[postgis-devel] Precision Reduction?

[Regina Obe]

-1 – please no more questionable functions  we might need to take out in 3.1.0 release.

 

Let’s be content with what we have already with the new gridSize args strk introduced and focus on bullet testing the geos 3.9.

 

The reality is a good chunk of people aren’t going to be using GEOS 3.9 in PostGIS 3.1 so you are creating a bunch of features that few can use and that yikes you might kick yourself in the ass when you realize you don’t like the new names you came up with.

Let’s just put all this dreamy talk into the 3.2 release milestone. PLEASE

 

Thanks,

Regina

 

From: postgis-devel [mailto:postgis-devel-bounces at lists.osgeo.org] On Behalf Of Darafei "Kom?pa" Praliaskouski
Sent: Wednesday, November 11, 2020 11:53 AM
To: PostGIS Development Discussion <postgis-devel at lists.osgeo.org>
Subject: Re: [postgis-devel] Precision Reduction?

 

Regarding naming: we already have duality of ST_Simplify <-> ST_SimplifyPreserveTopology, so if we want to follow it as an agreement, it has to be ST_SnapToGridPreserveTopology. If that looks wrong then we may want to consider the exposure of this function together with naming ST_SimplifyPreserveTopology.

 

Another recent precision reducer we have is ST_QuantizeCoordinates. Can GEOS be bent to provide the same behavior on a non-even grid with last bits set to 0 with guarantees on validity of output?

 

On Wed, Nov 11, 2020 at 7:27 PM Martin Davis <mtnclimb at gmail.com <mailto:mtnclimb at gmail.com> > wrote:

I am neutral about encoding precision in the data or in the schema.  As Darafei points out it raises a raft of tricky questions about what should be made precise, the semantics of spatial predicates, and what to do about mixed precision. It's a far bigger change than what Paul is asking about, since it requires many operations to be enhanced to support precise output.

 

Providing ST_PrecisionReduce may or may not be a small step in that direction.  But it lets users and the development team get some experience with how and when to enforce precision. 

 

On Wed, Nov 11, 2020 at 2:06 AM Sandro Santilli <strk at kbt.io <mailto:strk at kbt.io> > wrote:

I think we should add a precision field to the Geometry type
so that once precision is set on a geometry it sticks and can
be passed around, in particular to GEOS functions for retaining
the precision in output (basically using fixed precision all
the way).

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




 

-- 

Darafei "Komяpa" Praliaskouski
OSM BY Team - http://openstreetmap.by/

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.osgeo.org/pipermail/postgis-devel/attachments/20201111/6bc481be/attachment.html>