On 2015/10/01 22:31, Tom Lane wrote:
> Etsuro Fujita writes:
>> The following is a remark added to /src/backend/optimizer/README by
>> commit 8703059c6b55c427100e00a09f66534b6ccbfaa1, and IIUC, I think "LHS"
>> in the last sentence "We prevent that by forcing the min LHS for the
>> upper join to i
Etsuro Fujita writes:
> The following is a remark added to /src/backend/optimizer/README by
> commit 8703059c6b55c427100e00a09f66534b6ccbfaa1, and IIUC, I think "LHS"
> in the last sentence "We prevent that by forcing the min LHS for the
> upper join to include B." should be "RHS".
Mmm, yeah, tha
The following is a remark added to /src/backend/optimizer/README by
commit 8703059c6b55c427100e00a09f66534b6ccbfaa1, and IIUC, I think "LHS"
in the last sentence "We prevent that by forcing the min LHS for the
upper join to include B." should be "RHS".
The use of minimum Relid sets has some pitfal