Adding test coverage for RasterDataset.plot

[calebrob6]

Changing unrelated test data in #415 and #429 dropped test coverage for part of RasterDataset.plot. This PR adds a specific test for these lines.

[adamjstewart]

Once all GeoDataset datasets have their own plot method, RasterDataset.plot and VectorDataset.plot won't have any coverage at all. I think we'll need a PR like this in the meantime to prevent coverage from dropping.

In the long run, we have a few options:

1. Keep RasterDataset.plot and VectorDataset.plot but update the function signature to match
2. Make RasterDataset.plot and VectorDataset.plot abstract methods
3. Drop RasterDataset.plot and VectorDataset.plot

1 will result in dead code that isn't used by any dataset, so I don't think it's a great idea. 2 will allow us to enforce a consistent API for all plot methods, although it also means you can no longer instantiate a RasterDataset without first subclassing it and adding a plot method. We might have to go with 3.

[calebrob6]

We could do a mix of 1 and 2 and just raise a Not implemented error in RasterDataset.plot.

…

On Wed, Feb 23, 2022, 9:04 PM Adam J. Stewart ***@***.***> wrote: Once all GeoDataset datasets have their own plot method, RasterDataset.plot and VectorDataset.plot won't have any coverage at all. I think we'll need a PR like this in the meantime to prevent coverage from dropping. In the long run, we have a few options: 1. Keep RasterDataset.plot and VectorDataset.plot but update the function signature to match 2. Make RasterDataset.plot and VectorDataset.plot abstract methods 3. Drop RasterDataset.plot and VectorDataset.plot 1 will result in dead code that isn't used by any dataset, so I don't think it's a great idea. 2 will allow us to enforce a consistent API for all plot methods, although it also means you can no longer instantiate a RasterDataset without first subclassing it and adding a plot method. We might have to go with 3. — Reply to this email directly, view it on GitHub <#430 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AAIJUTX5OBAFVK36DSLBYZ3U4W37XANCNFSM5PFFWHTQ> . You are receiving this because you authored the thread.Message ID: ***@***.***>