Re: Refactor query normalization into core query jumbling

From: Lukas Fittl <lukas(at)fittl(dot)com>
To: Sami Imseih <samimseih(at)gmail(dot)com>
Cc: Michael Paquier <michael(at)paquier(dot)xyz>, zengman <zengman(at)halodbtech(dot)com>, pgsql-hackers <pgsql-hackers(at)postgresql(dot)org>, Julien Rouhaud <rjuju123(at)gmail(dot)com>
Subject: Re: Refactor query normalization into core query jumbling
Date: 2026-03-26 16:35:15
Message-ID: CAP53PkyY7AhQ8jZDwtHokyEfpuHLxxDvtY7EmcasR_n80yLiRQ@mail.gmail.com
Views: Whole Thread | Raw Message | Download mbox | Resend email
Thread:
Lists: pgsql-hackers

Hi Sami,

On Wed, Mar 25, 2026 at 8:31 PM Sami Imseih <samimseih(at)gmail(dot)com> wrote:
>
> v6 addresses the comments.
>

Great - I've tested that again and changes look good.

Two more thoughts:

1) I think "SetConstantLengths" should be renamed to
"GetConstantLengths", or "CalculateConstantLengths" in 0002, since
we're no longer modifying JumbleState

2) I wonder if we should export this function in 0002 - that would
specifically help pg_tracing, which also wants to extract inline
parameter values in addition to replacing them with a $n parameter
reference marker - I could also see that being useful for any other
extension that's interested in pulling out parameter values

I've also done some testing with two previously mentioned extensions,
pg_stat_monitor [0] and pg_tracing [1]. Both look like they can be
adapted to the new method with their regression tests succeeding.

Thanks,
Lukas

[0]: https://github.com/lfittl/pg_stat_monitor/commit/d3521de9f6736c1bb745bc31dae736540efe1781
[1]: https://github.com/lfittl/pg_tracing/commit/ecff95a87a1c60e8a5fe47662cc6a2148bd3632f

--
Lukas Fittl

In response to

Responses

Browse pgsql-hackers by date

  From Date Subject
Next Message Andrey Borodin 2026-03-26 16:37:18 Re: Add uuid_to_base32hex() and base32hex_to_uuid() built-in functions
Previous Message Sami Imseih 2026-03-26 16:34:32 Re: Clean up NamedLWLockTranche stuff