Re: Rework manipulation and structure of attribute mappings

On Tue, Dec 17, 2019 at 01:54:27PM +0900, Amit Langote wrote:
> Thanks for the updated patch. I don't have any comments, except that
> the text I suggested couple of weeks ago no longer reads clear:

I have spent a couple of extra hours on the patch, and committed it.
There was one issue in logicalrelation.h which failed to compile
standalone.

> + * by DDL operating on inheritance and partition trees to convert fully
> + * transformed expression trees from parent rowtype to child rowtype or
> + * vice-versa.
>
> Maybe:
>
> ...to adjust the Vars in fully transformed expression trees to bear
> output relation's attribute numbers.

I have used something more generic at the end:
+ * mappings by comparing input and output TupleDescs. Such mappings
+ * are typically used by DDL operating on inheritance and partition trees
+ * to do a conversion between rowtypes logically equivalent but with
+ * columns in a different order, taking into account dropped columns.
--
Michael