[Geoserver-devel] [jira] Created: (GEOS-2225) KML regionating does not return the top features in big datasets

KML regionating does not return the top features in big datasets
----------------------------------------------------------------

                 Key: GEOS-2225
                 URL: http://jira.codehaus.org/browse/GEOS-2225
             Project: GeoServer
          Issue Type: Bug
    Affects Versions: 1.7.0-RC2
            Reporter: Arne Kepp
            Assignee: Arne Kepp
             Fix For: 1.7.1

I'm not sure whether this qualifies as a 1.7.0 bug, so scheduling against 1.7.1. Consider this a scratchpad to organize my thoughts.

If I use the NYC buildings data in a versioned postgres datastore, with height attributes set (most buildings are 0, 160 of 950 000 have positive heights) and sorting externally in Postgresql, the top KML tiles do not contain any data. The data can be downloaded here: http://geo.openplans.org/~ak/democd_db_dump.tbz2
- I have verified that the H2 database is built correctly with the expected feature ids in the correct tiles (all the way up)
- The code does appear to build a (reasonable) list of features to exclude from tiles, based on the H2 database
- It does not matter whether the feature type is configured to return 150 or 5 features per tile.

- sf:streams, also in a versioned postgis database, does not exhibit this problem at all, so it's unlikely to be a direct consequence of wfsv.

--
This message is automatically generated by JIRA.
-
If you think it was sent incorrectly contact one of the administrators: http://jira.codehaus.org/secure/Administrators.jspa
-
For more information on JIRA, see: http://www.atlassian.com/software/jira