Release ggplot2 3.1.0

[clauswilke]

Issues to resolve before release:

• summarize_layout does not return final range values #2895 (brushing in shiny doesn't work for coord_trans())

PRs to complete before release:

• Update documentation to meet CII badging requirements #2792
• Make graticule labeling configurable #2849 (needs add explicit parsing option to labels in coord_sf() #2881 first)
• Improve error messages about conflicted parameters #2854
• Implement sec.axis for date, time, and datetime scales #2806
• Force arguments that are used in closures #2807
• Document now unused argument #2808
• add explicit parsing option to labels in coord_sf() #2881
• Fix back-transformation of ranges in coords #2821 (fixes summarize_layout does not return final range values #2895, has been rewritten to not introduce breaking changes)

Prepare for release:

• devtools::check_win_devel()
• rhub::check_for_cran()
• revdepcheck::revdep_check(num_workers = 4)
• Polish NEWS
• If new failures, update email.yml then revdepcheck::revdep_email_maintainers()

Perform release:

• Bump version (in DESCRIPTION and NEWS)
• devtools::check_win_devel() (again!)
• devtools::submit_cran()
• Approve email

Wait for CRAN...

• pkgdown::build_site()
• Tag release
• Merge RC back to master branch
• Bump dev version
• Write blog post
• Tweet
• Add link to blog post in pkgdown news menu

Template from r-lib/usethis#338

[josesho]

Is #2788 going to be part of ggplot 3.0.1? If not, can I request that it be included? (Working on fixing mission-critical function, and this breaks it in a massive way.)

[clauswilke]

@josesho If your code works with the current github master then it should work with ggplot2 3.0.1. If it doesn't work with the current github master then please file an issue with reprex.

[josesho]

Thanks for the heads-up, the current master does work and fixes the issue. Hopefully y'all can ggplot2 3.0.1 out soon! Looking forward 💪🏻

[clauswilke]

@topepo Could you run revdep_check() on the current ggplot2 master branch, please? All the pending patches have been merged and we are ready to proceed.

@hadley Could you run devtools::check_win_devel()? It looks like only the package maintainer can do it. And could you also run rhub::check_for_cran()? I got some strange error messages about a missing sf package and about long file names, and I'm not sure how to interpret them.

[clauswilke]

Here are the build logs from my check_for_cran() run:
https://builder.r-hub.io/status/ggplot2_3.0.0.9000.tar.gz-e0789aadf85149dea1cdf0e69ef0b65d
https://builder.r-hub.io/status/ggplot2_3.0.0.9000.tar.gz-a1b23d0e97de472f96fbacfcd8a15538
https://builder.r-hub.io/status/ggplot2_3.0.0.9000.tar.gz-4ef752110db84da59ff24e7b12f239a9

I'll rerun with _R_CHECK_FORCE_SUGGESTS_=false and see what happens.

[batpigandme]

Per @jennybc's suggestion in the OG thread: r-lib/usethis#338 (comment), I think it does make sense to do the pkgdown build after CRAN approval

[clauswilke]

@batpigandme Thanks, I've moved it down.

[clauswilke]

@hadley Still getting rhub errors even with _R_CHECK_FORCE_SUGGESTS_=false:

* checking package dependencies ... ERROR
Package required but not available: 'viridisLite'

https://builder.r-hub.io/status/ggplot2_3.0.0.9000.tar.gz-a168b584c5be475a94a8ff7e8163c249
https://builder.r-hub.io/status/ggplot2_3.0.0.9000.tar.gz-b9766fe583ad49f69a8fdf0599a8fcc6
https://builder.r-hub.io/status/ggplot2_3.0.0.9000.tar.gz-fd4aa9d2c01d493a9bc7ea8383ca33cf

[yutannihilation]

It seems dependency installation fails. I see errors around line 509, though it's not clear which package is wrong.

Error in desc[1L, "Type"] : subscript out of bounds
Calls: ... force -> i.p -> .install.winbinary -> unpackPkgZip -> %in%
In addition: Warning messages:
1: In unzip(zipname, exdir = dest) :
write error in extracting from zip file
2: In unzip(zipname, exdir = dest) :
write error in extracting from zip file
Execution halted

This seems an error from here: https://github.com/wch/r-source/blob/521c90a175d67475b9f1b43d7ae68bc48062d8e6/src/library/utils/R/windows/install.packages.R#L58

[topepo]

Done. I'll put in a PR in a bit

Broken (15)

package	version	error	warning	note
ActisoftR	0.0.1	+1	1 +1
AneuFinder	1.8.0	-1 +1	1 +2	1
BloodCancerMultiOmics2017	1.0.2	-1	+2	1
dartR	1.0.5		1 +1	1
GenomicDataCommons	1.4.3	-1	2 +1	1
ggridges	0.5.0	+1	1 +1	1
GOexpress	1.14.0		1	4 +1
MCbiclust	1.4.0	+1	1	1
mglR	0.1.0	-1	+1
RITAN	1.4.2	-1	1 +1	3
scDD	1.4.0	+1	2	2
seqsetvis	1.0.2	+2	1 +1
shadowtext	0.0.2	+1	1 +1
tricolore	1.2.0	+1	1	2
tsbox	0.0.3	+1	1

[topepo]

revdep db here

[clauswilke]

15 packages are newly broken in revdeps. That's not too bad. I suspect most of them will be due to the aesthetic name standardization that we introduced. That's what's causing the problems in my ggridges package. I used an aesthetic point_color and it is now internally converted to point_colour and thus some parts of the code can't find it. I'll fix ggridges and see if I can write recommendations for best practice.

@hadley @batpigandme @thomasp85 @karawoo @dpseidel @yutannihilation If you have time to look at any of the other broken packages to see what the issues are that would be great.