not all patches linked by MPG()

[achubaty]

Using the NA test case from #28, note that not all patches end up linked:

f <- "naErrorExample.asc"
r <- raster(f)
c <- r
p <- (r == 1)

mpg <- MPG(cost = c, patch = p)
plot(mpg)

[achubaty]

@samdoctolero @ecologics poking at this more, I think we mis-diagnosed this bug:

First, update to the latest version of the package to ensure the new test cases and test data are available to you.

library(igraph)
library(raster)
devtools::load_all("~/GitHub/grainscape")

naError <- raster(system.file("extdata/naErrorExample.asc", package = "grainscape"))
naErrorMPG <- MPG(cost = naError, patch = (naError == 1))

## plotting the result, it appears the top right patches aren't connected
plot(naErrorMPG)

## however, the underlying igraph object looks correct
g <- naErrorMPG$mpg
#g <- simplify(g, remove.multiple = TRUE, remove.loops = TRUE)
count_components(g) ## equals 1, despite the plot above showing 2 subgraphs
plot(g) ## also shows all nodes connected

To me it suggests that either the igraph object (naErrorMPG$mpg) differs from the raster that is produced (and then plotted) OR that the plotting is screwing up (though this seems unlikely because it's simply plotting the raster using base R, i.e., nothing we coded).

Thoughts?

[achubaty]

fixed in 5a5292a

[achubaty]

thanks @samdoctolero for tracking this down and fixing

[achubaty]

@cxfauvelle @ecologics please update your machines to the latest grainscape, which fixes the patch-linking bug

[ecologics]

No kidding @samdoctolero. Big thanks. Paul