LutaML for UML plugin

Purpose

Many standards organizations, such as OGC, ISO and IEC, rely on UML to describe and standardize conceptual or implementation models.

Standards authors who wish to describe a UML model in a standard document may take advantage of the Metanorma “model-based authoring” approach, where the author can programmatically generate content by having Metanorma navigate through the actual UML model.

The LutaML for UML plugin supports navigating through UML models stored in XMI files (version 2.4+).

Note
Currently, only XMI files exported via Enterprise Architect (versions 15.0+) from SparxSystems are officially supported. In particular, Enterprise Architect supports the export of a “Canonical UML XMI” model file, where the LutaML for UML plugin takes as input.

Basic UML concepts

An UML XMI model file can contain a combination of the following objects:

  • Package

  • Class / Enumerations / Classifiers

  • Attributes

  • Associations

  • Instances

Describing UML packages

General

A common use case in standards authoring is to iterate through a UML package and have the document perform the following:

  • display diagrams of the package

  • describe classes and associations in the package

    • for each class, describe its attributes

  • describe enumerations in the package

The following syntax allows for navigating within one or more UML packages in an XMI model file and generating text from them.

lutaml_uml_datamodel_description

Syntax

The lutaml_uml_datamodel_description command renders UML packages and its dependent objects from a XMI model file.

[lutaml_uml_datamodel_description, path/to/file.xmi, path/to/config.yml]
--
{block}
--

The lutaml_uml_datamodel_description command takes two arguments and a block:

  • path/to/file.xmi: location of the XMI file.

  • path/to/config.yml: (optional) configuration that specifies what packages and classes to iterate through.

  • {block}: (optional) provides the ability to insert specified content before or after designated packages.

Inserting text via the supplied block

Within the supplied block to the command, the following roles are available:

  • [.before]: macro to add additional text before the rendered output. There are two variants.

    • [.before]. When provided without the package attribute, specifies text to be inserted before the description of the first package.

      Note
      This role is only to be entered once per block, additional occurrences of the role will overwrite the previous content.
    • [.before, package="{package-name}"]. When provided with the package attribute to specify locality of the block, specifies text to be inserted before the description of the chosen package.

      Note
      This role is only to be entered once per package, additional occurrences of the role will overwrite the previous content.
  • [.after]: macro to add additional text after the rendered output. There are two variants.

    • [.after]. When provided without the package attribute, specifies text to be inserted after the description of the last package.

      Note
      This role is only to be entered once per block, additional occurrences of the role will overwrite the previous content.
    • [.after, package="{package-name}"]. When provided with the package attribute to specify locality of the block, specifies text to be inserted after the description of the chosen package.

      Note
      This role is only to be entered once per package, additional occurrences of the role will overwrite the previous content.

Selecting and ordering packages

The lutaml_uml_datamodel_description command supports selection and ordering of packages from the XMI file by providing a YAML configuration file.

The YAML configuration file is optional to the command.

The format of the configuration file is as follows:

---
packages:
  # includes these packages
  - "Package *"
  - two*
  - three
  # skips these packages
  - skip: four

Where:

  • packages: required, root element with an array of strings or objects.

    • string: treated as an regular expression pattern to be matched against package names for inclusion.

    • object: only the skip key is supported in an object, where the packages with names matching the value of the skip key are excluded.

  • the order of the array determines the order of packages rendered in text.

In this example,

  • "Package ": performs pattern matching, equal to the following regular expression: /^Package .$/

  • skip: four: excludes packages with a name that matches this criteria.

  • The command will read the supplied YAML file and arrange packages according to the order supplied in the config file.

Illustration

For example, given an XMI file named example.xmi which contains 2 UML packages, 'Package-One' and 'Package-Two', the following Metanorma AsciiDoc syntax can be used.

[lutaml_uml_datamodel_description, path/to/example.xmi]
--
[.before]
---
This text will be inserted before the description of all packages.
---

[.before, package="Package-One"]
---
This text will be inserted before the description of the package Package-One.
---

[.after, package="Package-One"]
---
This text will be inserted after the description of the package Package-One.
---

[.after, package="Package-Two"]
---
This text will be inserted after the description of the package Package-Two.
---

[.after]
---
This text will be inserted after the description of all packages.
---
--

Will translate into this output:

This text will be inserted before the description of all packages.

This text will be inserted before the description of the package Package-One.

=== Package-One

==== Package-One

==== Requirements

==== Class Definitions

.Classes used in Package-One
[cols="2a,6a",options="header"]
|===
|Class |Description


|<<AbstractAtomicTimeseries-section,AbstractAtomicTimeseries>>
«some-stereotype»
|


|<<AbstractTimeseries-section,AbstractTimeseries>>
«some-stereotype»
|

....
|===



.Data Types used in Package-One
[cols="2,6",options="header"]
|===
|Name |Description


|<<ADEOfAbstractAtomicTimeseries-section,ADEOfAbstractAtomicTimeseries>>
|


|<<ADEOfAbstractTimeseries-section,ADEOfAbstractTimeseries>>
|

....
|===



.Enumerated Classes used in Package-One
[cols="2a,6a",options="header"]
|===
|Name |Description

|<<TimeseriesTypeValue-section,TimeseriesTypeValue>>
|

|===


==== Additional Information

...


This text will be inserted after the description of the package Package-One.


=== Package-Two

==== Package-Two

==== Requirements

==== Class Definitions

==== Additional Information

This text will be inserted after the description of the package Package-Two.

This text will be inserted after the description of all packages.