As I said in this thread:
https://lists.apache.org/thread.html/r6tk31vn9pyrglt0lgym2093srjzx0pq
I hope the native query support can be included in this release.

Beside sys.server_properties which is included in #20183 as a
demonstration, I want the sys.tasks and sys.segments table to be included
too.
Actually the support for all existing system tables have been done for a
while on our own branches,
this means that the #20183 is the only blocker for us to move forward.
The changes in #20183 looks big, but it provides an opt-in way to query
system tables on the native engine while the current bindable path is still
the default one.

If we are not able to review the change line by line for this PR, and if
the general direction of the implementation is OK for us,
we can include this native query path as an Experimental feature so that we
don't block left work.

As for #20198, it was merged. And 4 or 5 follow up PRs will be submitted to
complete the whole refactoring.
The ultimate goal is to provide another system table
sys.query_context_parameters on the bindable path first (Development is
also done in my branch)

I hope we can achieve this goal in this release.


On Mon, Oct 12, 2026 at 8:50 PM Sasha Syrotenko <[email protected]> wrote:

> Hello, folks.
>
> On the 1th of October, we released Druid 38.
> To maintain the pace for quarterly releases, we would like to start
> preparing the release of Druid 39.0, targeting mid-November.
> I am volunteering to be the manager for this release.
>
> The tentative date for the code-freeze is October, 17th/18th.
>
> I am aware of the following important PRs, tagged for 39th release:
>
> 1. https://github.com/apache/druid/pull/20183 - native query support
> for system tables;
> 2. https://github.com/apache/druid/pull/20198 - typed query context
> parameters;
> 3. https://github.com/apache/druid/pull/20362 - reduce allocations for
> massive IN filters;
>
> You can find list of all PRs tagged for Druid 39.0 here:
> https://github.com/apache/druid/milestone/69.
>
> We can use this thread to discuss any high-priority issue or critical
> feature that must be included in Druid 39.0.
>
>
> Also, my apologies for the short notice, but we are already behind planned
> schedule.
> Let's ensure that all open PRs supposed to be included into Druid 39 are
> appropriately tagged and fully reviewed by then. Best wishes, Sasha.
>

Reply via email to