Comparison with HL7 Validator

The GoFHIR Validator aims to produce validation results equivalent to the HL7 FHIR Validator while taking advantage of Go’s performance characteristics. This page compares the two tools across features, performance, and CLI usage.

Feature Comparison

CLI Flag Equivalence

Both tools follow similar command-line conventions. The table below maps the most common flags:

Examples Side by Side

# Basic validation
gofhir-validator patient.json

# Validate against US Core
gofhir-validator -ig http://hl7.org/fhir/us/core/StructureDefinition/us-core-patient patient.json

# Disable terminology checks
gofhir-validator -tx n/a patient.json

# JSON output
gofhir-validator -output json patient.json

# Basic validation
java -jar validator_cli.jar patient.json -version 4.0.1

# Validate against US Core
java -jar validator_cli.jar patient.json -ig hl7.fhir.us.core#6.1.0

# Disable terminology checks
java -jar validator_cli.jar patient.json -tx n/a

# JSON output
java -jar validator_cli.jar patient.json -output json

When to Choose GoFHIR Validator

The GoFHIR Validator is a strong choice when:

• Startup time matters – In CI/CD pipelines, serverless functions, or CLI workflows where the Java validator’s 10-15 second startup is a bottleneck.
• Memory is constrained – In containerized environments or edge deployments where ~200 MB is preferable to ~600 MB+.
• Go ecosystem integration – When your application is written in Go and you want to embed validation as a library without JVM dependencies.
• Concurrent batch validation – GoFHIR leverages Go’s concurrency model for efficient parallel validation of large resource sets.
• Simple deployment – A single static binary with no runtime dependencies, compared to requiring a JVM installation.

The HL7 Validator remains the better choice when:

• Full FHIR version coverage is needed (R2, R3, R4, R4B, R5).
• Remote terminology server integration (-tx) is required for external code system validation.
• Reference conformance is critical – the HL7 Validator is the official reference implementation.

Performance Benchmarks

Typical performance on a modern machine (Apple M-series or equivalent x86-64):

These numbers are approximate and vary by machine, resource complexity, and profile depth. Run your own benchmarks with representative data for accurate comparisons.