How / Where does the record returned from the DB get mapped to a plain JS object?

[anthonyorona]

I am using slonik to handle some of my more complex queries. I would like to learn more about where the DB row gets mapped to a plain JS object

[nathanchapman]

Slonik is built on top of pg. Additional details here: https://github.com/gajus/slonik#pg-vs-slonik

pg handles the conversion from a db result to a JS object.

Here's a good starting point in slonik to better understand how the query is executing and the additional metadata on the query result:

slonik/src/connectionMethods/query.ts

Lines 10 to 41 in 1228dba

	const executionRoutine: ExecutionRoutine = async (
	finalConnection,
	finalSql,
	finalValues,
	) => {
	const result: PgQueryResult & { notices?: Notice[] } =
	await finalConnection.query(
	finalSql,
	// eslint-disable-next-line @typescript-eslint/no-explicit-any
	finalValues as any[],
	);
	const fields: Field[] = [];
	if (result.fields) {
	for (const field of result.fields) {
	fields.push({
	dataTypeId: field.dataTypeID,
	name: field.name,
	});
	}
	}
	return {
	command: result.command as QueryResult<unknown>['command'],
	fields,
	notices: result.notices ?? [],
	rowCount: result.rowCount || 0,
	rows: result.rows || [],
	type: 'QueryResult',
	};
	};

If you want to modify the behavior of slonik, you can use an interceptor, specifically a result parser interceptor or some of the community interceptors to modify how result objects are built