bart1

A quick addition, if this is going to be implemented it might be worth to also think about integrating the use of `units`, both `stars` and `sf` work nicely with...

This might relate to #255 as well

I have been caught by a similar issue the other way around where warnings appear that should not be there: ``` r require(dplyr) #> Loading required package: dplyr #> #>...

I just encountered an other case where this pull request could provide a big speedup. When using `print.sfc` it can be quite slow (20 seconds for 10^7 points) as the...

I just played around with this pull request out of curiosity what the consequences would be for [move2](https://cran.r-project.org/web/packages/move2/index.html). I see substantial memory usage reductions, in the example data from 7mb...

Thanks @edzer , I tested a bit more and noticed two other examples changing for unexpected behavior. Sorry for having these issues iteratively, the popup up in some of the...

That works for the examples above but filter fails if there is an empty location: ``` r require(sf) #> Loading required package: sf #> Linking to GEOS 3.10.2, GDAL 3.4.1,...

It seems dplyr filter drops the CRS: ``` r require(dplyr) require(sf) #> Loading required package: sf #> Linking to GEOS 3.10.2, GDAL 3.4.1, PROJ 8.2.1; sf_use_s2() is TRUE data(meuse, package...

Hi, I have been doing some more testing. In general I use the test suite from `move2` (most test/functions use `sf` in the background in some way) to find these...

Another small change/corner case casting and empty `sfc` fails: ``` r require(sf) #> Loading required package: sf #> Linking to GEOS 3.10.2, GDAL 3.4.1, PROJ 8.2.1; sf_use_s2() is TRUE st_cast(st_sfc(),"POINT")...