Added `no_data_value_u64`, `set_no_data_value_u64`, `no_data_value_i64` and `set_no_data_value_i64` to `RasterBand` #520

metasim · 2024-02-01T15:45:23Z

I agree to follow the project's code of conduct.
I added an entry to CHANGES.md if knowledge of this change could be valuable to users.

Adds RasterBand::no_data_value_u64 and RasterBand::no_data_value_i64, etc.

lnicola · 2024-02-01T15:47:54Z

Can you resample those to 1x1 and maybe merge them into a single file?

src/raster/rasterband.rs

metasim · 2024-02-01T15:51:55Z

Just realized I should add the setters too. Not enough coffee this morning...

lnicola · 2024-02-01T15:55:08Z

Good idea, I was actually wondering if there's any way to set it. Might make sense to use a larger value then, like 2^62.

metasim · 2024-02-01T15:55:36Z

Can you resample those to 1x1 and maybe merge them into a single file?

How about a 1x2 so we can have a data value and a no-data value?

CHANGES.md

metasim · 2024-02-01T15:56:46Z

Good idea, I was actually wondering if there's any way to set it. Might make sense to use a larger value then, like 2^62.

I basically neglected to see these

lnicola · 2024-02-01T15:57:46Z

Doesn't seem necessary to me, but sure it doesn't matter. Could come in handy in the future, who knows?

metasim · 2024-02-01T16:02:12Z

Doesn't seem necessary to me...

What are you referring to here? The comment sequence is confusing.

lnicola · 2024-02-01T16:07:20Z

I meant that about the second pixel with a valid value. I don't think it's needed to test these functions, but they could be useful in a simple warping/resampling test.

lnicola · 2024-02-01T19:12:46Z

src/raster/rasterband.rs

+    /// [`GDALDeleteRasterNoDataValue`](https://gdal.org/api/raster_c_api.html#_CPPv427GDALDeleteRasterNoDataValue15GDALRasterBandH)
+    #[cfg(all(major_ge_3, minor_ge_5))]
+    pub fn set_no_data_value_i64(&mut self, no_data: Option<i64>) -> Result<()> {
+        let rv = if let Some(no_data) = no_data {


Well, isn't this awkward...

Taking an Option seemed like a good idea at the time. Now we have three ways to clear it.

I still think there's value in keeping them consistent...

The alternative is to add GDALDeleteRasterNoDataValue

And now I think we should go for that (but in a separate PR).

…erNoDataValueAsUInt64`

src/raster/tests.rs

lnicola · 2024-02-01T19:45:11Z

Please squash at the end!

lnicola · 2024-02-01T19:56:22Z

🤦

EDIT: nvm, my bad.

Added bindings for GetNoDataValueAsInt64 and GetNoDataValueAsUInt64

29c1615

metasim requested a review from lnicola February 1, 2024 15:46