On 27.03.2025 10:48, Richard Guo wrote:
I'm wondering whether we should also remove parameter vardata1 from
eqjoinsel_semi. vardata2 is still needed though to clamp nd2 to be
not more than the rel's row estimate.
Thanks
Richard
Indeed, the parameter vardata1 in eqjoinsel_semi() is currently unused and could logically be removed. However, simply leaving a single parameter named vardata2 would appear strange and unintuitive, as it implicitly suggests the existence of a corresponding "first" parameter. I suggest renaming vardata2 to something more descriptive, such as rhs_vardata, clearly indicating its role related specifically to the right side of the join condition.
I attached v2 patch with changes.
Any thoughts?
--
Best regards,
Ilia Evdokimov,
Tantor Labs LLC.