New version > CRAN take two

[jooolia]

We were working towards a new version on CRAN in #71 but were waylaid by many of the NCBI functions that were no longer properly working. We will close #71 and prepare for a new release in this issue. So completeness I have kept the same list as in #71 even though I think several of the items are already complete.

From r-lib/usethis#338

Prepare for release:

• devtools::check_win_devel()
  1 Note: Maintainer: 'Julia Gustavsen j.gustavsen@gmail.com' seems to be ok based on
  https://mailman.stat.ethz.ch/pipermail/r-devel/2014-March/068497.html
• rhub::check_for_cran():
  Linux distros have the same note as above for the maintainer:
  rsnps 0.3.2.9121: NOTE Platform: | Ubuntu Linux 16.04 LTS, R-release, GCC
  https://builder.r-hub.io/status/rsnps_0.3.2.9121.tar.gz-e8127d069c0b4abc9657e2a1951bc3a6
  rsnps 0.3.2.9121: NOTE Platform: | Fedora Linux, R-devel, clang, gfortran
  https://builder.r-hub.io/status/rsnps_0.3.2.9121.tar.gz-4a669680d60c47ebb1c44d5885de7649
  Windows has this error:
  rsnps 0.3.2.9121: ERROR Platform: | Windows Server 2008 R2 SP1, R-devel, 32/64 bit
  checking package dependencies ... ERROR
  Package required but not available: 'data.table'
  https://builder.r-hub.io/status/rsnps_0.3.2.9121.tar.gz-a439a712d233486da4b75e700a6bb44d
  Lines 255-257:

#> Package which is only available in source form, and may need
#> These will not be installed
#> compilation of C/C++/Fortran: 'data.table'

• Trying again with
  rhub::check_for_cran(env_vars=c(R_COMPILE_AND_INSTALL_PACKAGES = "always") as {data.table} was not compiled and thus not installed.
  This command gave:
  rsnps 0.3.2.9121: NOTE (as above about maintainer) Platform: | Windows Server 2008 R2 SP1, R-devel, 32/64 bit
  https://builder.r-hub.io/status/rsnps_0.3.2.9121.tar.gz-a3cf0e4f3a314cdda42a83b1c48edb8b

• revdepcheck::revdep_check(num_workers = 4, timeout = as.difftime(360,units = "mins"))

  • revdepcheck no longer on CRAN? (opened issue package no longer on CRAN? r-lib/revdepcheck#263), installed with:
    devtools::install_github("r-lib/revdepcheck")
  • opened PR updated revdep folder readme.md after running revdepcheck #107 to merge in updates to revdep folder readme to document this check.

• If new failures, update email.yml then revdepcheck::revdep_email_maintainers() -no new failures

• Polish NEWS

  • opened PR updating news #111 for this.

• update comments-cran.md

  • opened PR update cran comments #112 for this.

• update codemeta.json using codemetar::write_codemeta(".") (from within project folder on dev branch)

  • opened PR updating codemeta for issue #102 #108 for this.

• update LICENSE to reflect new maintainers

  • opened PR update to correct year #109 for this.

• update COC to include only absolute not relative links

  • COC already only contains absolute links, but fixed a url in PR update COC #110

Perform release:

• Bump version (in DESCRIPTION and NEWS) PR updated version in DESCRIPTION to match the version in NEWS.MD #113
• Create RC branch (for bigger releases) opened PR Merge release candidate dev into master before submitting to CRAN #114 to merge dev into master
  • asked for review from Scott (received and very helpful). Opened PRs [IN PROGRESS] resolution of comments from Scott in PR #114. #116 and SR [IN PROGRESS] resolution of comments from Scott in PR #114 (Sina) #118 to resolve the problems found.
• devtools::check_win_devel() (again!)
• devtools::submit_cran() devtools::release()
• pkgdown::build_site() (for https://docs.ropensci.org/rsnps/ it is automatically done )
• Approve email

Wait for CRAN...

• Tag release
• Merge RC back to master branch ( merging dev into master before submitting to CRAN)
• Bump dev version
• Write blog post
• Tweet
• Add link to blog post in pkgdown news menu

Template from r-lib/usethis#338

[sinarueeger]

@jooolia shall we go ahead with

1. merging this into the master first (let Scott review) and
2. submitting to CRAN

?

[jooolia]

Hi @sinarueeger,
Yes sounds good. I guess we should also fix the version in the description to match what is in NEWS.md? I will do that right now.

[sinarueeger]

Hi Julia! Forgot about the version number... thanks for fixing that (just approved the PR).

[jooolia]

Just ran devtools::check_win_devel() again and now there is a new WARNING:
https://win-builder.r-project.org/iUiM5i0dzCRa/00check.log
checking re-building of vignette outputs ... [0s] WARNING
Error(s) in re-building vignettes:
--- re-building 'rsnps_vignette.Rmd' using knitr
Error: processing vignette 'rsnps_vignette.Rmd' failed with diagnostics:
The 'markdown' package should be declared as a dependency of the 'rsnps' package (e.g., in the 'Suggests' field of DESCRIPTION), because it contains vignette(s) built with the 'markdown' package. Please see yihui/knitr#1864 for more information.
--- failed re-building 'rsnps_vignette.Rmd'

SUMMARY: processing the following file failed:
'rsnps_vignette.Rmd'

So I will try following this advice and re-running to see if this solves this issue. If so I will open a PR.

UPDATE: Adding in {markdown} fixes the issue and now there is only 1 NOTES (about the maintainer, there is a second there because I had an uncommitted file in the package directory). https://win-builder.r-project.org/J4mZV0q1B1q2

[jooolia]

Hi @stefaniebutland,
@sinarueeger and I have released a new version of rsnps on CRAN. We were wondering if it might be interesting to write up a short post on this update as the package was not wholly working properly (due to a change in one of the APIs) and we have now fixed it.

Thanks, Julia

[stefaniebutland]

Sorry for my delay - on staycation will Sep 9 but couldn't resist coming here to say congratulations and a post sounds great.

Our Blog Guide has most of the info you should need. In this case, it sounds like a tech note might be most appropriate, unless you wanted to include more narrative. We haven't had a post about rnsps before and you did take on co-maintenance so there are opportunities to tell a story.

Let me know when you'd like to submit a draft and I can suggest a publication date.

[jooolia]

Thanks a lot @stefaniebutland! We will mull it over and get back to you. :)

[stefaniebutland]

@jooolia @sinarueeger thoughts on tech note vs blog post and when you might want to submit a draft? (no pressure)

[jooolia]

Hi @stefaniebutland thanks for following up. We will meet and discuss this a bit next week. Then we can let you know what we're thinking. :)

[stefaniebutland]

@jooolia have you made a decision about post or tech note? You can suggest a date for draft submission and based on that I'll propose a publication date.

[jooolia]

Thanks @stefaniebutland for following up. We will write a blog post when we will propose some new features.