[postgis-users] Help shaping the future: how do your use of spatial_ref_sys ?

[Paul Ramsey]

> On Feb 10, 2022, at 7:10 AM, Jim Klassen <klassen.js at gmail.com> wrote:
> 
> On 2/9/22 03:31, Sandro Santilli wrote:
>> Would you be disappointed/surprised by a PostGIS upgrade
>> replacing your custom entries above 200000 with some new
>> "official" entry ? What strategy did you set in place to
>> prevent this ?
> 
> I missed the second part in my first reply.
> 
> As it stands it would break things. The strategy at the time (again this was decided before 2010) was there weren't any EPSG codes in that range and the supplied spatial_ref_sys table at the time appeared to just mirror the EPSG codes.
> 
> How do other people handle conflicting srids?

My guess is that it's pretty rare, since most people who have defined custom ranges for operational use have done what you did, and looked at the ranges in use and given them a wide berth.

The exception, it seems to me, would be people who have found ERRORS in the epsg-derived entries in spatial_ref_sys, and rather than defining a custom, fixed entry on another srid, have changed the epsg entry in place. For those folks, updates will always be a bit fraught, since the reasonable expectation of most people (that updates bring along the latest-and-greatest data from epsg) will conflict with their desire (that the things they changed be kept static, and everything else updated).

My guess is that this second category is quite small, perhaps vanishingly so, and that that a more general "update all system entries" policy would be both easy to explain and not too hard to implement.

P