How cdcgov/ephtracking works
Project Type and Purpose
• This is an open-source “method and code” repository rather than a single deployable application.
• It houses:
• Data-validation protocols (R, SAS scripts)
• Dataset-creation guides (XSD schemas, PDFs)
• Code examples (Python API calls)
• Tracking-method documentation
• Its goal: define and share CDC’s end-to-end data-ingestion, validation, standardization, and publishing pipeline for environmental public-health measures.
File Structure
• Top-level folders map to pipeline stages and supporting materials:
• Data Validation
• Measure and Dataset Creation
• Tracking Data Methods
• Coding Examples (client/API examples)
• Data Explorer (map layers, points of interest)
• Configuration files (e.g., .env-template) and scripts (Python, SAS, R) live alongside protocol PDFs and XSD schema files.
• No single “frontend” or “backend” folder; this is a methods repo, not the actual web-app source.
Extract Architectural Insights from README
• The Tracking Network consists of:
1. Data ingestion from national/state/local sources
2. Validation & standardization
3. Measure calculation & XML generation
4. Data publishing to a repository & API
5. Web-based visualization (Data Explorer, dashboards)
• The repo provides the *methods* and *tools* used in steps (2) and (3); front-end and API code live elsewhere.
Main Components & Relationships
a. Data Sources
• National, state, local public-health databases, environmental sensors
b. ETL & Validation Layer
• Data Validation Protocol (PDF)
• SAS & R scripts (folder: Data Validation/Sample Validation Scripts)
c. Schema & Measure Creation
• XSD schema files (Measure and Dataset Creation/Schema Files)
• Style guides & calculation methods (PDFs)
• XML Generator Tool (external repo)
d. Standardized Data Repository
• Centralized data store (assumed database or file server)
e. Public API Layer
• API that serves standardized measures (external repo)
f. Visualization Layer
• National Data Explorer & Dashboards (external system)
g. Developer Toolkit & Documentation
• Coding Examples (Python API usage, .env templates)
• Tracking Data Methods scripts & docs
• Subcounty project resources
– Data flow: Sources → Validation → Schema mapping & measure calc → Repository → API → Explorer/Dashboards
– Supporting docs and example code sit alongside each pipeline stage.
Architectural Patterns & Design Principles
• Modular ETL pipeline: each stage encapsulated (validation, transformation, schema enforcement)
• Separation of concerns: methods repo (this project) vs. runtime systems (API, front end)
• Schema-driven design: XSD files drive data structure and enforce standards
• Extensibility: open-source scripts and docs allow external partners to plug in