7.4.0
📦 prismaView on GitHub →
✨ 4 features🐛 6 fixes🔧 4 symbols
Summary
Version 7.4.0 introduces a significant query compilation caching layer to improve performance under concurrency and adds support for Partial Indexes across several databases. Numerous community contributions fixed various bugs related to migrations, BigInt handling, and database connection stability.
Migration Steps
- To use Partial Indexes, enable the `partialIndexes` preview feature in your `generator client` block in your schema file.
✨ New Features
- Introduced a new caching layer in Prisma Client to store compiled query plans based on normalized query shapes, reducing event loop blocking during high concurrency.
- Added support for Partial Indexes (Filtered Indexes) for PostgreSQL, SQLite, SQL Server, and CockroachDB behind the `partialIndexes` preview feature.
- Partial Indexes can now be defined using raw SQL predicates in `@@unique` or `@@index` directives.
- Partial Indexes can now be defined using type-safe object literal syntax in `@@unique` or `@@index` directives (e.g., `where: { published: true }`).
🐛 Bug Fixes
- Fixed an issue in PostgreSQL migration scripts preventing the usage of `CREATE INDEX CONCURRENTLY`.
- Fixed BigInt precision loss in JSON aggregation for MySQL and CockroachDB by casting BigInt values to text.
- Fixed connection failures with non-ASCII database names by properly URL-decoding database names in connection strings.
- Fixed silent transaction commit errors in the PlanetScale adapter by ensuring COMMIT failures are properly propagated.
- Resolved race condition errors (EREQINPROG) in the SQL Server adapter by serializing commit/rollback operations using mutex synchronization.
- Fixed MSSQL connection string parsing to properly handle curly brace escaping for passwords containing special characters.