FEAT-012: Schema-Aware Validation

[ajitpratap0]

Summary

Add schema-aware validation to catch semantic errors (missing tables, wrong column types, etc.).

Problem

Current validation is syntax-only. Can't detect:

• References to non-existent tables
• References to non-existent columns
• Type mismatches in comparisons
• Foreign key violations

Action Items

1. Add schema definition support:

   • Load from DDL files
   • Load from database connection
   • Load from JSON schema
   • Manual schema definition

2. Validate queries against schema:

   • Table existence: All referenced tables exist
   • Column existence: All referenced columns exist in tables
   • Data type compatibility: Comparisons use compatible types
   • Foreign key relationships: JOINs respect FK constraints
   • Aggregate compatibility: GROUP BY includes all non-aggregated columns
   • Function signatures: Correct argument types

3. CLI integration:

   gosqlx validate --schema schema.sql query.sql
   gosqlx validate --db "postgres://..." query.sql

4. API for programmatic use

Acceptance Criteria

• Schema loading from multiple sources (DDL, database, JSON)
• Semantic validation implemented
• Helpful error messages with suggestions
• Examples for common use cases
• Performance: < 2x syntax-only validation

Technical Details

Priority: Low
Effort: Large (120h)
Phase: Phase 7 - Advanced Features
Dependencies: FEAT-001 (SQL-99 compliance)

Example

-- Schema
CREATE TABLE users (id INT, name VARCHAR);

-- Query
SELECT email FROM users;

-- Error
Column 'email' does not exist in table 'users'
Available columns: id, name
Did you mean 'name'?