Lineage Graph & Impact Analysis

Modifying a table or column in enterprise databases is often one of the riskiest tasks in software engineering. A seemingly harmless rename can invalidate views, break procedures across external schemas, and trigger silent failures.

LEAI addresses this challenge by compiling a cross-schema dependency graph and providing automated impact analysis.

🔍 How Lineage Tracing Works

The leai trace <OBJECT> command inspects both directions:

graph LR
    subgraph Upstream [What feeds into this object?]
        A[BASE_TABLE] --> B[CONSOLIDATED_VIEW]
    end

    subgraph Target [Target Object]
        B --> C[CALCULATION_PROCEDURE]
    end

    subgraph Downstream [What breaks if it changes?]
        C --> D[BILLING_PACKAGE]
        C --> E[AUDIT_TRIGGER]
        D --> F[NIGHTLY_JOB]
    end

    style C fill:#f96,stroke:#333,stroke-width:2px

Upstream (Ancestors): Which tables, views, and synonyms supply data or logic to this entity.
Downstream (Descendants): Which views, packages, triggers, and procedures will fail if this entity's signature or structure changes.

🚦 Automated Risk Scoring

For every traced object, LEAI calculates an automated impact severity score based on the count and criticality of downstream dependents:

Risk Level	Visual Indicator	Typical Criteria
`LOW`	🟢 Green	Leaf entity with zero downstream dependents (or a single simple view).
`MEDIUM`	🟡 Yellow	Moderate dependents (2 to 4), without mission-critical triggers or monolithic packages.
`HIGH`	🟠 Orange	Multiple materialized views, core packages, or cross-schema dependencies.
`CRITICAL`	🔴 Red	High-traffic hub object with dozens of downstream consumers, cascading triggers, or vital financial packages.

💻 CLI Usage Example

leai trace CONTRACTS_TB --depth 3

LEAI outputs: 1. A formatted summary table with calculated risk rating and total affected entities. 2. The hierarchical list of upstream and downstream dependencies. 3. The corresponding Mermaid.js diagram ready to embed into documents or PR reviews.