Error when using spatial_file_path

[dk-forestry]

I'm trying to download MODIS data using the spatial_file_path-option. When using the shapefile manually in the GUI to "Load extent from a spatial file", everything works as expected. However, when I provide it as an argument like this

MODIStsp(gui = FALSE, options_file = "options.json", spatial_file_path = "shapefile.shp")

I get the following error:

Error in as.vector(x) : no method for coercing this S4 class to a vector

Would you have any ideas what might be the problem here?

[ranghetti]

Dear @dk-forestry,
thank you for signaling it; i made a change in the code which should have solved the problem. Please try to update your MODIStsp installation from GitHub (devtools::install_github("lbusett/MODIStsp")).

[dk-forestry]

Thank you, it now works with the updated version. :)

By the way, I noticed two things related to this:

1. In the example for Non-Interactive Execution and scheduled processing the following code is used:

# Create a character array containing a list of shapefiles (or other spatial files)
extent_list = list.files("X:/path/containing/some/shapefiles/", "\\.shp$")

Shouldn't list.files include the argument full.names = TRUE?

2. If the argument for spatial_file_path points to a non-existing file, the user receives the following error:

Error in MODIStsp(gui = FALSE, options_file = option_file, spatial_file_path = single_shape) :
Error in bbox_from_file(file_path = spatial_file_path, out_crs = general_opts$user_proj4) :
File format not recognized by GDAL or OGR.
Details: Error in CRS(gsub("'", "", cmd_output)) :
PROJ4 argument-value pairs must begin with +: ERROR 1: ERROR - failed to load SRS definition from NA

Maybe it would make sense to check somewhere if the shapefile exists and give a clear error if not?

[lbusett]

Both good points ! Thanks for the suggestions !
We'll implement them in the next release.

Lorenzo

[dk-forestry]

I was working and downloading some data with the new version from above and noticed something else related to the spatial_file_path.

Assuming I have a shapefile called studyarea.shp, then a folder called studyarea is created in the Main output folder. So far, so good. However, if MODIStsp is run again (with exactly the same option file and shapefile), a new folder called studyarea_1 is created, and all data is reprocessed (even though the ReProcess Existing Data was set to No. If the command is run again, studyarea_2 is created, then studyarea_3, and so on.

I imagine that this behavior is unexpected, because during scheduled data downloads or this workaround the same data would be created over and over (until the HDD is at its limits ;) ). Is there maybe a fix or workaround?

[lbusett]

Definitely a bug in the spatial_file_path option. We'll investigate it ASAP.

[ranghetti]

This was an expected behaviour: the intention was to create separate folders in case this function was run in a cycle, in which spatial files had the same basename (being placed in different folders).
Actually this is an uncommon situation, while scheduled download appears to be a more common one. So this functionality was remove: now (with commit 099b432) the original folder name is maintained.

[dk-forestry]

Thanks for the explanation and the changes to the code! :)