[gdal-dev] Advice on `GDALCreateAndReprojectImage` and destination no-data value.

[Javier Jimenez Shaw]

I do not know that code. But having a look I have the impression that you
have to set the nodata also to papszCreateOptions, so the created dataset
has it.

On Fri, 3 Nov 2023 at 21:36, Fitch, Simeon via gdal-dev <
gdal-dev at lists.osgeo.org> wrote:

> Hi,
>
> I'm one of the maintainers at the `georust/gdal` bindings project. 👋
>
> I'm currently working on exposing some of the GDAL Warp functionality in
> Rust, starting with a simple wrapper around `GDALCreateAndReprojectImage`.
> I'm having trouble getting the output file to include a no-data value in
> the band's metadata, despite the Warp operation properly respecting no-data
> values specified in `GDALWarpOptions`.
>
> My C++ skills are very "rusty", so I'm hoping I can get away with
> explaining my problem with some Rust code that is pretty close to being 1:1
> with the C API. My suspicion is I'm missing a step in the `GDALWarpOptions`
> setup.
>
> This is the CLI operation I'm trying to replicate:
>
>     gdalwarp -t_srs EPSG:4269 -dstnodata 255 -r near input.tif output.tif
>
> Here's the bare-bones Rust bindgen version:
>
>     <https://gist.github.com/metasim/d2a288d5b6126438aaa84b9e95707aff>
>
> The single band in `input.tif` is described by `gdalinfo` like this:
>
>     Band 1 Block=186x44 Type=Byte, ColorInterp=Gray
>       NoData Value=255
>       Metadata:
>         CLASSES=3
>
> After running the Rust code, the results are correct (including Warp
> numerically respecting source and destination no-data values) _except_
> `output.tif` does not have a no-data value in the band metadata. Here's
> what `gdalinfo` on `output.tif` shows:
>
>     Band 1 Block=210x39 Type=Byte, ColorInterp=Gray
>
> The only way I've been able to get the behavior I need is to open the
> written dataset from the filesystem and set the no-data value explicitly in
> the band, which is inefficient and awkward. Given the CLI version does the
> right thing, it's obviously possible. I've spent several hours visually
> inspecting the C++ source code, but the CLI version doesn't call
> `GDALCreateAndReprojectImage` directly; rather it constructs the Warp
> pipeline itself.
>
> Anyone have tips that might help me run this down? Am I missing a no-data
> handling step required when working with Warp routines? Could
> `GDALCreateAndReprojectImage` have a bug?
>
> Thanks,
>
> Simeon
>
> --
> Simeon Fitch
> Co-founder & CTO
> Astraea, Inc.
>
>
> The content of this email is intended for the person or entity to which it
> is addressed only. This email may contain confidential information. If you
> are not the person to whom this message is addressed, be aware that any
> use, reproduction, or distribution of this message is strictly prohibited.
> If you received this in error, please contact the sender and immediately
> delete this email and any attachments.
> _______________________________________________
> gdal-dev mailing list
> gdal-dev at lists.osgeo.org
> https://lists.osgeo.org/mailman/listinfo/gdal-dev
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.osgeo.org/pipermail/gdal-dev/attachments/20231103/e0d6ac74/attachment.htm>