Edzer Pebesma comments

Results 708 comments of


                                            Edzer Pebesma

sf 1.0 will switch to using S2 geometry for ellipsoidal coordinates

Trying the path trough getting `s2_rebuild` inside `st_make_valid`, I stumble upon this: ```r library(sf) # Linking to GEOS 3.9.0, GDAL 3.2.1, PROJ 7.2.1 library(s2) data(World, package = "tmap") s2 =...

sf 1.0 will switch to using S2 geometry for ellipsoidal coordinates

As far as I can tell, yes.

sf 1.0 will switch to using S2 geometry for ellipsoidal coordinates

Now, `st_make_valid` makes unprojected geometries valid on S2, using `s2::s2_rebuild`, using a precision of 1e7 if no precision has been set, and when the argument `s2_options` has not been specified.

sf 1.0 will switch to using S2 geometry for ellipsoidal coordinates

I will start another round of reverse depency checks, and look at the diffs. For `tmap` and `tmaptools` for instance, @mtennekes, this implies that `st_make_valid` must be called on `World`...

sf 1.0 will switch to using S2 geometry for ellipsoidal coordinates

Thanks, @ranghetti - very helpful! In `sf` I now fixed this with a temporary (=slow) fix until we get this properly sorted in `s2`.

sf 1.0 will switch to using S2 geometry for ellipsoidal coordinates

it works for both, but differently: see e.g. https://keen-swartz-3146c4.netlify.app/spherical.html#validity-on-the-sphere After unprojecting `World`, it is interpreted on S2 and no longer valid, and needs to be pulled through `st_make_valid()`.

Postgis: add support for more geometry types

Is this still on your to-do list?

recording of talks, slides or other resources

Thanks for the clear list @kadyb ! Happy to be involved in your ocean hack week, remotely - I'm always on the watch for challenges and new use cases.

`st_transform(, ...pipeline=)` issue?

> curiously terra and sp agree on the x/y names of the bbox, but sf reverses them? ```r f

`st_transform(, ...pipeline=)` issue?

I guess the `st_transform` needs to have a bit of text in the Details section added on the axis order mess when `pipeline` is specified (as of: do it yourself).