Unnecessary has_cap checks for MySQL 4.1

[filosofo]

Now that we're using MySQL 4.1 as a minimum requirement, we don't have to do wpdb capability checks for its features.

Patch removes checks.

[ryan]

(In [12409]) Remove no longer needed DB has_cap() checks. Props filosofo. fixes #11443

[westi]

I am a bit wary of doing this this close to release.

While we require the mysql version for our wp-db class a drop in replacement class could be being used which doesn't have these caps.

I wonder whether we should be making this change at this point in the cycle.

Can we not leave taking this out till 3.0?

[ryan]

I doubt any of the drop-ins have the limitations our old MySQL requirement did, but point taken.

[ryan]

Drop-ins have to update with every release of WP anyway. Removing has_cap() checks that probably only have relevance to ancient MySQL versions probably won't affect them any more than the new SQL added to 2.9 will (such as ON DUPLICATE KEY UPDATE).

[filosofo]

For what it's worth, there are only 2 has_cap checks relevant to wpdb drop-ins (in wp-includes/query.php), and they both use what I believe is pretty standard SQL subquery syntax.

[Denis-de-Bernardy]

Replying to westi:

Can we not leave taking this out till 3.0?

I initially had the same thought when reading this ticket :-)

Then again...

While we require the mysql version for our wp-db class a drop in replacement class could be being used which doesn't have these caps.

In this case the recently introduced on duplicate key statement won't work either, meaning they won't be able to update options. Also, aren't we save to assume that anyone who is savvy enough to install a drop in replacement class of wpdb is also a) savvy enough to be using MySQL 5.1 or b) (see #10918) savvy enough to fix their query translator accordingly?

[ryan]

Indeed. I almost punted this but thought removing the cap checks would make us most consistent given the introduction of ON DUPLICATE. Any breakage that occurs from this could be a good thing because it would forcibly remind drop-in authors to update. I don't feel strongly either way, however.

[westi]

Replying to ryan:

Indeed. I almost punted this but thought removing the cap checks would make us most consistent given the introduction of ON DUPLICATE. Any breakage that occurs from this could be a good thing because it would forcibly remind drop-in authors to update. I don't feel strongly either way, however.

It is good to remind them to update.

Lets leave these changes in for now and just hi-light in the RC and Release announcements the new minimum mysql version requirement.

[miqrogroove]

Replying to westi:

I am a bit wary of doing this this close to release.

Good intuition. see #11666