format() will at least let us distinguish integer() from NULL.

Generated via commit 2eeb415

Download link for the artifact containing the test results: ↓ atime-results.zip

Time taken to finish the standard R installation steps: 11 minutes and 44 seconds

Time taken to run atime::atime_pkg on the tests: 3 minutes and 18 seconds

Maybe also add capabilities()["long.double"]?

I guess that's overkill, here's how that element of capabilities() is defined:

https://github.com/r-devel/r-svn/blob/e0f5c5e1c60bca2b7b2044eb0b4770e287cd770e/src/main/platform.c#L2627-L2628

Here's the LDOUBLE macro:

https://github.com/r-devel/r-svn/blob/1f537a2469956b505ad10135eeee282fe33a8475/src/include/Defn.h#L2407-L2411

In turn I think HAVE_LONG_DOUBLE gets set by ./configure.

Here's how .Machine$sizeof.longdouble is defined:

https://github.com/r-devel/r-svn/blob/1f537a2469956b505ad10135eeee282fe33a8475/src/main/machine.c#L139-L144

Here's the SIZEOF_LONG_DOUBLE macro:

https://github.com/r-devel/r-svn/blob/1f537a2469956b505ad10135eeee282fe33a8475/src/gnuwin32/fixed/h/config.h#L1182-L1191

So there might be some daylight between the two, WDYT?

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 98.69%. Comparing base (6f3fc8d) to head (fd41733).
Report is 778 commits behind head on master.

@@            Coverage Diff             @@
##           master    #6154      +/-   ##
==========================================
+ Coverage   97.51%   98.69%   +1.18%     
==========================================
  Files          80       79       -1     
  Lines       14979    14679     -300     
==========================================
- Hits        14607    14488     -119     
+ Misses        372      191     -181     

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.