Closed edzer closed 4 years ago
With new sf, we see
st_crs("+proj=longlat +datum=WGS84 +no_defs")$epsg
resulting in NA rather than 4326. This is GDAL, trying to guess the epsg from the WKT representation of the proj4string. I think this is wrong, but cannot be resolved by sf. In any case, it causes errors in your tests, and errors on CRAN once sf gets accepted. See https://win-builder.r-project.org/incoming_pretest/sf_0.9-0_20200319_213604/reverseDependencies/summary.txt and https://www.r-spatial.org/r/2020/03/17/wkt.html
Many thanks @edzer for the report and analysis.
With new sf, we see
resulting in NA rather than 4326. This is GDAL, trying to guess the epsg from the WKT representation of the proj4string. I think this is wrong, but cannot be resolved by sf. In any case, it causes errors in your tests, and errors on CRAN once sf gets accepted. See https://win-builder.r-project.org/incoming_pretest/sf_0.9-0_20200319_213604/reverseDependencies/summary.txt and https://www.r-spatial.org/r/2020/03/17/wkt.html