Jakub

@mhesselbarth @bitbacchus -- why do we split multipolygons to polygons before doing the sample calculations -- https://github.com/r-spatialecology/landscapemetrics/blob/acd3fd6dffa1908a8afb3cfaa5341cc50054e1ae/R/sample_lsm.R#L150 ? ``` r library(terra) library(landscapemetrics) library(sf) library(dplyr) library(tidyr) # prepare example data ----------------------------------------------------...

@MikeDTaylor -- I commented out the disaggregation line of code and rerun the calculations -- it seems to work as you want it to... However, to made any changes to...

@MikeDTaylor good to hear that you was able use your local version. Just in case, I also created a branch that you can use in the future -- `remotes::install_github("r-spatialecology/landscapemetrics@nodis")`.

Yep -- that would be my assumption as well. We probably should make a breaking change sometime in the future to allow the distinction between polygons and multipolygons.

@mhesselbarth should we make this change (removing `y

Then it would depend on the input geometry type. If the input has 7 POLYGONS, you will get 7 values as the result. But if the input is one MULTIPOLYGON...

@mhesselbarth See the results using `remotes::install_github("r-spatialecology/landscapemetrics@nodis")`: ``` r library(terra) library(landscapemetrics) library(sf) library(dplyr) library(tidyr) # prepare example data ---------------------------------------------------- r 6 1 class 3 NA pland 58.8 2 99.1 # MULTIPOLYGON...

Hi @bitbacchus and @mhesselbarth, I am following this discussion (and @bitbacchus progress) closely. My experience is similar to the @bitbacchus comments above - for a moving window problem, a lot...

FYI: One thing I found out today is that there is a quite overload caused by creating tibbles for each metric. The below line of code takes about 2.5 seconds...

Or we could make tibble optional... Or just return a vector of values from each main functions, and have another layer that creates a data.frame/tibble....