Re: [BUG] temporary file usage report with extended protocol and unnamed portals - Mailing list pgsql-hackers

From Mircea Cadariu
Subject Re: [BUG] temporary file usage report with extended protocol and unnamed portals
Date
Msg-id fa09a2d9-ccea-4e98-9e6f-5d442bb9b055@gmail.com
Whole thread Raw
In response to Re: [BUG] temporary file usage report with extended protocol and unnamed portals  (Guillaume Lelarge <guillaume.lelarge@dalibo.com>)
List pgsql-hackers
Shall we just look at the latest v13 and determine if it is a solution 
for showing the correct query we can stand behind, if it's not, explain 
why not, and in that case, possibly even figure out what preliminary 
refactoring is in order that would help us arrive at a solution in a 
more direct way?

I'd find this useful from an educational perspective, and possibly 
others as well which have taken the time to build a mental model of this 
area with this occasion.

The patch is more self-contained and less invasive than previous 
proposals, which makes potential reverts easier.

Maybe what's debatable is the temporary storage in 
saved_debug_query_string and back. The tests pass without this part 
indeed. IIUC the switch back happens "naturally" in PortalStart when the 
updated source text gets set.

But okay, just to be on the safe side we can keep the temporary storage 
and switch so that the "old" query is in the picture strictly as minimum 
as necessary.

Kind regards,

Mircea Cadariu





pgsql-hackers by date:

Previous
From: Andrey Borodin
Date:
Subject: Re: [PATCH] GROUP BY ALL
Next
From: Arseniy Mukhin
Date:
Subject: Re: GIN tries to form a tuple with a partial compressedList during insertion