On Sat, Jan 14, 2017 at 6:10 AM, Robert Haas <robertmhaas@gmail.com> wrote:
> On Fri, Jan 13, 2017 at 3:09 PM, Alvaro Herrera <alvherre@2ndquadrant.com> wrote:
>>
>> I'm just saying that the problem at hand is already solved for a related
>> feature, so ISTM this new code should use the recently added routine
>> rather than doing the same thing in a different way.
>
> Oh, I see. Amit, thoughts?
Hm, perhaps. The code in map_partition_varattnos() that creates the
map could be replaced by a call to the new
convert_tuples_by_name_map(). In fact, it could even have used the
old version of it (convert_tuples_by_name()). I guess I just aped
what other callers of map_variable_attnos() were doing, which is to
generate the map themselves (not that they ought to be changed to use
convert_tuples_by_name_map).
I will send a patch at my earliest convenience. Thanks to Alvaro for
pointing that out.
Thanks,
Amit