[PATCH] Radarsat-2 subdataset calibration not applied correctly

[aklein]

In RS2CalibRasterBand::IReadBlock, the look up table index is calculated using block offset instead of pixel offset. Additionally, the scanline length for reading byte data is incorrect. Updated code attached.

[aklein]

calibration fixes to r2 driver for gdal 1.8.1

[Even Rouault]

Hum, from my non-specialist eye, it looks reasonable, but unfortunately I have no test data to check the correctness of the patch, so I'm a bit hesitant to apply. Maybe Frank or Phil will be willing to have a look at this (I've attached the patch updated against trunk)

I have a few observations however :

• the patch assumes that the width of the raster is a multiple of the block width. Perhaps there should be some provision if the last block (at the right) is partial (like it is done for the y direction)
• there's no test to check if the number of elements of m_nfTable is the same as nRasterXSize, which according to this patch must be the case, otherwise bad things will happen.

[Even Rouault]

Patch against trunk r23569

[pvachon]

The patch looks OK, but give me a day to review this (I'll probably look at it tonight) -- I think there might need to be a bit more "love" given to this code at a glance.

P.

[Even Rouault]

This ticket has been automatically closed because Trac is no longer used for GDAL bug tracking, since the project has migrated to GitHub. If you believe this ticket is still valid, you may file it to ​https://github.com/OSGeo/gdal/issues if it is not already reported there.