Orville.PostgreSQL.Schema.TableDefinition

Since: 1.0.0.0

Documentation

data TableDefinition key writeEntity readEntity

Contains the definition of a SQL table for Orville to use for generating queries and marshalling Haskell values to and from the database.

• key is a Haskell type used to indicate whether the table has a primary key and what the type of the key is if so. See HasKey and NoKey for values to be used in this parameter.
• writeEntity is the Haskell type for values that Orville will write to the database for you (i.e. both inserts and updates).
• readEntity is the Haskell type for values that Orville will decode from the result set when entities are queried from this table.

Since: 1.0.0.0

data HasKey key

HasKey is a type with no constructors. It is used only at the type level as the key parameter to the TableDefinition type to indicate that the table has a primary key and what the Haskell type of the primary key is.

Since: 1.0.0.0

data NoKey

NoKey is a type with no constructors. It is used only at the type level as the key parameter to the TableDefinition type to indicate that the table does not have a primary key.

Since: 1.0.0.0

mkTableDefinition

Arguments

:: String

The name of the table

-> PrimaryKey key

Definition of the table's primary key

-> SqlMarshaller writeEntity readEntity

A SqlMarshaller to marshall table entities to and from the database

-> TableDefinition (HasKey key) writeEntity readEntity

Constructs a new TableDefinition with the basic fields required for operation. For convenience, this function accepts a PrimaryKey even though this is not required for all Orville operations to work. If you need to create a table without any primary key, see mkTableDefinitionWithoutKey.

Since: 1.0.0.0

mkTableDefinitionWithoutKey

Arguments

:: String

The name of the table

-> SqlMarshaller writeEntity readEntity

A SqlMarshaller to marshall table entities to and from the database

-> TableDefinition NoKey writeEntity readEntity

Constructs a new TableDefinition with the minimal fields required for operation. Note: tables created via this function will not have a primary key. Certain Orville functions require a primary key. Attempting to call functions requiring a primary key will fail to compile when using a table that has no key.

Since: 1.0.0.0

dropColumns

Arguments

:: [String]

Columns that should be dropped from the table

-> TableDefinition key writeEntity readEntity

-> TableDefinition key writeEntity readEntity

Annotates a TableDefinition with a direction to drop columns if they are found in the database. Orville does not drop columns during auto-migration unless they are explicitly requested to be dropped via dropColumns.

If you remove a reference to a column from the table's SqlMarshaller without adding the column's name to dropColumns, Orville will operate as if the column does not exist without actually dropping the column. This is often useful if you're not sure you want to lose the data in the column, or if you have zero down-time deployments, which requires the column not be referenced by deployed code before it can be dropped.

Since: 1.0.0.0

columnsToDrop :: TableDefinition key writeEntity readEntity -> Set String

Returns the set of columns that have been marked as dropped by dropColumns.

Since: 1.0.0.0

tableIdentifier :: TableDefinition key writeEntity readEntity -> TableIdentifier

Returns the table's TableIdentifier.

Since: 1.0.0.0

tableName :: TableDefinition key writeEntity readEntity -> Qualified TableName

Returns the table's name as an expression that can be used to build SQL statements. If the table has a schema name set, the name will be qualified with it.

Since: 1.0.0.0

setTableSchema :: String -> TableDefinition key writeEntity readEntity -> TableDefinition key writeEntity readEntity

Sets the table's schema to the name in the given String, which will be treated as a SQL identifier. If a table has a schema name set, it will be included as a qualifier on the table name for all queries involving the table.

Since: 1.0.0.0

tableConstraints :: TableDefinition key writeEntity readEntity -> TableConstraints

Retrieves all the table constraints that have been added to the table either via addTableConstraints or that are found on FieldDefinitions included with this table's SqlMarshaller.

Since: 1.0.0.0

addTableConstraints :: [ConstraintDefinition] -> TableDefinition key writeEntity readEntity -> TableDefinition key writeEntity readEntity

Adds the given table constraints to the table definition. It's also possible to add constraints that apply to only one column, adding them to the FieldDefinitions that are included in the table's SqlMarshaller.

