Re: [PATCH] Support % wildcard in extension upgrade filenames - Mailing list pgsql-hackers

From Sandro Santilli
Subject Re: [PATCH] Support % wildcard in extension upgrade filenames
Date
Msg-id 20230308122736.e2mx4e3paxbabcnw@c19
Whole thread Raw
In response to Re: [PATCH] Support % wildcard in extension upgrade filenames  (Robert Haas <robertmhaas@gmail.com>)
Responses Re: [PATCH] Support % wildcard in extension upgrade filenames
List pgsql-hackers
On Tue, Mar 07, 2023 at 12:39:34PM -0500, Robert Haas wrote:

> The thing that confuses me here is why the PostGIS folks are ending up
> with so many files.

We want to allow PostGIS users to upgrade from ANY previous version,
because different distribution or user habit may result in people
upgrading from ANY older release to the newest.

Now, PostGIS has released a total of 164 versions:

    curl -s https://download.osgeo.org/postgis/source/ |
      grep -c '.tar.gz<'

In *addition* to these, most developers end up installing a "dev"
suffixed version to distinguish it from the official releases,
which means there's a total of 164*2 = 328 version *per extension*.

Now, PostGIS doesn't install a single extension but multiple ones:

  - postgis
  - postgis_raster
  - postgis_topology
  - postgis_sfcgal
  - postgis_tiger_geocoder
  - address_standardizer

So we'd be up to 328 * 6 = 1968 files needed to upgrade from ANY
postgis extension version to CURRENT version.

The numbers are slightly less because not all versions of PostGIS
did support EXTENSION (wasn't even available in early PostGIS versions)

> We certainly don't have that problem with the
> extension that are being maintained in contrib, and I guess there is
> some difference in versioning practice that is making it an issue for
> them but not for us. I wish I understood what was going on there.

Our extension versioning has 3 numbers, but PostgreSQL doesn't really
care about this right ? It's just that we've had 328 different
versions released and more are to come, and we want users to be able
to upgrade from each one of them.

I guess you may suggest that we do NOT increas the *EXTENSION* version
number UNLESS something really changes at SQL level, but honestly I
doubt we EVER released a version with no changes at the SQL level
(maybe in some occasion for really bad bugs which were only fixed at
the C level).

--strk;



pgsql-hackers by date:

Previous
From: "Drouvot, Bertrand"
Date:
Subject: Re: Track IO times in pg_stat_io
Next
From: Sandro Santilli
Date:
Subject: Re: [PATCH] Support % wildcard in extension upgrade filenames