Re: PoC/WIP: Extended statistics on expressions

From: Tomas Vondra <tomas(dot)vondra(at)enterprisedb(dot)com>
To: Justin Pryzby <pryzby(at)telsasoft(dot)com>
Cc: Dean Rasheed <dean(dot)a(dot)rasheed(at)gmail(dot)com>, pgsql-hackers(at)postgresql(dot)org
Subject: Re: PoC/WIP: Extended statistics on expressions
Date: 2021-08-16 13:19:53
Message-ID: 58cbfa74-9356-778b-3e10-94e3075c5807@enterprisedb.com
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-hackers

On 8/16/21 3:31 AM, Justin Pryzby wrote:
> On 1/22/21 5:01 AM, Justin Pryzby wrote:
>>> In any case, why are there so many parentheses ?
>
> On Fri, Jan 22, 2021 at 02:09:04PM +0100, Tomas Vondra wrote:
>> That's a bug in pg_get_statisticsobj_worker, probably. It shouldn't be
>> adding extra parentheses, on top of what deparse_expression_pretty does.
>> Will fix.
>
> The extra parens are still here - is it intended ?
>

Ah, thanks for reminding me! I was looking at this, and the problem is
that pg_get_statisticsobj_worker only does this:

prettyFlags = PRETTYFLAG_INDENT;

Changing that to

prettyFlags = PRETTYFLAG_INDENT | PRETTYFLAG_PAREN;

fixes this (not sure we need the INDENT flag - probably not).

I'm a bit confused, though. My assumption was "PRETTYFLAG_PAREN = true"
would force the deparsing itself to add the parens, if needed, but in
reality it works the other way around.

I guess it's more complicated due to deparsing multi-level expressions,
but unfortunately, there's no comment explaining what it does.

regards

--
Tomas Vondra
EnterpriseDB: http://www.enterprisedb.com
The Enterprise PostgreSQL Company

In response to

Browse pgsql-hackers by date

  From Date Subject
Next Message Peter Eisentraut 2021-08-16 13:20:07 Some RELKIND macro refactoring
Previous Message Daniel Gustafsson 2021-08-16 11:42:01 Re: Slightly improve initdb --sync-only option's help message