System Requirements

Objective

Modular Python pipeline identifies 501(c)(3) entities by integrating IRS Business Master File (BMF) data with Form 990 XML filings. It ranks 5 types of non-profits (Higher Education, Lower Education, Community Foundations, Professional Societies, and Research) based on financial health from Form 990. Users customize the parameters.

Functional Requirements

• Data Acquisition: Must download IRS BMF files and TEOS Form 990 XML ZIPs using provided URL lists.
• Categorization: Must group organizations into five target categories: EduHigh, EduLow, Research, Foundation, and ProfSoc.
• XML Parsing: Must extract at least 25 specific financial and metadata fields from Form 990 schemas.
• Scoring Engine: Must compute a composite score based on:
  • Organization Age.
  • 5-year trends (increases) in grants and endowment balances.
  • 2-year gaps in net assets and grants paid.
  • Peer-relative performance (top-half ranking) across investment income and revenue.

Technical Requirements

• Python 3.x: Primary language.
• SQLite3: For local data persistence.
• 7-Zip (7z): Must be in the system PATH to handle IRS ZIP files using Deflate64 compression.
• Dependencies: requests, lxml, and pathlib.
• No Hardcoding: All URLs and task toggles must reside in the input/ directory.