Skip to content

fhir-candle - A FHIR development and test server implemented in C#

License

Notifications You must be signed in to change notification settings

FHIR/fhir-candle

Repository files navigation

fhir-candle

Tests Publish dotnet tool Publish Docker image to ghcr.io Deploy to argo.run

When you need a small FHIR.

fhir-candle is a small in-memory FHIR server that can be used for testing and development. It is NOT intended to be used for production workloads.

The project is intended to serve as a platform for rapid development and testing for FHIR - both for features in the core specification as well as Implementation Guide development.

While there are many existing OSS FHIR servers, somewhere between most and all of them are intended to support production workloads. In my own work on Reference Implementations, I often found it challenging to add the types of features I wanted due to the conflicts that causes. To that end, here are some principles I generally use while developing this project:

  • No database / persisted state
  • Fast startup
  • Dynamically apply changes (e.g., search parameters)
  • House features that would not be appropriate in production
    • E.g., provide feedback on SMART tokens to help developers

FHIR Foundation Project Statement

Contributing

There are many ways to contribute:

This project has adopted the Microsoft Open Source Code of Conduct. For more information see the Code of Conduct FAQ or contact opencode@microsoft.com with any additional questions or comments.

Security

To report a security issue, please use the GitHub Security Advisory "Report a Vulnerability" tab.

For more information, please see the Security Readme.

Documentation

Get Started with .Net

Install .NET 8 or newer and run this command:

dotnet tool install --global fhir-candle

Note that this software is still under heavy development.

Start a FHIR server and open the browser by running:

fhir-candle -o

Get Started with Docker

Install Docker and run these commands:

docker pull ghcr.io/fhir/fhir-candle:latest
docker run -p 8080:5826 ghcr.io/fhir/fhir-candle:latest

This will run the docker image with the default configuration, mapping port 5826 from the container to port 8080 in the host. Once running, you can access http://localhost:8080/ in the browser to access the fhir-candle's UI or access the default endpoints:

Note that additional arguments can be passed directly via the docker run command. For example, to run the server with only an R4 endpoint named 'test':

docker run -p 8080:5826 ghcr.io/fhir/fhir-candle:latest --r4 test

Get Started via cloning this repository

To run the default server from the command line:

dotnet run --project src/fhir-candle/fhir-candle.csproj

To pass arguments when using dotnet run, add an extra --. For example, to see help:

dotnet run --project src/fhir-candle/fhir-candle.csproj -- --help

To build a release version of the project:

dotnet build src/fhir-candle/fhir-candle.csproj -c Release

The output of the release build can be run (from the root directory of the repo)

  • on all platforms:
dotnet ./src/fhir-candle/bin/Release/net8.0/fhir-candle.dll
  • if you built on Windows:
.\src\fhir-candle\bin\Release\net8.0\fhir-candle.exe
  • if you built on Linux or MacOs:
./src/fhir-candle/bin/Release/net8.0/fhir-candle

FHIR Tenants

By default, this software loads three FHIR 'tenants':

  • a FHIR R4 endpoint at /r4,
  • a FHIR R4B endpoint at /r4b, and
  • a FHIR R5 endpoint at /r5.

The tenants can be controlled by command line arguments - note that manually specifying any tenants overrides the default configuration and will only load the ones specified. To load only an R4 endpoint at 'fhir', the arguments would include --r4 fhir. You can specify multiple tenants for the same version, for example --r5 fhir --r5 also-fhir will create two endpoints.

Loading Initial Data

The server will load initial data specified by the --fhir-source argument. If the path specified is a relative path, the software will look for the directory starting at the current running path.

If the system is loading multiple tenants, it will check the path for additional directories based on the tenant names. For example, a path like fhirData passed into the default server will look for fhirData/r4, fhirData/r4b, and fhirData/r5. If tenant directories are not found, all tenants will try to load resources from the specified path.

Subscriptions Reference Implementation

This project also contains the reference stack for FHIR Subscriptions. To use the default landing page of the subscriptions RI, the following command can be used:

fhir-candle --reference-implementation subscriptions --load-package hl7.fhir.uv.subscriptions-backport#1.1.0 --load-examples false --protect-source true -m 1000

Using OpenTelemetry

OpenTelemetry instrumentation can be enabled via either the --otel-otlp-endpoint argument or the OTEL_EXPORTER_OTLP_ENDPOINT environment variable. For example, to send traces to a Jaeger instance running on localhost:4317: fhir-candle -o --otel-otlp-endpoint http://localhost:4317.

For local testing, you can run a Jaeger instance in Docker with the following command:

docker run --rm --name jaeger -p 4317:4317 -p 4318:4318 -p 5778:5778 -p 16686:16686 -p 14250:14250 jaegertracing/all-in-one:latest

To-Do

Note: items are unsorted within their priorities

High priority

  • Feature/module definitions for selective loading Build interfaces for Hosted Services, etc. Add module tag to Operation, etc. Conditional loading based on discovery within types
  • Persistent 'unsubscribe' list
  • Finish search evaluators (remaining modifier combinations)
  • Save/restore points
  • Versioned Resource support
  • Resource display / edit in UI
  • Resource editor design improvements
  • Add loading packages/profiles to CapabilityStatement

Mid Priority

  • SMART support
  • Transaction support
  • OpenAPI generation
  • Compartments
  • Contained resources
  • Subscription websocket support

The long tail

  • Non-terminology validation
  • Link to terminology server for full validation
  • _filter support
  • Runtime named queries
  • GraphQL support

More Information

FHIR® is the registered trademark of HL7 and is used with the permission of HL7.