start geopandas extension #728
Conversation
|
resolves #136 |
|
Keep me in the loop. Happy to do a proper review of this from the geopandas perspective once @kylebarron is happy. |
|
i think this should be ready, though I'm not the best with type hints or google docstrings |
|
Can you check why pre-commit is still failing? You should be able to run it locally with |
|
@martinfleis do you have any comments on this PR? I'd like to get in #753, which in general raises the code quality of the code base, and then ensure that this PR meets that bar as well. |
lonboard/geopandas.py
Outdated
| elevation : Name of column on the dataframe used to extrude each geometry or | ||
| an array-like in the same order as observations. Defaults to None. | ||
| extruded : Whether to extrude geometries using the z-dimension. |
There was a problem hiding this comment.
I probably don't follow here. Shouldn't extruded be automatically True when elevation is provided? Like the two are tied together, no? Or can I use extruded without elevation? What does it extrude on, then, Z coordinate? How?
There was a problem hiding this comment.
i agree this is a little awkward. There's probably no need for the bool, though I was just trying to follow the viz/Layer API which treats them separately. Happy to go with consensus here.
Another small wrinkle is that sometimes in GIS or CAD software (IIRC anyway) 'elevation' sometimes refers to the height beneath the polygon, e.g. so you can drop the extruded poly onto a DEM for things like viewsheds. In that case I want to say this parameter is called 'extrusion_height' or something? No real opinion here, just something to consider in case folks are coming from that world
There was a problem hiding this comment.
i removed the bool so that its turned on automatically if elevation is passed
|
fixed annotations in the notebook so im not sure what the lingering failure is |
|
copiolt thinks the failure is a caching issue (apparently), so i'd guess it will pass if re-run on its own |
|
looks like maybe some other parts of the stack are moving at the moment, but let me know if there's anything I can do to help move this forward :) |
| def explore( # noqa: PLR0913 | ||
| self, | ||
| column: str | None = None, | ||
| cmap: str | None = None, | ||
| scheme: str | None = None, |
There was a problem hiding this comment.
Can we make some of these parameters keyword-only? Perhaps change to
| def explore( # noqa: PLR0913 | |
| self, | |
| column: str | None = None, | |
| cmap: str | None = None, | |
| scheme: str | None = None, | |
| def explore( # noqa: PLR0913 | |
| self, | |
| column: str | None = None, | |
| *, | |
| cmap: str | None = None, | |
| scheme: str | None = None, |
so that everything after column can only be passed as a named kwarg?
| def _dexplore( # noqa: C901, PLR0912, PLR0913, PLR0915 | ||
| gdf, # noqa: ANN001 | ||
| *, | ||
| column, # noqa: ANN001 | ||
| cmap, # noqa: ANN001 | ||
| scheme, # noqa: ANN001 | ||
| k, # noqa: ANN001 | ||
| categorical, # noqa: ANN001 | ||
| elevation, # noqa: ANN001 | ||
| elevation_scale, # noqa: ANN001 | ||
| alpha, # noqa: ANN001 | ||
| layer_kwargs, # noqa: ANN001 | ||
| map_kwargs, # noqa: ANN001 | ||
| classification_kwds, # noqa: ANN001 | ||
| nan_color, # noqa: ANN001 | ||
| color, # noqa: ANN001 | ||
| vmin, # noqa: ANN001 | ||
| vmax, # noqa: ANN001 | ||
| wireframe, # noqa: ANN001 | ||
| tiles, # noqa: ANN001 | ||
| highlight, # noqa: ANN001 | ||
| m, # noqa: ANN001 | ||
| ) -> Map: |
There was a problem hiding this comment.
It's not ok to ignore all of these errors. At a minimum we need every function parameter and return type to be correctly typed so that we can catch internal type errors.
There was a problem hiding this comment.
Instead of defining this as a separate function; put this implementation inside of LonboardAccessor.explore. There's not really any reason to have this defined as a standalone function?
| return new_m | ||
|
|
||
|
|
||
| def _get_categorical_cmap(categories, cmap, nan_color, alpha): # noqa: ANN001, ANN202 |
| from matplotlib import colormaps | ||
| except ImportError as e: | ||
| raise ImportError( | ||
| "this function requires the `lonboard` package to be installed", |
There was a problem hiding this comment.
| "this function requires the `lonboard` package to be installed", | |
| "this function requires the `matplotlib` package to be installed", |
| elevation, # noqa: ANN001 | ||
| elevation_scale, # noqa: ANN001 | ||
| alpha, # noqa: ANN001 | ||
| layer_kwargs, # noqa: ANN001 |
There was a problem hiding this comment.
It looks like layer_kwargs is something you're adding that doesn't exist in the upstream GeoDataFrame.explore. In that case, you should follow the same behavior as in viz and use the same parameters:
Lines 92 to 95 in c6c3209
This will also fix your type hinting below:
There was a problem hiding this comment.
Sorry that this has sat for a while without a review. I have too many projects and haven't dedicated time to Lonboard recently.
Because I have too many projects, I also have a high bar for contributions so that they don't grow to become a larger maintenance burden in the future.
There's still a decent amount of work here before it's review ready. In particular, there should be virtually no type: ignore statements.
|
no sweat! thanks for the comments right there with you and similarly spread thin :). Static typing is pretty extraneous in most of my projects so not very practiced; will take another pass, or several |
to begin review. The notebook at examples/explore should run through and demonstrate functionality. I need to take a look at how you prefer to test against Maps