Goodbye Stamen

[emanuel-schmid]

Changes proposed in this PR:

• replace the default tiles provider to use OpenStreetMap instead of StamenTerrain

This PR fixes #796

PR Author Checklist

• Read the Contribution Guide
• Correct target branch selected (if unsure, select develop)
• Descriptive pull request title added
• Source branch up-to-date with target branch
• Documentation updated
• Tests updated
• Tests passing
• No new linter issues
• Changelog updated

PR Reviewer Checklist

• Read the Contribution Guide
• CLIMADA Reviewer Checklist passed
• Tests passing
• No new linter issues

[peanutfun]

Looks good, thanks for fixing this! ☺️

[rogerilesdr]

Hi Lukas, I'm running the latest core CLIMADA release v4.01 on Anaconda which still has the Stamen issue. I was just interested to know when an update with the resolution will be released so I can decide whether to look for a temporary workaround?
Thanks, Roger

[peanutfun]

@rogerilesdr The workaround is to specify a different provider in the Impact method parameters, e.g.,

import contextily as ctx
impact.plot_basemap_eai_exposure(url=ctx.providers.CartoDB.Positron)

[peanutfun]

@gaelleclain @rogerilesdr On second thought, I am not sure if that fix helps because the code in impact.py still includes the provider which does not exist anymore. You might need to use the development version of Climada, following the "Advanced Instructions" for installation until our next release.

[rogerilesdr]

That's OK, I've done a dirty workaround for now, I've manually updated the files changed, rather than using the development version for now. I'll overwrite these files when the next stable release arrives.

[gaelleclain]

@rogerilesdr can you please give a few details on your workaround?

• Did you update any other file than impact.py?
• Did you do anything else before testing?

Thanks

[rogerilesdr]

I updated the .py files listed on the Files tab above, which also shows where they're located... namely...

• test_plot.py
• impact.py
• base.py
• api_client.py

[peanutfun]

@rogerilesdr @gaelleclain Another fix I missed before is to revert xyzservices to version 2023.7.0 after installing Climada. This will still require to use the fix above for the plots themselves, but at least you should be able to load and execute Climada modules without any code modification.

[peanutfun]

@rogerilesdr @gaelleclain The latest update to our Conda package reverts xyzservices in the dependency list and should resolve the issue. Update the package with mamba update -n climada_env climada