Database.Persist.SqlBackend.Internal

This function should prepare a Statement in the target database, which should allow for efficient query reuse.

This function generates the SQL and values necessary for performing an insert against the database.

SQL for inserting many rows and returning their primary keys, for backends that support this functionality. If Nothing, rows will be inserted one-at-a-time using connInsertSql.

Some databases support performing UPSERT _and_ RETURN entity in a single call.

This field when set will be used to generate the UPSERT+RETURN sql given * an entity definition * updates to be run on unique key(s) collision

When left as Nothing, we find the unique key from entity def before * trying to fetch an entity by said key * perform an update when result found, else issue an insert * return new entity from db

Some databases support performing bulk UPSERT, specifically "insert or replace many records" in a single call.

This field when set, given * an entity definition * number of records to be inserted should produce a PUT MANY sql with placeholders for records

When left as Nothing, we default to using defaultPutMany.

A reference to the cache of statements. Statements are keyed by the Text queries that generated them.

Close the underlying connection.

This function returns the migrations required to include the EntityDef parameter in the [EntityDef] database. This might include creating a new table if the entity is not present, or altering an existing table if it is.

A function to begin a transaction for the underlying database.

A function to commit a transaction to the underlying database.

A function to roll back a transaction on the underlying database.

A function to extract and escape the name of the column corresponding to the provided field.

A function to extract and escape the name of the table corresponding to the provided entity. PostgreSQL uses this to support schemas.

A function to escape raw DB identifiers. MySQL uses backticks, while PostgreSQL uses quotes, and so on.

A tag displaying what database the SqlBackend is for. Can be used to differentiate features in downstream libraries for different database backends.

Attach a 'LIMIT/OFFSET' clause to a SQL query. Note that LIMIT/OFFSET is problematic for performance, and indexed range queries are the superior way to offer pagination.

A log function for the SqlBackend to use.

Some databases (probably only Sqlite) have a limit on how many question-mark parameters may be used in a statement

Some databases support performing bulk an atomic+bulk INSERT where constraint conflicting entities can replace existing entities.

This field when set, given * an entity definition * number of records to be inserted should produce a INSERT sql with placeholders for primary+record fields

When left as Nothing, we default to using defaultRepsertMany.

Carry arbitrary payloads for the connection that may be used to propagate information into hooks.

Instrumentation hooks that may be used to track the behaviour of a backend.