Multiplatform Persistence
summary
Shows concrete multiplatform SQLite work: SQLDelight 1.0 generates Kotlin from .sq files, uses expect/actual to let Android supply a SupportSQLite driver and iOS use Narc SQLite driver, and exposes a small runtime plus column adapters. Migrations gain a schema+verify task; an in‑memory JDBC driver enables JVM tests. Queries become extension‑backed observables (RX Java kept out of core). Kotlin/Native limits (no generics, no value types yet) force awkward bridges but tests run on Android and native.
AI-written summary · about
chapters (18)
- 0:00Introduction
- 1:11Square Cache ecosystem
- 3:57Data flow in Cache
- 6:25Modularization benefits
- 9:55Current Cache dependencies
- 11:19Defining module boundaries
- 14:05Platformless persistence goal
- 15:07SQL Delight multiplatform
- 18:16Multiplatform setup
- 21:46Benefits of common code
- 24:13Query observables with RX
- 25:37iOS value types support
- 27:19Pros and cons of multiplatform
- 29:04Cross-platform testing
- 29:25SQL Delight migrations
- 31:10Testing with JDBC driver
- 31:51Future of Cache modules
- 33:16SDK Search example
