simplifier-ig 0.2.0

CLI tool for generating Simplifier.net supported Implementation Guide structures from an IG publisher like input folder

simplifier-ig

CLI tool for generating Simplifier.net Implementation Guide structures from a IG publisher like input folder. The tool supports initializing a new IG input structure, validating the input, and generating the IG output. It also allows saving configuration for repeated use.

Installation

Via pip (recommended for developers)

pip install simplifier-ig

Via standalone binary (no Python required)

Download the pre-built binary for your OS from the GitHub Actions artifacts:

1. Go to the Actions tab → Build Binaries workflow

2. Click on the latest successful run (tagged with a version like v0.1.0)

3. Scroll down to Artifacts and download:

   • simplifier-ig-linux (Linux)
   • simplifier-ig-macos (macOS)
   • simplifier-ig-windows (Windows)

4. Extract the binary and make it executable (Linux/macOS only):

   # Linux/macOS
   chmod +x simplifier-ig
   ./simplifier-ig --help
   
   # Or add to PATH
   sudo mv simplifier-ig /usr/local/bin/
   

   # Windows (PowerShell)
   .\simplifier-ig.exe --help
   
   # Or add to PATH by moving to a directory in your PATH
   

Note: Binaries are currently stored as GitHub Actions artifacts (90-day retention). For permanent releases, they can be manually attached to GitHub Releases or automated via an additional workflow.

From source

git clone https://github.com/ArdonToonstra/simplifier-ig.git
cd simplifier-ig
pip install .

As a GitHub Action

Use this action in your workflows to generate IGs automatically:

# .github/workflows/generate-ig.yml
name: Generate IG
on: [push]

jobs:
  generate:
    runs-on: ubuntu-latest
    steps:
      - uses: actions/checkout@v4

      - name: Generate Implementation Guide
        uses: ArdonToonstra/simplifier-ig@v1
        with:
          input: ./input
          output: ./guides

      - name: Upload IG output
        uses: actions/upload-artifact@v4
        with:
          name: ig-output
          path: ./guides

Action inputs

Input	Description	Default
command	Command to run: generate, validate, ig-resource	generate
input	Path to the IG input folder	./input
output	Path to the output folder	./guides
skip-validation	Skip input validation (true/false)	false
no-ig-resource	Skip ImplementationGuide.json generation (true/false)	false
python-version	Python version to use	3.12

Validate only

- uses: ArdonToonstra/simplifier-ig@v1
  with:
    command: validate
    input: ./input

Usage

Initialize a new IG input folder

simplifier-ig init --path ./my-ig --name "My Implementation Guide"

Option	Description	Default
--path	Where to create the IG input structure	./input
--name	Name for the Implementation Guide	My Implementation Guide
--style	Style folder name	custom
--force	Allow init in a non-empty directory

Validate an IG input folder

simplifier-ig validate --input ./my-ig

Generate an IG

simplifier-ig generate --input ./my-ig --output ./output

By default, an ImplementationGuide.json FHIR R4 resource is generated alongside the IG. Only FHIR R4 (4.0.1) is currently supported for IG resource generation. This requires the following fields in guide.yaml:

id: my.ig.id
status: draft          # draft | active | retired | unknown
fhirVersion: 4.0.1
canonical: https://example.org/fhir

If these fields are missing, the tool warns and skips IG resource generation automatically.

Option	Description	Default
--input	Path to the input folder	saved path or ./input
--output	Path to the output folder	./guides
--skip-validation	Skip input validation
--no-ig-resource	Skip ImplementationGuide.json resource generation

Generate IG resource separately

simplifier-ig ig-resource --path ./output/my-ig --input ./my-ig

Configuration

simplifier-ig config   # Show current configuration
simplifier-ig clear    # Clear saved configuration

Input folder structure

After running init, the input folder will have this structure:

my-ig/
├── guide.yaml                  # IG configuration (title, url-key, menu, etc.)
├── variables.yaml              # Template variables
├── resources/                  # FHIR conformance resources (.json)
├── examples/                   # FHIR example resources (.json)
├── fsh-generated/              # (optional) SUSHI output
│   └── resources/              # FSH-compiled FHIR resources (.json)
├── pages/                      # Markdown documentation pages
├── images/                     # (optional) Images
├── pagetemplates/              # (optional) Reusable page snippets
├── pagetemplates-artifacts/    # Per-resource-type page templates
└── styles/
    └── custom/
        ├── master.html
        ├── settings.style
        └── style.css

FHIR Shorthand (FSH) support

If your project uses FHIR Shorthand, run SUSHI before running this tool:

sushi .
simplifier-ig generate --input ./my-ig --output ./output

Resources from fsh-generated/resources/ are automatically picked up alongside resources/. If the same resource (same resourceType + id) exists in both folders, the one in resources/ takes priority.

Development

git clone https://github.com/ArdonToonstra/simplifier-ig.git
cd simplifier-ig
pip install -e .

You can also run the tool as a module:

python -m simplifier_ig --help

Building binaries

The GitHub Actions workflow builds standalone binaries for Linux, macOS, and Windows using PyInstaller when a version tag is pushed (e.g., v0.1.0).

To build binaries for a release:

git tag v0.1.0
git push origin v0.1.0

This triggers the Build Binaries workflow which creates three artifacts:

• simplifier-ig-linux (Ubuntu)
• simplifier-ig-macos (macOS)
• simplifier-ig-windows (Windows .exe)

Artifacts are available in the Actions tab → Build Binaries workflow for 90 days. To attach binaries permanently to GitHub Releases, either:

• Download artifacts and manually attach to the release, or
• Extend the workflow to automatically create a GitHub Release with attached binaries

Publishing to PyPI

Tag a release to trigger the publish workflow:

git tag v0.1.0
git push origin v0.1.0

This triggers the Publish to PyPI workflow, which uses OpenID Connect (Trusted Publisher) — no API token or secret is required.

Dependencies

• Python >= 3.9
• PyYAML

License

MIT