forestgeo/fgeo.plot

Release 1.1.6

Closed this issue · 0 comments

Prepare for release:

  • Check that description is informative
  • Update packages and package-versions
  • Check licensing of included files
  • Check that all @authors mentioned in docs are listed in DESCRIPTION
  • usethis::use_cran_comments()
  • devtools::check()
  • rhub::check_for_cran()
    • windows
    • ubuntu
    • fedora
  • devtools::check_win_devel()
  • rhub::check_on_macos()
  • rhub::check(platform = 'ubuntu-rchk')
  • rhub::check_with_sanitizers()
  • rhub::check(platform = "fedora-clang-devel")
  • rhub::check(platform = "fedora-gcc-devel")
  • X rhub::check_on_solaris()
  • Polish pkgdown reference index

Submit to CRAN:

  • Remove badges
  • usethis::use_version('major')
  • Update cran-comments.md
  • devtools::release()
  • Approve email

Wait for CRAN...

Rejected:

* using log directory 'd:/RCompile/CRANincoming/R-devel/fgeo.plot.Rcheck'
* using R Under development (unstable) (2019-05-12 r76489)
* using platform: x86_64-w64-mingw32 (64-bit)
* using session charset: ISO8859-1
* checking for file 'fgeo.plot/DESCRIPTION' ... OK
* this is package 'fgeo.plot' version '1.1.4'
* package encoding: UTF-8
* checking CRAN incoming feasibility ... NOTE
Maintainer: 'Mauro Lepore <leporem@si.edu>'

New submission
* checking package namespace information ... OK
* checking package dependencies ... ERROR
Package required but not available: 'fgeo.tool'

See section 'The DESCRIPTION file' in the 'Writing R Extensions'
manual.
* DONE
Status: 1 ERROR, 1 NOTE

The problem was that, by the time I submitted fgeo.plot, fgeo.tool had not been installed in at least one CRAN machine. There was nothing I could do to fix the problem. Just wait.

Also, Uwe Ligges aked if I really needed so many fgeo.* packages or could reduce them to one or just a few. I explained my reasons and he agreed to keep fgeo packages as is. The relevant text follows.

RE: Do you really need so many packages? This creates quite some overhead. Can't you restrict this to one or at least few packages?
I understand your concern. I'll explain my thinking.

We designed a collection of R packages, fgeo, to deal with ForestGEO data (https://forestgeo.si.edu/). ForestGEO has been collecting tree-census data regularly since 1980. The network and data continuously grow. We now have 65 forest plots in 27 countries. The data has a very specific structure and, until now, there was no R package on CRAN to access, manipulate, analyze, and visualize ForestGEO data. fgeo fills this gap.

To manage complexity and leave room for growth, fgeo is modular:

  • fgeo.x (on CRAN) contains small ForestGEO datasets, for examples and tests.
  • fgeo.tool (on CRAN) contains general functions to import and manipulate ForestGEO data.
  • fgeo.plot (did not pass the incoming checks automatically because fgeo.tool wasn't available) contains tools for plots.
  • fgeo.analyze (did not pass the incoming checks automatically because fgeo.tool wasn't available) contains functions for common analyses of ForestGEO data.
  • fgeo is a meta-package (not yet submitted) to install and load all other packages with a single R command. Most of our users should interact with this package only.
    Any new code should fit in one of these packages. We expect the code base to grow but the package-ecosystem should be stable.

I value your experience. Would you suggest a different approach? What next steps do you recommend?

...

Uwe answered:

OK, then let us go ahead.

  • Recheck and resubmit

Wait for CRAN...

Thanks, please replace \dontrun{} by \donttest{} or unwap the examples
if they can be executed in less than 5 sec per Rd-file.

Please ensure that your functions do not write by default or in your
examples/vignettes/tests in the user's home filespace. That is not allow
by CRAN policies. Please only write/save files if the user has specified
a directory. In your examples/vignettes/tests you can write to
tempdir(). E.g.

pdf(file.path(tempdir(), "map.pdf"), paper = "letter", height = 10.5,
width = 8)

Wait for CRAN...

  • Accepted
  • Tag
  • GitHub release
  • usethis::use_dev_version()
  • Restore badges in README with git revert --no-edit <sha that removed badges>
  • Update install instructions in README
  • Commit check-with-r-3.4
  • Merge

From forestgeo/learn#212
Adapted from https://github.com/r-lib/vroom/issues/83