Programmatic API

Alongside command line, you can use node-pg-migrate also programmatically. It exports runner function, which takes options argument with the following structure (similar to command line arguments):

Options

NOTE

If you use dbClient, you should not use databaseUrl at the same time and vice versa.

Option	Type	Description
databaseUrl	string or object	Connection string or client config which is passed to new pg.Client
dbClient	pg.Client	Instance of new pg.Client. Instance should be connected to DB, and after finishing migration, user is responsible to close connection
migrationsTable	string	The table storing which migrations have been run
migrationsSchema	string	The schema storing table which migrations have been run (defaults to same value as schema)
schema	string or array[string]	The schema on which migration will be run (defaults to public)
dir	string or array[string]	The directory containing your migration files. This path is resolved from cwd(). Alternatively, provide a glob pattern or an array of glob patterns and set useGlob = true. Note: enabling glob will read both, dir and ignorePattern as glob patterns
useGlob	boolean	Use glob to find migration files. This will use dir and ignorePattern to glob-search for migration files. Note: enabling glob will read both, dir and ignorePattern as glob patterns
checkOrder	boolean	Check order of migrations before running them
direction	enum	up or down
count	number	Amount of migration to run
timestamp	boolean	Treats count as timestamp
ignorePattern	string or array[string]	Regex pattern for file names to ignore (ignores files starting with . by default). Alternatively, provide a glob pattern or an array of glob patterns and set isGlob = true. Note: enabling glob will read both, dir and ignorePattern as glob patterns
file	string	Run-only migration with this name
singleTransaction	boolean	Combines all pending migrations into a single transaction so that if any migration fails, all will be rolled back (defaults to true)
createSchema	boolean	Creates the configured schema if it doesn't exist
createMigrationsSchema	boolean	Creates the configured migration schema if it doesn't exist
noLock	boolean	Disables locking mechanism and checks
lockValue	number	Value to use for the lock
fake	boolean	Mark migrations as run without actually performing them (use with caution!)
dryRun	boolean
log	function	Redirect log messages to this function, rather than console
logger	object with debug/info/warn/error methods	Redirect messages to this logger object, rather than console
verbose	boolean	Print all debug messages like DB queries run (if you switch it on, it will disable logger.debug method)
decamelize	boolean	Runs decamelize on table/column/etc. names
migrationLoaderStrategies	MigrationLoaderStrategy[]	Allows custom loading strategies based on file extensions. If omitted, default behavior is used. See Migration Loading Strategies.

MigrationLoaderStrategy

export interface MigrationLoaderStrategy {
  // File extensions handled by this strategy.
  extensions: string[];

  /**
   * Loader that handles conversion of file paths to migration units.
   *
   * @param filePaths - The file paths to load migrations from.
   * @returns The migration units.
   */
  loader: MigrationLoader | 'default' | 'legacySql' | 'sql';
}

MigrationUnit

export interface MigrationUnit {
  // The unique identifier for the migration unit. Represents the significant part of the file name used for tracking which migrations have been performed.
  id: string;

  // File paths that are part of the migration unit.
  filePaths: string[];

  // The migration builder actions that are contained within the migration files.
  actions: MigrationBuilderActions;
}