UNPKG

typeorm

Version:

Data-Mapper ORM for TypeScript and ES2021+. Supports MySQL/MariaDB, PostgreSQL, MS SQL Server, Oracle, SAP HANA, SQLite, MongoDB databases.

297 lines (296 loc) • 10.6 kB
import { ObjectLiteral } from "../../common/ObjectLiteral"; import { DataSource } from "../../data-source/DataSource"; import { ColumnMetadata } from "../../metadata/ColumnMetadata"; import { EntityMetadata } from "../../metadata/EntityMetadata"; import { QueryRunner } from "../../query-runner/QueryRunner"; import { RdbmsSchemaBuilder } from "../../schema-builder/RdbmsSchemaBuilder"; import { TableColumn } from "../../schema-builder/table/TableColumn"; import { Driver } from "../Driver"; import { ColumnType } from "../types/ColumnTypes"; import { CteCapabilities } from "../types/CteCapabilities"; import { DataTypeDefaults } from "../types/DataTypeDefaults"; import { MappedColumnTypes } from "../types/MappedColumnTypes"; import { ReplicationMode } from "../types/ReplicationMode"; import { PostgresConnectionCredentialsOptions } from "./PostgresConnectionCredentialsOptions"; import { PostgresConnectionOptions } from "./PostgresConnectionOptions"; import { PostgresQueryRunner } from "./PostgresQueryRunner"; import { Table } from "../../schema-builder/table/Table"; import { View } from "../../schema-builder/view/View"; import { TableForeignKey } from "../../schema-builder/table/TableForeignKey"; import { UpsertType } from "../types/UpsertType"; /** * Organizes communication with PostgreSQL DBMS. */ export declare class PostgresDriver implements Driver { /** * Connection used by driver. */ connection: DataSource; /** * Postgres underlying library. */ postgres: any; /** * Pool for master database. */ master: any; /** * Pool for slave databases. * Used in replication. */ slaves: any[]; /** * We store all created query runners because we need to release them. */ connectedQueryRunners: QueryRunner[]; /** * Connection options. */ options: PostgresConnectionOptions; /** * Version of Postgres. Requires a SQL query to the DB, so it is not always set */ version?: string; /** * Database name used to perform all write queries. */ database?: string; /** * Schema name used to perform all write queries. */ schema?: string; /** * Schema that's used internally by Postgres for object resolution. * * Because we never set this we have to track it in separately from the `schema` so * we know when we have to specify the full schema or not. * * In most cases this will be `public`. */ searchSchema?: string; /** * Indicates if replication is enabled. */ isReplicated: boolean; /** * Indicates if tree tables are supported by this driver. */ treeSupport: boolean; /** * Represent transaction support by this driver */ transactionSupport: "nested"; /** * Gets list of supported column data types by a driver. * * @see https://www.tutorialspoint.com/postgresql/postgresql_data_types.htm * @see https://www.postgresql.org/docs/9.2/static/datatype.html */ supportedDataTypes: ColumnType[]; /** * Returns type of upsert supported by driver if any */ supportedUpsertTypes: UpsertType[]; /** * Gets list of spatial column data types. */ spatialTypes: ColumnType[]; /** * Gets list of column data types that support length by a driver. */ withLengthColumnTypes: ColumnType[]; /** * Gets list of column data types that support precision by a driver. */ withPrecisionColumnTypes: ColumnType[]; /** * Gets list of column data types that support scale by a driver. */ withScaleColumnTypes: ColumnType[]; /** * Orm has special columns and we need to know what database column types should be for those types. * Column types are driver dependant. */ mappedDataTypes: MappedColumnTypes; /** * The prefix used for the parameters */ parametersPrefix: string; /** * Default values of length, precision and scale depends on column data type. * Used in the cases when length/precision/scale is not specified by user. */ dataTypeDefaults: DataTypeDefaults; /** * Max length allowed by Postgres for aliases. * @see https://www.postgresql.org/docs/current/sql-syntax-lexical.html#SQL-SYNTAX-IDENTIFIERS */ maxAliasLength: number; isGeneratedColumnsSupported: boolean; cteCapabilities: CteCapabilities; constructor(connection?: DataSource); /** * Performs connection to the database. * Based on pooling options, it can either create connection immediately, * either create a pool and create connection when needed. */ connect(): Promise<void>; /** * Makes any action after connection (e.g. create extensions in Postgres driver). */ afterConnect(): Promise<void>; protected enableExtensions(extensionsMetadata: any, connection: any): Promise<void>; protected checkMetadataForExtensions(): Promise<{ hasUuidColumns: boolean; hasCitextColumns: