Fixes for CRAN resubmission

[b-rodrigues]

#327

[b-rodrigues]

I think that the commit cdffb86 deals with the DESCRIPTION file (title, and write software names with single quotes). AFAIK, the point regarding references is not relevant here? Or should we cite Eelco's paper on Nix? We could I guess.
The final point I think might be a misunderstanding: they functions don't write by default, since overwrite = FALSE. The user needs to explicitly set this argument to TRUE, so I think we need to explain this in cran-comments.md.

[b-rodrigues]

In case we want to cite Eelco's paper:

Eelco et al. (2004) <doi:10.5555/1052676.1052686>

what do you think @philipp-baumann ?

[philipp-baumann]

Reproducible Software Environments with 'Nix'

[philipp-baumann]

@b-rodrigues there are a couple of lint problems that need a manual fix

let's use

# nolint start: line_length_linter
# nolint end

or

# nolint next: line_length_linter

or

# nolint

if stuff cannot be broken down to 100 characters (urls)...

[philipp-baumann]

In case we want to cite Eelco's paper:

Eelco et al. (2004) <doi:10.5555/1052676.1052686>

what do you think @philipp-baumann ?

yeah sure, sounds good to me!

[b-rodrigues]

That’s not going to fly:

Please ensure that your functions do not write by default or in your examples/vignettes/tests in the user's home filespace (including the package directory and getwd()). This is not allowed by CRAN policies.
Please omit any default path in writing functions. In your
examples/vignettes/tests you can write to tempdir().
-> R/rix.R; R/rix_init.R

it’s ok for nix_build() because it’s not writing, only looking for a default.nix

[philipp-baumann]

@b-rodrigues ready for your review ;-)