Add scalar row factory

[embecka]

Please, add scalar_row factory to return a scalar value instead of a tuple.

Usage:

with connection.cursor(row_factory=scalar_row()) as cursor:
	cursor.execute("SELECT username FROM user WHERE id = %(id)s", {"id": 1})
	
	print(cursor.fetchone())  # Prints "username" instead of ("username",)

Possible implementation:

def scalar_row(index: int = 0) -> BaseRowFactory[Any]:
	def scalar_row_(cursor: BaseCursor[Any, Any]) -> RowMaker[Any]:
		res = cursor.pgresult
		if not res:
			return no_result
		
		nfields = _get_nfields(res)
		if nfields is None:
			return no_result
		
		if index >= nfields:
			return no_result
		
		def scalar_row__(values: Sequence[Any]) -> Any:
			return values[index]
		
		return scalar_row__
	
	return scalar_row_

[dvarrazzo]

Hello,

funnily enough, I needed myself something similar to this. I solved it by using:

cur = conn.cursor(row_factory=args_row(lambda x: x))

to return the first field of the record as a scalar. It is a bit too magic, I understand, so I was thinking that a "scalar_row" type of thing could be useful.