[postgis-users] St_makevalid issue
Paul van der Linden
paul.doskabouter at gmail.com
Thu Mar 26 10:54:19 PDT 2020
> > Any advice on how to further debug this?
>
> IF you know docker ...
> THEN there are some postgis docker images for testing
> so you can test your code/data with different environments easily.
>
> https://github.com/postgis/docker-postgis
> - postgis/postgis:12-master ( Latest postgis,geos )
>
> IF "postgis/postgis:12-master" is OK ,
> THEN this is fixed in somewhere ; good news , you have to upgrade your
> environment,
> or you can test with different images ( postgis/postgis:11-2.5
> )
> ELSE - please create a minimal working example[1] and probably somebody
> will debug.
>
> [1] https://en.wikipedia.org/wiki/Minimal_working_example
>
> imre
I'm really not able to do any upgrades, so that's out of the question
And the sample below is already the minimum to get that behavior. The
inconsistency between runs makes it impossible to pinpoint on one or 2
geometries.
No experience with docker though
>
> Imre Samu <pella.samu at gmail.com> ezt írta (idopont: 2020. márc. 26.,
Cs,
> 17:26):
>
> > > Doesn't anybody have an idea? Or is at least curious?
> > >"POSTGIS="2.5.1 r17027" [EXTENSION] PGSQL="110" GEOS="3.7.0-CAPI-1.11.0
> > 3.7.1"
> >
> > imho: try to update your environment with the latest patches/fixes and
> > please re-check.
> > Postgis: from 2.5.1 -> to 2.5.4 (
> > https://postgis.net/2020/02/28/postgis-2.5.4/
> > https://git.osgeo.org/gitea/postgis/postgis/raw/tag/2.5.4/NEWS )
> > Geos: from 3.7.0 -> to the latest (3.8.1 ) (
> > https://github.com/libgeos/geos/blob/master/NEWS )
> >
> > Imre
> >
> > Paul van der Linden <paul.doskabouter at gmail.com> ezt írta (idopont:
2020.
> > márc. 26., Cs, 15:54):
> >
> >> Doesn't anybody have an idea? Or is at least curious?
> >>
> >> Especially the select st_isvalid(geo) where not st_valid returning a
> >> "true" row?
> >>
> >> >I have a query that gives inconsistent results.
> >> >It almost seems like st_makevalid doesn't always produce the same
results
> >> >given the same input, and sometimes even returns invalid geometries...
> >> >
> >> >Results are
> >> >f;1159320407;"Bangladesh"
> >> >f;1159321073;"Mozambique"
> >> >f;1159316737;"Nord"
> >> >f;1159314827;"Uusimaa"
> >> >
> >> >second run:
> >> >f;1159321073;"Mozambique"
> >> >t;1159315455;"Syddanmark"
> >> >t;1159314827;"Uusimaa"
> >> >t;1159317795;"Gerona"
> >> >
> >> >third run:
> >> >f;1159321073;"Mozambique"
> >> >t;1159315455;"Syddanmark"
> >> >
> >> >note the 't' in the second and third run...
> >> >
> >> >Query:
> >> >
> >> >SELECT ST_isvalid(simplifiedgeo2),* FROM (
> >> >SELECT
> >> > foo2.ne_id,
> >> > foo2.name,
> >> > CASE
> >> > WHEN ST_isvalid(foo2.simplifiedgeo) THEN foo2.simplifiedgeo
> >> > ELSE ST_makevalid(foo2.simplifiedgeo)
> >> > END AS simplifiedgeo2
> >> > FROM (
> >> > SELECT
> >> > ST_simplifypreservetopology(foo.geom,
> >> > LEAST(
> >> > ST_xmax(foo.shiftedgeo::box3d) -
> >> ST_xmin(foo.shiftedgeo::box3d),
> >> >ST_ymax(foo.shiftedgeo::box3d) - ST_ymin(foo.shiftedgeo::box3d))
> >> > ) AS simplifiedgeo,
> >> > foo.ne_id,
> >> > foo.name,
> >> > foo.geom
> >> > FROM (
> >> > SELECT
> >> > natearth.ne_id,
> >> > natearth.geom,
> >> > natearth.name,
> >> > CASE
> >> > WHEN (
> >> > ST_xmax(ST_shiftlongitude(natearth.geom)::box3d) -
> >> >ST_xmin(ST_shiftlongitude(natearth.geom)::box3d))
> >> > <
> >> > (ST_xmax(natearth.geom::box3d) -
> >> >ST_xmin(natearth.geom::box3d)
> >> > ) THEN ST_shiftlongitude(natearth.geom)
> >> > ELSE natearth.geom
> >> > END AS shiftedgeo
> >> > FROM (
> >> > SELECT
> >> > ne_10m_admin_0_countries_lakes.ne_id,
> >> > ne_10m_admin_0_countries_lakes.geom,
> >> > ne_10m_admin_0_countries_lakes.name
> >> > FROM natural_earth.ne_10m_admin_0_countries_lakes
> >> > UNION ALL
> >> > SELECT
> >> > ne_10m_admin_1_states_provinces_lakes.ne_id,
> >> > ne_10m_admin_1_states_provinces_lakes.geom,
> >> > ne_10m_admin_1_states_provinces_lakes.name
> >> > FROM natural_earth.ne_10m_admin_1_states_provinces_lakes
> >> > ) natearth
> >> > ) foo
> >> > ) foo2
> >> >) as fff
> >> >WHERE NOT ST_isvalid(simplifiedgeo2)
> >> >
> >> >Any advice on how to further debug this?
> >> >
> >> >"POSTGIS="2.5.1 r17027" [EXTENSION] PGSQL="110"
GEOS="3.7.0-CAPI-1.11.0
> >> >3.7.1" PROJ="Rel. 4.9.3, 15 August 2016" GDAL="GDAL 2.2.4, released
> >> >2018/03/19 GDAL_DATA not found" LIBXML="2.7.8" LIBJSON="0.12"
> >> >LIBPROTOBUF="1.2.1" (core procs from "2.5.0 r16836" nee (...)"
> >> >"PostgreSQL 11.2, compiled by Visual C++ build 1914, 64-bit"
> >> >
> >> >Running on windows 64-bit.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.osgeo.org/pipermail/postgis-users/attachments/20200326/24817270/attachment.html>
More information about the postgis-users
mailing list