sqlite2j

sqlite2j is a Java port of SQLite 2.0 focused on behavioral and functional parity, not a redesign of SQL database internals.

• Port SQLite 2.0 core architecture from C to Java.
• Preserve SQLite 2.0 behavior, SQL semantics, transaction model, and storage expectations as closely as practical.
• Keep implementation pure Java (no JNI/native dependency in production code).

sqlite2j follows a layered design:

1. SQL frontend (tokenizer + parser)
2. Compiler/code generator (VDBE-like bytecode)
3. Virtual machine interpreter
4. Storage stack (B-tree + pager + rollback journal + locking)
5. Public interfaces (Java API + CLI)
6. Compatibility/differential testing harness

See full details in docs/PLAN.md.

The implementation roadmap is phased:

• Phase 0: Foundations (tooling, contracts, determinism)
• Phase 1: Minimal vertical slice (CREATE TABLE, INSERT, SELECT *)
• Phase 2+: Core SQL semantics, transactions, indexes, compatibility hardening

Detailed phase planning:

• docs/phase-0-foundations.md
• docs/phase-1-todo.md

• Architecture boundaries: docs/architecture/module-boundaries.md
• Package layout: docs/architecture/package-layout.md
• Coding standards: docs/coding-conventions.md
• Compatibility tracking: docs/sqlite2-compatibility-matrix.md
• Deterministic test rules: docs/testing-determinism.md

Planning and documentation phase in progress. Implementation has not started.