Schema fixes #31
Schema fixes #31
Conversation
Postgres can efficiently use any prefix of an index.
And reorder one UNIQUE key to have the foreign key column first
It's only queries together, so having it in the same index is more efficient than BitmapAnd-ing two indexes.
|
|
||
| my $schema = $self->result_source->schema; | ||
| $schema->resultset('Coverage')->search( | ||
| $self->coverages->search( |
There was a problem hiding this comment.
I do not understand how this change works, can you explain it?
There was a problem hiding this comment.
The has_many declaration in Schema::Run creates an accessor that's equivalent to ->related_resultset($name).
|
FYI I have "merged" this with some creative rebasing in this repo/tree here: https://github.com/Test-More/Test2-Harness/tree/2.0_merge_t2h I inserted them after merging the Test2::Harness::UI history into Test2::Harness, but before all the big changes I have made to where files are located. To my amazement there were no conflicts and the rebase was clean. I am leaving this PR open due to my request to understand one specific change (see inline comment). But once Test2::Harness 2.0 is shipped it will contain these improvements. Thank you, I greatly appreciate your help and contribution on this! |
No description provided.