New versions of rgdal, rgeos and maptools have just been published on CRAN. Their startup messages on attachment now repeat Edzer Pebesma's announcement in his useR! 2021 keynote that their planned retirement will occur at the latest at the end of 2023. See https://www.youtube.com/watch?v=cK08bxUJn5A&list=PL4IzsxWztPdmHoJwIVa4um44w2GMjctmP&index=6, about 23 minutes in.

I maintain these packages, in addition to a number of others, but now much of their functionality is available in sf, terra and vapour (among others), which link directly to one or more of the external libraries GDAL, PROJ and GEOS. As can be seen from my signature, I am myself retired, but will continue to contribute to implementing transitions.

The sp package will receive the coercion methods for spatstat now in maptools, and possibly other functions. sp will be modified to use sf rather than rgdal for coordinate reference systems, and coercion between sf and sp vector objects is reliable. Work will continue on instantiating sp raster objects from stars and terra (and other) rasters.

The maintainers of packages using rgdal, rgeos and maptools should heed the gentle suggestion in their startup messages. Deprecation will follow during 2022. Securing reproducibility will be harder as the APIs of PROJ, GDAL or GEOS shift, making it impossible to install rgdal or rgeos against future versions of these libraries. However, they are all open source, so reproduction will involve installing legacy external libraries from source, followed by R packages from source.

I hope that this doesn't spoil your day, but planning transition is at least less painful than an unmanaged break of service.

Roger

--
Roger Bivand
Emeritus Professor
Department of Economics, Norwegian School of Economics,
Postboks 3490 Ytre Sandviken, 5045 Bergen, Norway.
e-mail: roger.biv...@nhh.no
https://orcid.org/0000-0003-2392-6140
https://scholar.google.no/citations?user=AWeghB0AAAAJ&hl=en

_______________________________________________
R-sig-Geo mailing list
R-sig-Geo@r-project.org
https://stat.ethz.ch/mailman/listinfo/r-sig-geo

Reply via email to