Re: Move bki file pre-processing from initdb to bootstrap - Mailing list pgsql-hackers

From Krishnakumar R
Subject Re: Move bki file pre-processing from initdb to bootstrap
Date
Msg-id CAPMWgZ-KFTaHbdsGHim86d39TGrOafPuUtBFOC47Ank7vNYfGQ@mail.gmail.com
Whole thread Raw
In response to Re: Move bki file pre-processing from initdb to bootstrap  (Peter Eisentraut <peter@eisentraut.org>)
Responses Re: Move bki file pre-processing from initdb to bootstrap
List pgsql-hackers
Thank you for review, Peter.

Makes sense on the split part. Was starting to think in same lines, at the end of last iteration. Will come back shortly. 

On Fri, Nov 10, 2023 at 12:48 AM Peter Eisentraut <peter@eisentraut.org> wrote:
On 17.10.23 03:32, Krishnakumar R wrote:
>> The version comparison has been moved from initdb to bootstrap. This
>> created some compatibility problems with windows tests. For now I kept
>> the version check to not have \n added, which worked fine and serves
>> the purpose. However hoping to have something better in v3 in addition
>> to addressing any other comments.
>
> With help from Thomas, figured out that on windows fopen uses binary
> mode in the backend which causes issues with EOL. Please find the
> attached patch updated with a fix for this.

I suggest that this patch set be split up into three incremental parts:

1. Move some build-time settings from initdb to postgres.bki.
2. The database locale handling.
3. The bki file handling.

Each of these topics really needs a separate detailed consideration.

pgsql-hackers by date:

Previous
From: Christoph Berg
Date:
Subject: Re: pgsql: Don't trust unvalidated xl_tot_len.
Next
From: "Tristan Partin"
Date:
Subject: Re: Failure during Building Postgres in Windows with Meson