drug-api

HealthPorta's drug reference ingestion service for normalized NDC products, packages, labels, and RxNorm-linked lookup support.

What This Repository Covers

This service maintains the drug-side canonical tables used for:

• NDC product and package search
• FDA label retrieval
• RxNorm-linked product and package lookup
• DailyMed/OpenFDA label-derived condition evidence
• downstream crosswalk support for other HealthPorta services

Documentation

Public documentation lives in docs/README.md.

Key pages:

• Data sources
• Import index
• NDC import
• Label import
• Drug indications import
• Drug indications DevOps
• Architecture
• Readability budget
• Test coverage ratchet
• Commit message style

Source Families

This repository ingests from public FDA/OpenFDA source systems:

• OpenFDA (drug/ndc and drug/label payloads)
• FDA download catalog (partition discovery)

Reference terminology/context used by the data model:

• RxNorm
• DailyMed

See the canonical source registry in docs/data-sources.md.

Commercial Usage

For production documentation and managed commercial access, see HealthPorta Docs.

HealthPorta can be used as:

• a hosted API layer for current drug reference, NDC, package, and label data
• an MCP-backed data service for AI agents and internal enterprise workflows
• a downstream integration point for customer applications, clinical products, and analytics systems that need fresh drug data

For AI-agent connectivity, see HealthPorta MCP.

Local Setup

Use .env.example as the configuration reference.

Local prerequisites:

• PostgreSQL
• Redis
• Python virtual environment with dependencies installed

Typical startup flow:

python main.py server start --host 0.0.0.0 --port 8080

Import Quick Start

Run the imports separately.

NDC / product import:

python main.py start ndc
python main.py worker process.NDC --burst

Label import:

python main.py start label
python main.py worker process.Labeling --burst

Drug indication mapping import:

python main.py start drug-indications --test
python main.py start drug-indications

Each import rebuilds staging tables and then swaps them into the live rx_data schema.

Operational Notes

• NDC import publishes product and package together.
• Label import publishes label separately.
• Drug indications import publishes drug_condition_evidence from local label data plus official clinical terminology relationships.
• The project uses dedicated ARQ queues for NDC and label imports by default.
• RxNorm lookup support depends on a successful NDC import because product.rxnorm_ids is populated from OpenFDA payloads during that import.
• DailyMed/NLM-derived outputs must preserve the required NLM attribution statement from the import docs.