Fix R_UnboundValue NOTE

[Enchufa2]

Closes #1465.

For promises, we have here a temporary workaround, but this is going to come back at us, because we are using a bunch of functions (PRVALUE, PRENV...) that are not part of the API, even if we don't see a NOTE yet. The problem is that I don't see alternatives yet.

Checklist

• Code compiles correctly
• R CMD check still passes all tests
• Preferably, new tests were added which fail without the change
• Document the changes by file in ChangeLog

[eddelbuettel]

Looks good to me. It will get in the queue of reverse depends runs to do.

[kevinushey]

LGTM!

[eddelbuettel]

This branch constantlty times out at GH when running the battery of Docker containers. That ... has never happened before.

When I try to repro, I see htop with a line from tinytest hanging, and I when I kill it (with '9') it sits at

[...]
  test_function.R...............   17 tests OK 
  test_function.R...............   18 tests OK 
  test_function.R...............   19 tests OK Killed
* checking for unstated dependencies in vignettes ... OK
* checking package vignettes ... OK
* checking running R code from vignettes ... SKIPPED
* checking re-building of vignette outputs ... SKIPPED

so maybe one of these is regressing. An earlier one killed with 15 has

* checking for unstated dependencies in ‘tests’ ... OK
* checking tests ...
  Running ‘tinytest.R’
sh: 0: getcwd() failed: No such file or directory
Warning message:
In file.rename(outfile, paste0(outfile, ".fail")) :
  cannot rename file 'tinytest.Rout' to 'tinytest.Rout.fail', reason 'No such file or directory'
 ERROR
* checking for unstated dependencies in vignettes ... OK

but that may be due to having run two concurrently as root from the same directory.

[eddelbuettel]

It is these two lines / this test:

expect_error(function_cons_ns("sourceCpp", "Rcppp"),
             info = "namespace-lookup constructor: fail when ns does not exist")

[Enchufa2]

Good catch, fixed now!

BTW, and it freezes instead of failing? It may be a tinytest bug, right?

[eddelbuettel]

BTW, and it freezes instead of failing? It may be a tinytest bug, right?

I am not sure. I actually first went that route as I know Mark just updated tinytest for the first time in years, that made no difference. The test predicate is expect_error() and we did not supply one for R < 4.5.0 ... so what is it to do? But feel free to poke him at the repro. Maybe it should fail eventually? Not sure how easy that is to do.