Documentation voc2skosmos

This repository contains two reusable actions that can be used to build a skos-compatible vocabulary. Either you can build from a .csv or you can start out from a .yaml file with LInkML schema-compatible content.

Publish Vocabulary Workflow from CSV

This GitHub workflow automates the process of converting vocabulary data from a CSV source into SKOS (Simple Knowledge Organization System) format using YARRRML → RML → RDF → SKOS transformations.

The generated SKOS vocabulary can be uploaded as an artifact and optionally published when triggered on a release event.

---

Workflow Overview

This workflow is designed to be reusable via workflow_call so it can be invoked from other workflows in the repository.

Key features

• Cancel any previously running workflow instances
• Convert YARRRML mappings into RML
• Use RML Mapper to generate RDF from CSV data
• Run skosify to produce SKOS vocabularies
• Optionally publish SKOS files on release events

---

Inputs

Name	Required	Type	Default	Description
ref	✅	string	—	Git reference (branch, tag, or commit SHA) to use.
version	❌	string	—	Version of the vocabulary being published.
namespace_label	✅	string	—	Label for the SKOS namespace.
namespace	✅	string	—	Namespace URI for the SKOS vocabulary.
mappings_file	✅	string	—	Path to the YARRRML mapping file.

---

Secrets

Name	Required	Description
token	✅	GitHub token for repository checkout.

---

Usage Example

To call this workflow from another workflow:

name: CSV to SKOS Conversion

on:
  workflow_dispatch:

jobs:
  csv-to-skos:
    uses: ./.github/workflows/csv-to-skos.yml
    with:
      ref: main
      namespace_label: "My Vocabulary"
      namespace: "https://example.org/vocab/"
      mappings_file: "mappings/mappings.yml"
    secrets:
      token: ${{ secrets.GITHUB_TOKEN }}

Publish Vocabulary Workflow from YML

This GitHub workflow automates the process of converting vocabulary data from a YAML source into SKOS (Simple Knowledge Organization System) format using LinkML → RDF → SKOS transformations.

The resulting SKOS vocabulary file can be uploaded as an artifact and optionally published on release events.

---

Workflow Overview

This workflow is designed to be reusable via workflow_call, making it easy to integrate into other workflows in the repository.

Key features

• Supports both local files and artifacts as data sources
• Processes LinkML schemas (from local paths or URLs)
• Adds SKOS and language annotations
• Converts YAML input into RDF, then to SKOS
• Validates the generated SKOS file
• Optionally publishes SKOS files on release events

---

Inputs

Name	Required	Type	Default	Description
ref	✅	string	—	Git reference (branch, tag, or commit SHA) to use.
version	❌	string	—	Version of the vocabulary being published.
namespace	✅	string	—	Base namespace for RDF resources.
voc_uri	✅	string	—	Namespace URI for the SKOS vocabulary.
entity_name	✅	string	—	Name of the entity class in the LinkML schema.
linkml_schema	✅	string	—	Path or URL to the LinkML schema file.
data	❌	string	—	Path to the YAML data file containing vocabulary terms.
artefact_name	❌	string	—	Name of an artifact containing the data file (alternative to data).

---

Secrets

Name	Required	Description
token	✅	GitHub token for repository checkout.

---

Usage Example

To call this workflow from another workflow:

name: YML to SKOS Conversion

on:
  workflow_dispatch:

jobs:
  yml-to-skos:
    uses: ./.github/workflows/yml-to-skos.yml
    with:
      ref: main
      namespace: "https://example.org/base/"
      voc_uri: "https://example.org/vocab/"
      entity_name: "MyEntity"
      linkml_schema: "schema/my_schema.yml"
      data: "data/vocabulary.yml"
    secrets:
      token: ${{ secrets.GITHUB_TOKEN }}