[postgis-tickets] [PostGIS] #4825: Memory Leak with ST_MakeValid() and invalid geometry?

PostGIS trac at osgeo.org
Tue Jan 5 05:49:49 PST 2021


#4825: Memory Leak with ST_MakeValid() and invalid geometry?
----------------------+---------------------------
  Reporter:  neumann  |      Owner:  pramsey
      Type:  defect   |     Status:  new
  Priority:  medium   |  Milestone:  PostGIS 3.1.1
 Component:  postgis  |    Version:  master
Resolution:           |   Keywords:
----------------------+---------------------------

Comment (by strk):

 Running the query under valgrind against PotgreSQL-10,
 PostGIS-3.1.0rc1-17-g591a307ae and GEOS-3.10.0dev-CAPI-1.15.0 reports no
 problems. I suggest the following steps:

   1. Further reduce the test dataset by _removing_ records where
 ST_IsValid(geometrie) -- does the problem persist after this step ?
   2. Further reduce the test dataset by ONLY including {{{id}}} and
 {{{geom}}} as the columns, and provide no defaults (also helps restoring
 the dump)
   3. Report here what differences you see in the first and subsequent runs
 of the query

 I tried spotting differences in ST_MakeValid output using this approach:
 {{{
 CREATE TABLE test(id int, g geometry, vg1 geometry, vg2 geometry);
 INSERT INTO test(id, g) SELECT t_id, geometrie FROM
 arp_npl.erschlssngsplnung_erschliessung_linienobjekt;
 UPDATE test set vg1 = ST_MakeValid(g);
 UPDATE test set vg2 = ST_MakeValid(g);
 SELECT id FROM test WHERE NOT ST_Equals(vg1, vg2);
 }}}

 I get not results from last query, do you ?

-- 
Ticket URL: <https://trac.osgeo.org/postgis/ticket/4825#comment:6>
PostGIS <http://trac.osgeo.org/postgis/>
The PostGIS Trac is used for bug, enhancement & task tracking, a user and developer wiki, and a view into the subversion code repository of PostGIS project.


More information about the postgis-tickets mailing list