CRS returns NA when rgdal is not installed

[edzer]

This came up here: https://github.com/rvalavi/blockCV/issues/18

> as(st_crs(4326), "CRS")
CRS arguments: NA 

which calls CRS(SRS_string = st_crs(4326)$wkt) - this should stop or accept the wkt without checking. Also setting doCheckCRSArgs=FALSE does not help.

[rsbivand]

My commits/pushes to my forks of sf and sp: https://github.com/rsbivand/sf/commit/cd26c08c34781206ef932c32ec0d0797ab321e7f https://github.com/rsbivand/sp/commit/fbe5b883cde864861ac7202bad3122aec824b03e give (with no installed rgdal - the fixed just trust sf):

> library(sf)
Linking to GEOS 3.10.0dev, GDAL 3.3.0, PROJ 8.0.1
> as(st_crs(4326), "CRS")
Coordinate Reference System:
Deprecated Proj.4 representation: +proj=longlat +datum=WGS84 +no_defs 
WKT2 2019 representation:
GEOGCRS["WGS 84",
    ENSEMBLE["World Geodetic System 1984 ensemble",
        MEMBER["World Geodetic System 1984 (Transit)"],
        MEMBER["World Geodetic System 1984 (G730)"],
        MEMBER["World Geodetic System 1984 (G873)"],
        MEMBER["World Geodetic System 1984 (G1150)"],
        MEMBER["World Geodetic System 1984 (G1674)"],
        MEMBER["World Geodetic System 1984 (G1762)"],
        ELLIPSOID["WGS 84",6378137,298.257223563,
            LENGTHUNIT["metre",1]],
        ENSEMBLEACCURACY[2.0]],
    PRIMEM["Greenwich",0,
        ANGLEUNIT["degree",0.0174532925199433]],
    CS[ellipsoidal,2],
        AXIS["geodetic latitude (Lat)",north,
            ORDER[1],
            ANGLEUNIT["degree",0.0174532925199433]],
        AXIS["geodetic longitude (Lon)",east,
            ORDER[2],
            ANGLEUNIT["degree",0.0174532925199433]],
    USAGE[
        SCOPE["Horizontal component of 3D system."],
        AREA["World."],
        BBOX[-90,-180,90,180]],
    ID["EPSG",4326]] 

with an extra argument to sp::CRS(). I think that in sf/R/sp.R:

nm <- "CRS"
attr(nm, "package") <- "sp"
obj <- new(nm, projargs=from$proj4string)
comment(obj) <- from$wkt

will also work, but may make sf depend on 'methods - untried.

[rsbivand]

I reverted the additional sp::CRS() argument to avoid depending on a specific sp version; the use of new("CRS", ...) did not show any extra problems in check: https://github.com/rsbivand/sf/commit/44bbae31c82a2855d0c25b19e9194fe688d71c4b https://github.com/rsbivand/sp/commit/8f3be3e9eac5dbd9aae4716922a41bc094979edf So new() is my preference and also signals a start to using PROJ/GDAL through sf even when constructing an sp "CRS" object. The commit and reversion for sp are in PR #103.

[edzer]

Thanks, makes a lot of sense!