Horton

Re: making EXPLAIN extensible

Andrei Lepikhov <lepihov@gmail.com>

View thread
From: Andrei Lepikhov <lepihov@gmail.com>
To: Robert Haas <robertmhaas@gmail.com>
Cc: Jeff Davis <pgsql@j-davis.com>, Thom Brown <thom@linux.com>, PostgreSQL Hackers <pgsql-hackers@lists.postgresql.org>, Tom Lane <tgl@sss.pgh.pa.us>
Date: 2025-03-18T12:02:08Z
Lists: pgsql-hackers

Commits

Same data as JSON: GET /api/v1/messages/:b64id/commits the thread's linked commits as JSON, with link sources. API reference →

  1. pg_overexplain: Use PG_MODULE_MAGIC_EXT.

  2. pg_overexplain: Call previous hooks as appropriate.

  3. pg_overexplain: Additional EXPLAIN options for debugging.

  4. Add an additional hook for EXPLAIN option validation.

  5. Add some new hooks so extensions can add details to EXPLAIN.

  6. Make it possible for loadable modules to add EXPLAIN options.

On 3/7/25 16:05, Robert Haas wrote:
I have attempted to use hooks, proposed in 0002, in my extensions.
At first, it worked great. My patch reduced a lot, and the only things 
that I need in the planner to improve its predictions are the 
selectivity hook and the create_plan hook - the last one needed to pass 
data from the best_path chosen to the plan and needed to compare 
prediction and reality at the end of execution.

Some questions:
1. I think, hooks ExplainOneQuery_hook_type, explain_per_plan_hook_type, 
explain_per_node_hook_type deserve to be moved to explain_format.h
At least, inside the hook, we usually use functions like ExplainProperty.
2. In my patch I inserted the hook before the line 1894:
/* in text format, the first line ends here */
Why have you chosen a different way? I don't have specific reasons to 
insist, except the extension data right under the node looks better to 
me personally.

Some changes of the explain format have already been made in the current 
master. Applying this proposal now would reduce additional work for 
extension (and fork) maintainers in the next version.

-- 
regards, Andrei Lepikhov