L1B driver/gdalwarp: coordinate system should be defined

[Markus Neteler]

Frank,

as being curious if the new gdalwark manages the AVHRR reprojection
better than GRASs, I tried on a L1B file:

gdalinfo A1726874.L2460424 |head
Driver: L1B/NOAA Polar Orbiter Level 1b Data Set
Size is 2048, 269
Coordinate System is `'
Origin = (0.000000,0.000000)
Pixel Size = (1.000000,1.000000)
GCP Projection = GEOGCS["WGS 72",DATUM["WGS_1972",SPHEROID["WGS
72",6378135,298.26,AUTHORITY["EPSG",7043]],TOWGS84[0,0,4.5,0,0,0.554,0.2263],AUTHORITY["EPSG",6322]],PRIMEM["Greenwich",0,AUTHORITY["EPSG",8901]],UNIT["degree",0.0174532925199433,AUTHORITY["EPSG",9108]],AXIS["Lat","NORTH"],AXIS["Long","EAST"],AUTHORITY["EPSG",4322]]
GCP[  0]: Id=(null), Info=
          (2022.5,268.5) -> (35.8455,47.5516,0)
GCP[  1]: Id=(null), Info=
          (1982.5,268.5) -> (33.9067,47.5479,0)
[...]

the gdalwarp:

gdalwarp -order 3 -of gtiff A1726874.L2460424 A1726874.L2460424.tif
Creating output file is that 2059P x 452L.
:0...10...20...30...40...50...60...70...80...90...100 - done.

gdalinfo A1726874.L2460424.tif
Driver: GTiff/GeoTIFF
Size is 2059, 452
Coordinate System is `'
Origin = (-0.642078,50.420825)
Pixel Size = (0.017700,-0.017700)
Corner Coordinates:
Upper Left  (  -0.6420778,  50.4208248)
Lower Left  (  -0.6420778,  42.4205880)
Upper Right (  35.8014789,  50.4208248)
Lower Right (  35.8014789,  42.4205880)
Center      (  17.5797006,  46.4207064)
Band 1 Block=2059x1 Type=UInt16, ColorInterp=Gray
Band 2 Block=2059x1 Type=UInt16, ColorInterp=Gray
Band 3 Block=2059x1 Type=UInt16, ColorInterp=Gray
Band 4 Block=2059x1 Type=UInt16, ColorInterp=Gray
Band 5 Block=2059x1 Type=UInt16, ColorInterp=Gray

While gdalinfo sees the coordinate system, gdalwarp seems
to forget about it (in this case only?).

That renders the gdalwarp result partially unusable for
import into a GIS. If possible, please allow gdalwarp to
copy the coordinate system definition from source to target.

Ah, maybe copying is wrong but it should use the "GCP Projection"
string from the source as "coordinate system" for the target.

Thanks in advance,

 Markus

[warmerdam]

Markus, 

Your analysis is quite correct.  I have fixed the bug. 

Best regards,