Opened 15 years ago
Closed 15 years ago
#2888 closed defect (invalid)
ogr2ogr conversion with EPSG:28992 not correctly
| Reported by: | oxygen | Owned by: | warmerdam |
|---|---|---|---|
| Priority: | high | Milestone: | |
| Component: | Utilities | Version: | unspecified |
| Severity: | major | Keywords: | FWTools (win) 2.2.8 |
| Cc: |
Description
Conversion between KML and ESRI shapefile (and back) using EPSG:28992 results in wrong placement.
We used a correct KML (in Google Earth / Maps) converted it into ESRI shape, read it in Quantum GIS with a Dutch WMS server We used a correct ESRI Shapefile converted into KML and viewed it in Google Earth.
Displacement is approx. 67 meter.
Change History (2)
comment:1 by , 15 years ago
comment:2 by , 15 years ago
| Resolution: | → invalid |
|---|---|
| Status: | new → closed |
Given Even's analysis this is not a GDAL problem. It is a user coordinate system specification issue.
I would add it is also not a PROJ.4 problem so don't file a ticket there unless it is one providing a compelling case to override the datum definition associated with EPSG:28992 with a particular towgs84 solution with lots of supporting information on why that is the right single choice for everyone in the world.
Seems to be a proj4 related issue or use issue... I see that the proj4 definition for EPSG:28992 use the Bessel ellipsoid, but does not specify a datum. So as you use PROJ4.6 or later (FWTools 2.2.8 ships with 4.6.1), no ellipsoid to ellipsoid will be made. My guess would be that it is exactly the issue discussed in http://n2.nabble.com/GDAL---PROJ.4-Transformation-(cs2cs)---several-results-by-%C2%A0PROJ.4-Versions-(DHDN-GK2-%3D%3E-ETRS89-UTM32)-td2466135.html.