If you wish to constrain multiple columns with a single constraint (e.g. a multi-column unique constraint), you must use addTableConstraints.

Note: If multiple constraints are added with the same ConstraintMigrationKey, only the last one that is added will be part of the TableDefinition. Any previously-added constraint with the same key is replaced by the new one.

Since: 1.0.0.0

tableIndexes :: TableDefinition key writeEntity readEntity -> Map IndexMigrationKey IndexDefinition

Retrieves all the table indexes that have been added to the table via addTableIndexes.

Since: 1.0.0.0

addTableIndexes :: [IndexDefinition] -> TableDefinition key writeEntity readEntity -> TableDefinition key writeEntity readEntity

Adds the given table indexes to the table definition.

Note: If multiple indexes are added with the same IndexMigrationKey, only the last one that is added will be part of the TableDefinition. Any previously-added index with the same key is replaced by the new one.

Since: 1.0.0.0

tablePrimaryKey :: TableDefinition (HasKey key) writeEntity readEntity -> PrimaryKey key

Returns the primary key for the table, as defined at construction via mkTableDefinition.

Since: 1.0.0.0

tableMarshaller :: TableDefinition key writeEntity readEntity -> AnnotatedSqlMarshaller writeEntity readEntity

Returns the marshaller for the table, as defined at construction via mkTableDefinition.

Since: 1.0.0.0

mapTableMarshaller :: (SqlMarshaller readEntityA writeEntityA -> SqlMarshaller readEntityB writeEntityB) -> TableDefinition key readEntityA writeEntityA -> TableDefinition key readEntityB writeEntityB

Applies the provided function to the underlying SqlMarshaller of the TableDefinition.

Since: 1.0.0.0

mkInsertExpr :: ReturningOption returningClause -> TableDefinition key writeEntity readEntity -> NonEmpty writeEntity -> InsertExpr

Builds an InsertExpr that will insert the given entities into the SQL table when it is executed. A RETURNING clause will either be included to return the inserted rows or not, depending on the ReturningOption given.

Since: 1.0.0.0

mkCreateTableExpr :: TableDefinition key writeEntity readEntity -> CreateTableExpr

Builds a CreateTableExpr that will create a SQL table matching the given TableDefinition when it is executed.

Since: 1.0.0.0

mkTableColumnDefinitions :: TableDefinition key writeEntity readEntity -> [ColumnDefinition]

Builds the ColumnDefinitions for all the fields described by the table definition's SqlMarshaller.

Since: 1.0.0.0

mkTablePrimaryKeyExpr :: TableDefinition key writeEntity readEntity -> Maybe PrimaryKeyExpr

Builds the PrimaryKeyExpr for this table, or none if this table has no primary key.

Since: 1.0.0.0

mkInsertColumnList :: SqlMarshaller writeEntity readEntity -> InsertColumnList

Builds an InsertColumnList that specifies the columns for an insert statement in the order that they appear in the given SqlMarshaller.

In normal circumstances you will want to build the complete insert statement via mkInsertExpr, but this is exported in case you are composing SQL yourself and need the column list of an insert as a fragment.

Since: 1.0.0.0

mkInsertSource :: SqlMarshaller writeEntity readEntity -> NonEmpty writeEntity -> InsertSource

Builds an InsertSource that will insert the given entities with their values specified in the order that the fields appear in the given SqlMarshaller (which matches the order of column names produced by mkInsertColumnList).

In normal circumstances you will want to build the complete insert statement via mkInsertExpr, but this is exported in case you are composing SQL yourself and need the column list of an insert as a fragment.

Since: 1.0.0.0

mkTableReturningClause :: ReturningOption returningClause -> TableDefinition key writeEntity readEntty -> Maybe ReturningExpr

When WithReturning is given, builds a ReturningExpr that will return all the columns in the given TableDefinition.

Since: 1.0.0.0