SQLite connector

SQLite uses the jdbc:sqlite: scheme with the org.sqlite.JDBC driver. It can be a source or sink. As a source, all modes are available. As a sink, it supports complete and incremental modes. The sink does not support complete-atomic mode.

Connection

Item	Value
Source or sink URL	jdbc:sqlite:${SQLITE_PATH}
Driver class	org.sqlite.JDBC
Connection parameters	busy_timeout is set to 30000 ms by ReplicaDB

Source options

Property	Description
source.table	Table to read.
source.columns	Comma-separated column list to read.
source.where	SQL filter applied to the source read.
source.query	Free-form SQL query used as the source.

Sink options

Property	Description
sink.table	Target table. A schema prefix such as main.t_sink is allowed.
sink.columns	Comma-separated column list to write.
sink.staging.schema	Schema used for staging tables during incremental merges, for example main.
mode	complete or incremental. complete-atomic is not supported for this sink.

Locking and concurrency

SQLite is a single-file database, so writes use file locking. ReplicaDB sets busy_timeout to 30000 ms so that a connection waits for a lock instead of failing immediately. When the connection is closed, ReplicaDB runs a WAL checkpoint to release locks.

Incremental mode

Incremental merges match rows on the sink primary key, so the sink table must define one.

Example

Replicate a table from a source database into a SQLite file:

source.connect=jdbc:postgresql://${PGHOST}:5432/catalog
source.user=${PGUSER}
source.password=${PGPASSWORD}
source.table=t_source

sink.connect=jdbc:sqlite:${SQLITE_PATH}
sink.table=main.t_sink
sink.staging.schema=main

mode=complete