r: check finds 'abort' and 'stderr' in Flight SQL driver

[lidavidm]

What happened?

* checking compiled code ... WARNING
Warning: File ‘adbcflightsql/libs/adbcflightsql.so’:
  Found ‘abort’, possibly from ‘abort’ (C)
    Object: ‘go/libadbc_driver_flightsql.a’
  Found ‘stderr’, possibly from ‘stderr’ (C)
    Object: ‘go/libadbc_driver_flightsql.a’

Compiled code should not call entry points which might terminate R nor
write to stdout/stderr instead of to the console, nor use Fortran I/O
nor system RNGs nor [v]sprintf.

Haven't looked at why; also should those .a files be there in the first place? Maybe we can just delete them?

Stack Trace

No response

How can we reproduce the bug?

No response

Environment/Setup

No response

[paleolimbot]

This was a new check that was added in R 4.5.0. It is, in my opinion, a check that embodies the reason why nobody wraps code for R...most external libraries have an abort or a stderr sitting around (it's just that the good ones provide a way to avoid them by setting a custom logger or something). For Go drivers we don't publish them to CRAN so it doesn't matter. I think the only way we can fix this by ignoring warnings for those drivers (because even if we clean up the static libraries after linking, adbcflightsql.so will still be there.

[lidavidm]

Ah thanks...that's rather annoying :/