wrapper for gdalmdimtranslate

[ChristianBeilschmidt]

• 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.

---

I created a wrapper around GDALMultiDimTranslate. I used part of the logic for the wrapper for the VRT program.

[lnicola]

Does this work?

Suggested change

	fn try_from(value: Vec<&str>) -> std::result::Result<Self, Self::Error> {
	fn try_from(value: Vec<&str>) -> Result<Self> {

[ChristianBeilschmidt]

23957df

[lnicola]

Should probably be called MultiDimTranslateOptions, since there can be multiple options inside.

[ChristianBeilschmidt]

23957df

[ChristianBeilschmidt]

I forgot to replace the Cell<bool> with a plain boolean, I guess it can be this simple 🙈 .

[lnicola]

bors r+

[bors]

Build failed:

• ci ubuntu-lts

[lnicola]

thread 'config::tests::test_error_handler' panicked at 'assertion failed: `(left == right)`
  left: `[(Warning, 1, "GPKG: bad application_id=0x00000000 on '/tmp/.tmpPTiAzK.gpkg'"), (Failure, 42, "foo"), (Warning, 1, "bar")]`,
 right: `[(Failure, 42, "foo"), (Warning, 1, "bar")]`', src/config.rs:230:9

Is this some sort of new GDAL warning?

[ChristianBeilschmidt]

thread 'config::tests::test_error_handler' panicked at 'assertion failed: `(left == right)`
  left: `[(Warning, 1, "GPKG: bad application_id=0x00000000 on '/tmp/.tmpPTiAzK.gpkg'"), (Failure, 42, "foo"), (Warning, 1, "bar")]`,
 right: `[(Failure, 42, "foo"), (Warning, 1, "bar")]`', src/config.rs:230:9

Is this some sort of new GDAL warning?

Since it appeared in the LTS-CI, I guess not.

What could have happened:
The tests run in parallel. Some test sets the global error handler. Then some other test fails… which lands in the error handler. And this test gets some strange results.

I'm not sure, but we could get races on all the global stuff.
Maybe we need some global Mutexes on some test stuff that must not interfere with something else 🤷.

What do you think?

[jdroenner]

bors merge

[bors]

Build succeeded:

• ci ubuntu-lts
• ci ubuntu-lts (--all-features)
• clippy

[lnicola]

I'm a bit worried about the test failure here.

[jdroenner]

i guess the problem is not really related to this PR but to gdal and cargo test in general. Cargo runs all the tests in parallel and when two tests fail at the same time we have a problem with global state.
So we need to find a solution for that. We could use a mutex for reading / writing global state...

[lnicola]

I think we can move them to an integration test.