#3543 closed defect (fixed)
Upcoming PDS raster not correctly read
Reported by: | michaelaye | Owned by: | warmerdam |
---|---|---|---|
Priority: | normal | Milestone: | |
Component: | GDAL_Raster | Version: | unspecified |
Severity: | normal | Keywords: | PDS |
Cc: |
Description
The DAWN Framing Camera (Mission to Asteroids) will soon release data in PDS format, an example is attached. The errors vary, depending on the header and formats implemented. For the attached file the failure mode is, that the read data becomes of dtype int16, where it should have been uint16.
Change History (5)
comment:1 by , 14 years ago
Status: | new → assigned |
---|
comment:2 by , 14 years ago
Resolution: | → fixed |
---|---|
Status: | assigned → closed |
I encountered three immediate problems, fixed in trunk (r19475)
- support quoted SAMPLE_TYPE.
- Check for UNSIGNED in SAMPLE_TYPE for UInt16
- support files where scanlines are broken over several records.
Note that we still only support the first image while the file contains several other pseudo-image products.
This image has been placed at:
http://download.osgeo.org/gdal/data/pds/FC11A0000338_07344220211F2C.IMG
for reference.
comment:3 by , 14 years ago
A suggestion/question: If it is true, that PDS does not support bands anyway, could the other binaries (outer rims of the CCD with different lighting conditions, e.g. refractive light, only bias etc.) maybe support by the driver as bands?
comment:4 by , 14 years ago
Michael,
We can't represent these auxiliary products as bands because they are not the same size as the base image. If we want to support them we would need to go to the subdataset approach used for hdf, netcdf, etc. This would be fairly involved.
comment:5 by , 14 years ago
Keywords: | PDS added |
---|
The header text looks like this: