Engineer(s): Slav Babanin, Valentin Kavalenka 2026-02-12  - 🟢 On Track What was accomplished since the last update: Created performance test infrastructure to compare 2VL vs 3VL execution. Started testing function handling under ternary logic; built a POC for strLength. What’s the focus over the next two weeks: Continue performance testing and evaluate performance implications for functions. Complete UPDATE ... SET translation to test NULL behavior and its implications. Any risks/blockers/impediments: None. Engineer(s): Slav Babanin, Valentin Kavalenka 2026-01-29 What was accomplished since the last update: Completed join testing to validate NULL / 3VL behavior. Implemented partial POC for update queries (UPDATE ... SET). What’s the focus over the next two weeks: Complete UPDATE ... SET POC. Move to schema-based AST generation to avoid pervasive NULL checks. Continue evaluating feasibility of De Morgan rewrites and treating UNKNOWN as FALSE in predicate contexts; validate fallback to ad-hoc $expr where rewrites don’t apply. Continue performance testing and analyze index usage differences between 3VL MQL and 2VL MQL translations. Any risks/blockers/impediments: None. Engineer(s): Slav Babanin, Valentin Kavalenka 2026-01-16  What was accomplished since the last update: Created compatibility tests against Postgres and MongoDB to validate equivalent behavior for the currently available POCs. Built a POC for joins to validate NULL / 3VL behavior. Performed a small initial performance analysis. What’s the focus over the next two weeks: Continue evaluating feasibility of De Morgan rewrites and treating UNKNOWN as FALSE in predicate contexts; validate fallback to ad-hoc $expr where rewrites don’t apply. Continue performance testing and analyze index usage differences between 3VL MQL and 2VL MQL translations. Extend coverage beyond current POCs (joins / GROUP BY / aggregates) and iterate based on compatibility/perf findings. Any risks/blockers/impediments: None. Engineer(s): Slav Babanin, Valentin Kavalenka 2025-12-18: What was accomplished since the last update: HIBERNATE-140 implemented and moved to review to enable testing Hibernate across multiple Java versions. Started designing an approach for emulating ternary logic in MQL to preserve query semantics/compatibility; explored limitations, constraints, and options. Investigated existing products (e.g., MongoSQL) and how their AST translation and “algebralizer” work. Built a simple POC 3VL translation using $expr for simple queries. What’s the focus over the next two weeks: Explore ternary logic handling for GROUP BY, joins, aggregate functions. Create compatibility tests against Postgres and MongoDB to validate equivalent behavior. Evaluate feasibility of De Morgan rewrites and treating UNKNOWN as FALSE in predicate contexts; test fallback to ad-hoc $expr where rewrites don’t apply. Test performance and analyze index usage differences between 3VL MQL and 2VL MQL translations. Any risks/blockers/impediments: None.