Closed Mervap closed 4 years ago
Please close this bug report and open one on pkgKitten
.
You are free to uninstall pkgKitten
should you have it installed. It is just a suggested package -- if you don't want it, don't use it.
And yes, I want it. I wrote it for Rcpp and also use it for RcppArmadillo and RcppEigen, and it has been used in the present form for many years. It's an optional feature.
Okey
But it seems very strange to me that the behavior described above does not lead to an error in Rcpp
and does it on RcppArmadillo
and RcppEigen
Shouldn't the behavior be similar?
Let me try to explain what is the problem Consider the following sequence of operations from the console:
After that, it is obvious that
tmp
contains only themyPackage
folder and themyPackage
folder is emptyNow I want to create a new R package inside the
myPackage
folderNow consider 2 cases:
The
pkgKitten
package is not installed on the current interpreter I run following codeand get the next one output
Let's look inside the
myPackage
folderGreat! The package was created
The
pkgKitten
package is installed on the current interpreter now Let's tryOops, error. And the directory is of course empty.
I think this behavior is unacceptable. It depends on some external factors. Are you sure you want to use the
pkgKitten
package? Or maybe you should give the user a choice whether to use it? The directory where I want to create the package may be not empty. I would like to be able to create a package without first deletingpkgKitten
every time. Thanks