[postgis-devel] PSC Vote: Make spatial_ref_sys as a view

[Paul Ramsey]

> On Feb 7, 2022, at 3:02 PM, Sandro Santilli <strk at kbt.io> wrote:
> 
> On Mon, Feb 07, 2022 at 02:35:27PM -0800, Paul Ramsey wrote:
>>> On Feb 7, 2022, at 2:27 PM, Sandro Santilli <strk at kbt.io> wrote:
>>> 
>>>  3.2.1dev 3.2.0-28-g4443d6989
>> 
>> POSTGIS="3.2.1dev 3.2.0-339-gafaebf921" [EXTENSION] PGSQL="140" GEOS="3.11.0dev-CAPI-1.16.0" PROJ="8.1.0" LIBXML="2.9.4" LIBJSON="0.15" LIBPROTOBUF="1.4.0" WAGYU="0.5.0 (Internal)"
> 
> There seem to be NO CHANGE between mine and yours spatial_ref_sys.sql
> in the source repository:
> 
>  git diff afaebf921 4443d6989 spatial_ref_sys.sql # none
> 
>> But I "see" now they are in fact different. There's two one-character differences in the SRTEXT.
>> 
>> So mystery solved.
> 
> Well, next mystery is WHY there are those one-character
> differences if the spatial_ref_sys.sql in the repository
> if it hadn't changed between your version and my version ?
> 

Checking the spatial_ref_sys.sql in the 3.2 branch and master, the difference is right there in the files. Someone clever re-generated the spatial_ref_sys in master and changed the multiple INSERTS into one huge multi-valued insert. Either during that process, or just because it's regenerated and the EPSG data is slightly updated, in (for example) 4893, the string "USA - CONUS and Alaska; PRVI" has been changed to "USA - CONUS and Alaska, PRVI" in the master version.

So... the update is detecting the system-side changes as user changes and preserving them? I mean, that seems impossible to do practically, if the user actually ALTERS the system codes, there's no way to distinguish between changes the user created and changes the EPSG has introduced.

P.



> --strk;
> _______________________________________________
> postgis-devel mailing list
> postgis-devel at lists.osgeo.org
> https://lists.osgeo.org/mailman/listinfo/postgis-devel