geocompy Test and adapt to geopandas 1.0

https://martinfleischmann.net/geopandas-1.0-is-coming.-what-will-change/

May 05 '24 11:05 Nowosad

From a glance doesn't seem anything affects our code directly. Thoughts @anitagraser and @michaeldorman ? I guess we need to just test as @Nowosad says. A dev-packages.yml workflow could help with that.

May 07 '24 06:05 Robinlovelace

There are some code changes required. For example, .unary_union is flagged to be deprecated, replaced by .union_all() (which doesn't exist in geopandas==0.14.*). I suggest we switch to the newer functions whenever possible, what do you think? (If so, I will be happy if you can please update the package version on GitHub so that the book is rendered and tested with geopandas==1.0.0 on GitHub as well)

May 08 '24 07:05 michaeldorman

There are some code changes required. For example, .unary_union is flagged to be deprecated, replaced by .union_all() (which doesn't exist in geopandas==0.14.*). I suggest we switch to the newer functions whenever possible, what do you think? (If so, I will be happy if you can please update the package version on GitHub so that the book is rendered and tested with geopandas==1.0.0 on GitHub as well)

Agreed, let's go with the later version. I'm time limited and most geocompx attention on geocompr this week, if anyone else has a chance to look at the .yml or Docker build settings that would be most appreciated.

If not I should be able to look at this next week.

May 08 '24 08:05 Robinlovelace

Also advices from Martin seems to be useful: https://fosstodon.org/@martinfleis/112433669245696411

May 19 '24 12:05 Nowosad

Also advices from Martin seems to be useful: https://fosstodon.org/@martinfleis/112433669245696411

Now read the document, as far as I can tell we're already implementing all of the relevant advice

Jul 01 '24 13:07 michaeldorman