Thanks Tom.
On Thu, Nov 24, 2016 at 2:58 AM, Tom Lane wrote:
> Ashutosh Bapat writes:
>> On Sat, Nov 5, 2016 at 2:16 AM, Tom Lane wrote:
>>> Ashutosh Bapat writes:
Also, the way this code has been written, the declaration of variable
sjinfo masks the earlier declaration with the same
Ashutosh Bapat writes:
> On Sat, Nov 5, 2016 at 2:16 AM, Tom Lane wrote:
>> Ashutosh Bapat writes:
>>> Also, the way this code has been written, the declaration of variable
>>> sjinfo masks the earlier declaration with the same name.
>> Hmm, yeah, that's probably not terribly good coding practi
On Sat, Nov 5, 2016 at 2:16 AM, Tom Lane wrote:
> Ashutosh Bapat writes:
>> There's code in add_paths_to_joinrel() which computes the set of
>> target relations that should overlap parameterization of any proposed
>> join path.
>> ...
>> The calculations that follow are based on joinrel->relids (
Ashutosh Bapat writes:
> There's code in add_paths_to_joinrel() which computes the set of
> target relations that should overlap parameterization of any proposed
> join path.
> ...
> The calculations that follow are based on joinrel->relids (baserels
> covered by the join) and SpecialJoinInfo list
Hi,
There's code in add_paths_to_joinrel() which computes the set of
target relations that should overlap parameterization of any proposed
join path.
120 /*
121 * Decide whether it's sensible to generate parameterized
paths for this
122 * joinrel, and if so, which relations such p