Deal with ::: nonsense

[dill]

At the moment there is an issue when submitting dsm to CRAN that the dsm.var.prop method uses ::: to import functions from mrds that are not in its NAMESPACE. One way to get around this would be to export these functions in mrds and just give them the keyword internal in their documentation.

Is this a reasonable solution?

[LHMarshall]

Which functions from dsm are required in this way from mrds?

[dill]

From R CMD CHECK dsm_2.2.11.tar.gz:

* checking dependencies in R code ... NOTE
Unexported objects imported by ':::' calls:
  ‘mrds:::assign.par’ ‘mrds:::create.ddfobj’ ‘mrds:::detfct’ ‘mrds:::process.data’
  See the note in ?`:::` about the use of this operator.

[dill]

To add a little more to this...

Used by dsm.var.prop: mrds:::assign.par Used by generate.ds.uncertainty: mrds:::detfct mrds:::create.ddfobj mrds:::process.data

So there are two separate issues here -- one with variance propagation and one with the detection function uncertainty when doing bootstraps.

[dill]

assign.par is now exported from mrds (as of 7c1c5ce), code fixed in dsm as of 21b1720.

[dill]

b54c213 removes the dependency on mrds:::process.data from generate.ds.uncertainty.

[dill]

Fixed in 2103a63, taking into account mrds changes in e4e8d50.