Even Rouault

@tbonfort Given the last answer in the libjxl ticket "-Would the encoder be allowed to reorder bands ? - I don't think the encoder should be allowed to do that,...

> Thoughts? LGTM. (Is there a standard way for Python doc, similar to Doxygen for C/C++, to document parameters and return values ? I somehow remember the answer is "not...

for the cppcheck warnings about unused private members (https://github.com/OSGeo/gdal/runs/7781642042?check_suite_focus=true), you can use patterns like ``` CPL_IGNORE_RET_VAL(poDS->sHeader.foo); ```

> GetSpatialRef (gdalinfo crashes though) The crash is completely unrelated: ``` $ valgrind gdalinfo nt_20220409_f18_nrt_s.bin ==701859== Invalid write of size 4 ==701859== at 0x5387E0C: NSIDCbinRasterBand::GetNoDataValue(int*) (nsidcbindataset.cpp:185) ==701859== by 0x5A98616: GDALNoDataMaskBand::GDALNoDataMaskBand(GDALRasterBand*)...

https://github.com/OSGeo/gdal/runs/7857751581?check_suite_focus=true: ``` /__w/gdal/gdal/doc/source/drivers/raster/nsidcbin.rst:55: WARNING: Literal block ends without a blank line; unexpected unindent. /__w/gdal/gdal/doc/source/drivers/raster/nsidcbin.rst:60: ERROR: Unexpected indentation. /__w/gdal/gdal/doc/source/drivers/raster/nsidcbin.rst:66: WARNING: Definition list ends without a blank line; unexpected unindent. /__w/gdal/gdal/doc/source/drivers/raster/nsidcbin.rst:89: WARNING:...

We now have linting rules of autotest python files You should install & setup the pre-commit utility with (once for all for the repository): ``` python -m pip install pre-commit...

> I'd still like this legacy driver to be included if there are significant amount of past data using the legacy format and likely not being converted and published into...

@mdsumner any update ?

> I'd still like it to be included in a future GDAL release ok, makes sense

I'm happy with the principle of partial PRs if they are not too intrusive