Method _refresh_for_new_column

lib/sqlalchemy/sql/selectable.py:1053–1078

Given a column added to the .c collection of an underlying selectable, produce the local version of that column, assuming this selectable ultimately should proxy this column. this is used to "ping" a derived selectable to add a new column to its .c. collection when a

(self, column: ColumnElement[Any])

        """

    def _refresh_for_new_column(self, column: ColumnElement[Any]) -> None:
        """Given a column added to the .c collection of an underlying
        selectable, produce the local version of that column, assuming this
        selectable ultimately should proxy this column.

        this is used to "ping" a derived selectable to add a new column
        to its .c. collection when a Column has been added to one of the
        Table objects it ultimately derives from.

        If the given selectable hasn't populated its .c. collection yet,
        it should at least pass on the message to the contained selectables,
        but it will return None.

        This method is currently used by Declarative to allow Table
        columns to be added to a partially constructed inheritance
        mapping that may have already produced joins.  The method
        isn't public right now, as the full span of implications
        and/or caveats aren't yet clear.

        It's also possible that this functionality could be invoked by
        default via an event, which would require that
        selectables maintain a weak referencing collection of all
        derivations.

        """
        self._reset_column_collection()

    def _anonymous_fromclause(
        self, *, name: Optional[str] = None, flat: bool = False