boolean; hasHstoreColumns: boolean; hasCubeColumns: boolean; hasGeometryColumns: boolean; hasLtreeColumns: boolean; hasExclusionConstraints: boolean; hasExtensions: boolean; }>; /** * Closes connection with database. */ disconnect(): Promise<void>; /** * Creates a schema builder used to build and sync a schema. */ createSchemaBuilder(): RdbmsSchemaBuilder; /** * Creates a query runner used to execute database queries. */ createQueryRunner(mode: ReplicationMode): PostgresQueryRunner; /** * Prepares given value to a value to be persisted, based on its column type and metadata. */ preparePersistentValue(value: any, columnMetadata: ColumnMetadata): any; /** * Prepares given value to a value to be persisted, based on its column type or metadata. */ prepareHydratedValue(value: any, columnMetadata: ColumnMetadata): any; /** * Replaces parameters in the given sql with special escaping character * and an array of parameter names to be passed to a query. */ escapeQueryWithParameters(sql: string, parameters: ObjectLiteral, nativeParameters: ObjectLiteral): [string, any[]]; /** * Escapes a column name. */ escape(columnName: string): string; /** * Build full table name with schema name and table name. * E.g. myDB.mySchema.myTable */ buildTableName(tableName: string, schema?: string): string; /** * Parse a target table name or other types and return a normalized table definition. */ parseTableName(target: EntityMetadata | Table | View | TableForeignKey | string): { database?: string; schema?: string; tableName: string; }; /** * Creates a database type from a given column metadata. */ normalizeType(column: { type?: ColumnType; length?: number | string; precision?: number | null; scale?: number; isArray?: boolean; }): string; /** * Normalizes "default" value of the column. */ normalizeDefault(columnMetadata: ColumnMetadata): string | undefined; /** * Compares "default" value of the column. * Postgres sorts json values before it is saved, so in that case a deep comparison has to be performed to see if has changed. */ private defaultEqual; /** * Normalizes "isUnique" value of the column. */ normalizeIsUnique(column: ColumnMetadata): boolean; /** * Returns default column lengths, which is required on column creation. */ getColumnLength(column: ColumnMetadata): string; /** * Creates column type definition including length, precision and scale */ createFullType(column: TableColumn): string; /** * Obtains a new database connection to a master server. * Used for replication. * If replication is not setup then returns default connection's database connection. */ obtainMasterConnection(): Promise<[any, Function]>; /** * Obtains a new database connection to a slave server. * Used for replication. * If replication is not setup then returns master (default) connection's database connection. */ obtainSlaveConnection(): Promise<[any, Function]>; /** * Creates generated map of values generated or returned by database after INSERT query. * * todo: slow. optimize Object.keys(), OrmUtils.mergeDeep and column.createValueMap parts */ createGeneratedMap(metadata: EntityMetadata, insertResult: ObjectLiteral): ObjectLiteral | undefined; /** * Differentiate columns of this table and columns from the given column metadatas columns * and returns only changed. */ findChangedColumns(tableColumns: TableColumn[], columnMetadatas: ColumnMetadata[]): ColumnMetadata[]; private lowerDefaultValueIfNecessary; /** * Returns true if driver supports RETURNING / OUTPUT statement. */ isReturningSqlSupported(): boolean; /** * Returns true if driver supports uuid values generation on its own. */ isUUIDGenerationSupported(): boolean; /** * Returns true if driver supports fulltext indices. */ isFullTextColumnTypeSupported(): boolean; get uuidGenerator(): string; /** * Creates an escaped parameter. */ createParameter(parameterName: string, index: number): string; /** * Loads postgres query stream package. */ loadStreamDependency(): any; /** * If driver dependency is not given explicitly, then try to load it via "require". */ protected loadDependencies(): void; /** * Creates a new connection pool for a given database credentials. */ protected createPool(options: PostgresConnectionOptions, credentials: PostgresConnectionCredentialsOptions): Promise<any>; /** * Closes connection pool. */ protected closePool(pool: any): Promise<void>; /** * Executes given query. */ protected executeQuery(connection: any, query: string): Promise<unknown>; /** * If parameter is a datetime function, e.g. "CURRENT_TIMESTAMP", normalizes it. * Otherwise returns original input. */ protected normalizeDatetimeFunction(value: string): string; /** * Escapes a given comment. */ protected escapeComment(comment?: string): string | undefined; }