <module_path> argument is a colon-separated list of directory names. By default, the value of the YANG_MODPATH environment variable is used if it is set, otherwise the module path contains only the current directory.

All YANG modules specified in YANG library need to be located in one of these directories, and their file names have to be in the following form:

module-or-submodule-name ['@' revision-date] '.yang'
-P, --pickled

This option indicates that <in_file> contents is to be interpreted as a serialized (pickled) data model object. See also --dump option.

-s <validation_scope>, --scope <validation_scope>

This option specifies validation scope, and is only relevant when used with the --validate operation. The choices for the validation_scope argument are as follows:

  • syntax – schema constraints (including when and if-feature conditions) and data types;

  • semantics – must constraints, uniqueness of list keys, unique constraints in lists, integrity of leafref and instance-identifier references;

  • all – all of the above.

The default value is all. See also InstanceNode.validate().

-S <subschema>, --subschema <subschema>

Parse and validate the instance object against a subschema (RPC or notification). The subschema argument is a prefixed name of the selected RPC or notification.

In this case, the instance object has to be input and/or output payload of the selected RPC enclosed in <modulename>:input or <modulename>:output, or the notification payload.

Note that validation may fail if the RPC input/output or notification payload contains XPath of leafref references to configuration or state data outside the selected RPC or notification.

Environment Variables

YANG_MODPATH

A colon-separated list of directories that is used as the default module path, see the --path option.

Exit Status

Value

Meaning

0

No errors

1

Problem with reading or decoding JSON files

2

YANG library or data model problem

3

Validation of the instance object failed

Author

Ladislav Lhotka <ladislav@lhotka.name>