gee-community / pygaul Goto Github PK

It will virtually return the complete dataset but who cares ?

replace get_items by an Items object. It will offer more flexibility for extention. same for get_names.

The current display is based on the geemap lib. Unfortunately the latter is displaying FeatureCollection as image which results in grey background in the dark theme.

We should rely on pure ipyleaflet methods to represent these datasets

GEE cannot connect to jupyter-sphinx, I need to adapt the trick I use in sepal-ui

If one use exotic ways to connect to GEE (as in a private company where we connect with a service account) running ee.Initialize at the top of every file may end up by killing the process.

Instead the user should be responsible for initializeing ee and/or we should be able to connect to it whitouh conflict (more dificult)

get_names could also be used to know where an AOI coming from (which country, whoch state). In the current implementation, this information is hidden by default

I need to use the same trick as in sepal-ui

• conda-forge/staged-recipes#23457

centerObject is not managing the zoom

check all continent shapes and remove the one that should not be here

• africa
• asia
• europe
• oceania
• north america
• south america
• antartica

we now get gull featureColection for each tests, it's useless and take time

one of the tools in @sepal-contrib was using a previous implementation of sepal_ui where the gaul_database was exposed, now that it has been removed in favor to the pygaul package, there's no way to directly access to this file.

Would it make sense to expose the database as a dataframe?, for example:

from pygaul.data import gaul_dadtabase

My current workaround imports the file using pkgs_resources:

resource_path = "pygaul/data/gaul_database.parquet"
content = pkg_resources.resource_filename("pygaul", resource_path)

verify that everything still works since the move to gee-community:

• github actions
• devcontainer
• RDT
• coverage
• pypi
• conda

yes I know it's the third time I change but I think it's the lats because it cannot be simplified more and it looks robust enough to be understood by anyone.