Framework issues

[sckott]

Framework issues

[sckott]

Thanks for your interest! This is a complex piece of software with many parts, so I've been hesitant to spend time on this unless there was interest in it. So if this worked you would use it? Do you know any others that would?

So the package doesn't work quite yet as a normal package would with exported functions, but you can run that workflow stuff in the readme https://github.com/ropensci/mutant#current-workflow if you clone the repo and then load all functions, e.g., pkgload::load_all() in the dir for mutant on your machine. Of course use a path to a different package you have on your machine rather than randgeo. That workflow at least demonstrates all the basic parts of mapping a package's functions, mutating functions, re-writing a package, and running the test suite of the mutated pkg.

[eduardokapp]

@sckott, thank you for quickly answering. Yes, I would very much like to use this package. My team is trying to improve our overall test quality. We do unit testing through testthat and use lintr for style. On top of that, we use covr for our CI. However, we recently decided to try for unit test mutations, but couldn't find any package that could do that other than this one.

I'm pretty confident that if this package gets a bit more sophisticated it would certainly receive praise/attention from the R community. Considering that most R users are not exactly developers or actually use R for a living, such improvement would help those that are and actually expand the language's "frontiers", so to speak :) .

[sckott]

Glad to hear there's interest. Sounds like I may be able to ping you when there's something to test

[jkamins7]

Might be off topic, but you can ping me when there's something to test too.

[sckott]

@jkamins7 would be happy to do so, but i'm changing jobs, so very very unlikely to work on this at all moving forward. If anyone wants to pick this project up let me know, or create your own elsewhere