Changing the name of an existing R package is annoying but common
task in the early stages of package development (at least for me).
changer tries to automate this task:
Validity and availability of the new package name is checked by
available from available).
All complete words matching the package name are
replaced in all R scripts, C/C++/Fortran/Stan source codes, markdown
files, and typical files in R packages, i.e. files with extension
.bib, as well as files
Change file names containing the package name. Only checks files
with extensions listed above, as well as files
TRUE (default), old
Rd files are removed.
All files with extensions
/src are removed to force a rebuild of
Name of the package directory is changed.
TRUE (default), the
remote url of the git repository is changed to match the new name. Note
that you still need to change the name of the repository in
Github/Bitbucket etc manually. For example in Github: Go to the URL
of your Github package, click Settings near the top-right,
change the name under Repository name, and click
documentation and rebuild package.
Inspired by back-and-forth naming of the package particlefield) and Nick Tierney’s blog post.
If the package is already published in CRAN, then you should first consult folks at CRAN whether they will accept the name change.
It is strongly recommended to have a backup backup before proceeding.
Note that if the old package name is a common word, it will be replaced in the docs etc as well!
You can install the developed version of changer from Github with:
There is just one function in the package:
replace_allyou can now replace also partial matches instead of just the whole word matches.
changernow tests that the supplied path actually contains an R package by checking the existence of the