Closed mapninja closed 2 years ago
Might be related to proj and handling esri's code, as I assume they might be closed source or some other limiting factor related to them.
I can also confirm the issue on Windows with QGIS 3.22.10: when a vector layer (with any CRS) is exported as an ESRI Shapefile layer selecting the "ESRI:102999 - NAD_1983_2011_StatePlane_California_III_FIPS_0403" CRS, then the exported layer CRS is recognised by QGIS as "EPSG:6419 - NAD83(2011) / California zone 3" instead.
The exported ESRI Shapefile layer *.prj sidecar file contains the PROJCS["NAD_1983_2011_StatePlane_California_III_FIPS_0403",GEOGCS["GCS_NAD_1983_2011", ...
string.
The same issue occurs when a new empty ESRI Shapefile layer is created in QGIS selecting the "ESRI:102999 - NAD_1983_2011_StatePlane_California_III_FIPS_0403" CRS: the created layer is added in the map with the "EPSG:6419 - NAD83(2011) / California zone 3" CRS instead.
The issue doesn't occur if the layer is exported e.g. as GeoPackage, GeoJSON or FlatGeobuf: in this cases the exported layer CRS is recognised by QGIS as "ESRI:102999 - NAD_1983_2011_StatePlane_California_III_FIPS_0403"
It seems to me this is a GDAL/OGR and/or PROJ issue.
EPSG:6419 and ESRI:102999 are aliases of each other. When writing a shapefile .prj, ESRI WKT must be used, and the representation of them is identical, and ESRI WKT doesn't contain a AUTHORITY:CODE. When reading back, OGR will prefer the EPSG code to the ESRI one when 2 CRS match the ESRI WKT definition, because that's more interoperable So for me works as intented, and there's not much that can be done to change that behavior.
What is the bug or the crash?
Exporting a shapefile using
ESRI:102999
results in a shapefile withEPSG:6419
CRS defined.Steps to reproduce the issue
Exporting an
EPSG:4269
shapefile to a new shapefile usingESRI:102999 - NAD_1983_2011_StatePlane_California_III_FIPS_0403
as the output CRS results in a shapefile withEPSG:6419 - NAD83(2011) / California zone 3
as it's CRS. Not a show stopper, but definitely confusing and not the expected behavior.Discovered while writing this, so here is the full workflow: https://github.com/mapninja/Earthsys144/blob/master/Labs/Week_02/Coordinate_Systems_Geodesy.md
Versions
MacOS Monterey QGIS 3.26.3
Supported QGIS version
New profile
Additional context
No response