Re: plpgsql.warn_shadow

From: Marko Tiikkaja <marko(at)joh(dot)to>
To: Pavel Stehule <pavel(dot)stehule(at)gmail(dot)com>
Cc: Florian Pflug <fgp(at)phlo(dot)org>, PostgreSQL-development <pgsql-hackers(at)postgresql(dot)org>, Jim Nasby <jim(at)nasby(dot)net>
Subject: Re: plpgsql.warn_shadow
Date: 2014-01-15 13:58:58
Message-ID: 52D69422.60809@joh.to
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-hackers

On 1/15/14 2:27 PM, Pavel Stehule wrote:
> 2014/1/15 Marko Tiikkaja <marko(at)joh(dot)to>
>> Yeah, me neither, it's just something that needs to be communicated very
>> clearly. So probably just a list plpgsql.warnings would be the most
>> appropriate then.
>>
>
> I am thinking so the name is not good. Changing handling warnings is messy
> - minimally in Postgres, where warnings and errors are different creatures.
>
> what about
>
> plpgsql.enhanced_checks = (none | warning | error)

You crammed several suggestions into one here:

1) You're advocating the ability to turn warnings into errors. This
has been met with some resistance. I think it's a useful feature, but I
would be happy with just having warnings available.
2) This syntax doesn't allow the user to specify a list of warnings
to enable. Which might be fine, I guess. I imagine the normal approach
would be to turn all warnings on anyway, and possibly fine-tune with
per-function directives if some functions do dangerous things on purpose.
3) You want to change the name to "enhanced_checks". I still think
the main feature is about displaying warnings to the user. I don't
particularly like this suggestion.

Regards,
Marko Tiikkaja

In response to

Responses

Browse pgsql-hackers by date

  From Date Subject
Next Message Pavel Stehule 2014-01-15 14:09:30 Re: plpgsql.warn_shadow
Previous Message Robert Haas 2014-01-15 13:55:02 Re: [Lsf-pc] Linux kernel impact on PostgreSQL performance