Database.Persist.Postgresql.Internal.Migration

Returns a structured representation of all of the DB changes required to migrate the Entity from its current state in the database to the state described in Haskell.

Returns a structured representation of all of the DB changes required to migrate the listed entities from their current state in the database to the state described in Haskell. This function avoids N+1 queries, so if you have a lot of entities to migrate, it's much faster to use this rather than using migrateStructured in a loop.

Returns a structured representation of all of the DB changes required to migrate the Entity to the state described in Haskell, assuming it currently does not exist in the database.

In order to ensure that generating migrations is fast and avoids N+1 queries, we split it into two phases. The first phase involves querying the database to gather all of the information we need about the existing schema. The second phase then generates migrations based on the information from the first phase. This data type represents all of the data that's gathered during the first phase: information about the current state of the entities we're migrating in the database.

The state of a particular entity (i.e. table) in the database; we generate migrations based on the diff of this versus an EntityDef.

Information about an existing table in the database

Query a database in order to assemble a SchemaState containing information about each of the entities in the given list. Every entity name in the input should be present in the returned Map.

Check for the existence of each of the input tables. The keys in the returned Map are exactly the entity names in the argument; True means the table exists.

Get all columns for the listed tables from the database, ignoring foreign key references (those are filled in later).

Get all constraints for the listed tables from the database, except for foreign keys and primary keys (those go in the Column data type)

Get foreign key constraint information for all columns in the supplied tables from the database. We return a list of references per column because there may be duplicate FK constraints in the database.

Note that we only care about FKs where the column in question has ordinal position 1 i.e. is the first column appearing in the FK constraint. Eventually we may want to fill this gap so that multi-column FK constraints can be dealt with by this migrator, but for now that is not something that persistent-postgresql handles.

Indicates whether a Postgres Column is safe to drop.

Represents a change to a Postgres column in a DB statement.

Represents a change to a Postgres table in a DB statement.

Represents a change to a Postgres DB in a statement.

Create a table if it doesn't exist.

Postgres' default maximum identifier length in bytes (You can re-compile Postgres with a new limit, but I'm assuming that virtually noone does this). See https://www.postgresql.org/docs/11/sql-syntax-lexical.html#SQL-SYNTAX-IDENTIFIERS

Intelligent comparison of SQL types, to account for SqlInt32 vs SqlOther integer

Generate the default foreign key constraint name for a given source table and source column name. Note that this function should generally not be used except as an argument to postgresMkColumns, because if you use it in other contexts, you're likely to miss nonstandard constraint names declared in the persistent models files via `constraint=`

Check if a column name is listed as the "safe to remove" in the entity list.

Get the references to be added to a table for the given column.