hl7v2-server

HTTP/gRPC runtime service for HL7 v2 validation and evidence workflows.

hl7v2-server is the deployable sidecar/edge-guard crate in the hl7v2-rs workspace. It is separate from the hl7v2 library crate because it carries runtime concerns: HTTP/gRPC serving, metrics, authentication, rate limits, config loading, and filesystem output roots.

Sidecar workflows

Useful release-candidate workflows include:

• hl7v2-server --print-config for sanitized effective configuration output.
• GET /ready for startup readiness checks.
• POST /hl7/validate for typed validation reports.
• POST /hl7/validate-redacted for validate-and-redact workflows.
• POST /hl7/bundle for server-created redacted evidence bundles under a configured bundle output root.
• POST /hl7/ack-policy for policy-driven ACK/NAK decisions.
• /metrics for Prometheus metrics.

The server reuses the same evidence contracts as the CLI and hl7v2 library instead of defining a server-only report language.

Usage

Inspect effective configuration without leaking secrets:

hl7v2-server --print-config

Run readiness after startup:

curl http://127.0.0.1:8080/ready

Create a redacted validation report:

curl -X POST http://127.0.0.1:8080/hl7/validate-redacted \
  -H "Content-Type: application/json" \
  -d '{
    "message": "MSH|^~\\&|App||Fac||20250128||ADT^A01|123|P|2.5.1\rPID|1||MRN123||Doe^John\r",
    "profile": "name: ADT_A01\nmessage_type: ADT^A01\nsegments: []\n",
    "redaction_policy": "[[rules]]\npath = \"PID.3\"\naction = \"hash\"\nreason = \"patient identifier\"\n[[rules]]\npath = \"PID.5\"\naction = \"drop\"\nreason = \"patient name\"\n"
  }'

For a full operator walkthrough, see the Deploy Validation Sidecar guide. For source examples, see the examples/ directory in the root of the repository.