Re: Collation versioning - Mailing list pgsql-hackers

From Julien Rouhaud
Subject Re: Collation versioning
Date
Msg-id 20200311084427.GA12358@nol
Whole thread Raw
In response to Re: Collation versioning  (Thomas Munro <thomas.munro@gmail.com>)
Responses Re: Collation versioning
List pgsql-hackers
On Wed, Mar 11, 2020 at 04:13:17PM +1300, Thomas Munro wrote:
> On Wed, Mar 4, 2020 at 10:01 PM Julien Rouhaud <rjuju123@gmail.com> wrote:
> > On Thu, Feb 27, 2020 at 08:45:35AM +0100, Julien Rouhaud wrote:
> > > On Thu, Feb 27, 2020 at 04:10:14PM +1300, Thomas Munro wrote:
> > > > On Thu, Feb 27, 2020 at 3:29 AM Julien Rouhaud <rjuju123@gmail.com> wrote:
> > > > > [v10]
> > > [...]
> > >
> > > Fixed in v11 by changing pg_collation_actual_version() to handle default
> > > collation too, as it seems a better behavior.
> >
> > Rebased v12 due to conflict with the recent command tag commit (2f9661311b), no
> > other changes.
>
> I'm still reviewing and testing the code, but here's a new patch set
> with a bunch of small changes to documentation, tests and commit
> messages:


Thanks a lot Thomas!  FTR one of the patch has a minor conflict due to
3c173a53a82, so I'm attaching a v14 to make cfbot happy and also benefit from
windows coverage just in case.  I also tried to add some missing reviewers to
the patches, I hope I didn't forget anyone.

>
> 0001-Remove-pg_collation.collversion-v13.patch
>  * removed ALTER COLLATION ... REFRESH VERSION from collate.linux.utf8.sql
>    (oops, that wasn't running on my FreeBSD system, or on cfbot's Ubuntu
>    image because it doesn't have "locales-all" installed; I will make a
>    note to fix that on cfbot)


Wow I don't know how that happened since I did remove that in v2 ([1]). Thanks!

>
> 0002-Add-pg_depend.refobjversion-v13.patch
>  * added new column to the documentation of the pg_depend catalog
>
> 0003-Implement-type-regcollation-v13.patch
>  * added regcollation to a list of such types in datatype.sgml
>
> 0004-Track-collation-versions-for-indexes-v13.patch
>  * added a note to the documentation's list of reasons why REINDEX
> might be needed
>
> 0005-Preserve-index-dependencies-on-collation-during--v13.patch
>  * minor rewording of the docs, I hope you like it
>
> 0006-Add-ALTER-INDEX-.-ALTER-COLLATION-.-REFRESH-VERS-v13.patch
>  * minor rewording of the docs, I hope you like it
>
> 0007-doc-Add-Collation-Versions-section-v13.patch
>  * new: a brief introduction to this topic


I'm quite happy with all the changes, and especially the new section about
collation version in 0007.

Minor gripe in pgugprade.sgml, as Peter E. mentioned in [2] the documentation
shouldn't focus on pg13- / pg13+ upgrade only.  Or should we keep that for when
additional system will be supported post pg13?

[1] https://www.postgresql.org/message-id/CAOBaU_aL%3D4f67L%2Bm2s28DmiaacZ%3DDn75BZY-HGmEq1WquGa-Jg%40mail.gmail.com
[2] https://www.postgresql.org/message-id/3ede40e7-5799-7096-dc5f-d7beda8e7145%402ndquadrant.com

Attachment

pgsql-hackers by date:

Previous
From: Prabhat Sahu
Date:
Subject: Re: [Proposal] Global temporary tables
Next
From: Amit Kapila
Date:
Subject: Re: [HACKERS] Moving relation extension locks out of heavyweight lock manager