Date: Fri, 21 Apr 2023 13:44:51 -0700
From: Nathan Bossart <nathandbossart@gmail.com>
To: Daniel Gustafsson <daniel@yesql.se>
Cc: Yurii Rashkovskii <yrashk@gmail.com>,
	pgsql-hackers@lists.postgresql.org
Subject: Re: [PATCH] Extend the length of BackgroundWorker.bgw_library_name
Message-ID: <20230421204451.GC1440148@nathanxps13>
References: <CA+RLCQyjFV5Y8tG5QgUb6gjteL4S3p+1gcyqWTqigyM93WZ9Pg@mail.gmail.com>
 <20230313173528.GA195204@nathanxps13>
 <D32826BC-97A9-4D3D-8BD7-3639044F8D4B@yesql.se>
 <20230420233243.GB1437940@nathanxps13>
 <D326A802-7DDF-4B65-9AC7-1083F10294D5@yesql.se>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <D326A802-7DDF-4B65-9AC7-1083F10294D5@yesql.se>
Archived-At: <https://www.postgresql.org/message-id/20230421204451.GC1440148%40nathanxps13>
Precedence: bulk

On Fri, Apr 21, 2023 at 10:49:48AM +0200, Daniel Gustafsson wrote:
> On 21 Apr 2023, at 01:32, Nathan Bossart <nathandbossart@gmail.com> wrote:
>> I am -0.5 for this.  If you are writing a new background worker, it's
>> probably reasonable to expect that you can locate the definition of
>> BGW_MAXLEN.  
> 
> Of course.  The question is if it's a helpful addition for someone who is
> reading the documentation section on implementing background workers where we
> explicitly mention BGW_MAXLEN without saying what it is.

IMHO it's better to have folks use the macro so that their calls to
snprintf(), etc. are updated when BGW_MAXLEN is changed.  But I can't say
I'm strongly opposed to adding the value to the docs if you think it is
helpful.

>> Also, I think there's a good chance that we'd forget to update
>> such documentation the next time we adjust it.
> 
> There is that, but once set to MAXPGPATH it seems unlikely to change
> particularly often so it seems the wrong thing to optimize for.

True.

-- 
Nathan Bossart
Amazon Web Services: https://aws.amazon.com