[Bioc-devel] Warnings in "checking compiled code" of R CMD check

[Oleksii Nikolaienko]

Hi all,
it seems that the devel version of R CMD check has raised the level of
messages in "checking compiled code" from NOTE to WARNING. My package
contains

entry points for abort, exit, stdout, stderr and sprintf, while I use only
snprintf and don't terminate or write out anything in a non recommended
way. I can see similar warnings in checks of Rsamtools and rtracklayer as
well.
Is there anything one needs to do about this? Would it become a problem in
the future?

Best,
Oleksii

[[alternative HTML version deleted]]

___
Bioc-devel@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/bioc-devel

Re: [Bioc-devel] Warnings in "checking compiled code" of R CMD check

[Hervé Pagès]

Hi Oleksii,

I was contacted off-list by the maintainer of the FLAMES package about 
this. The FLAMES package has the same "checking compiled code" WARNING.

I'm just copying my answer below.

Best,

H.

-

Yes the warning for the C code in Rhtslib actually propagates to any 
package that is linked to Rhtslib (via the LinkingTo field). See for 
example the CHECK result for Rsamtools:

https://bioconductor.org/checkResults/3.17/bioc-LATEST/Rsamtools/nebbiolo1-checksrc.html

This is because Rsamtools.so and FLAMES.so are statically linked to 
Rhtslib, so even if the C code in Rsamtools and FLAMES don't use these 
forbidden symbols, they still end up in Rsamtools.so and FLAMES.so.

Nothing can be done on Rsamtools' or FLAMES' side about this, except 
using dynamic linking. But this is discouraged because it introduces a 
whole set of other problems.

So a proper solution would need to originate in the Rhtslib package 
itself. However note that these symbols are not known to actually cause 
problems in practice so the situation is not that bad.

All this to say that it's ok to ignore the warning for FLAMES.

Thanks for your contribution to Bioconductor.

-

On 26/01/2023 06:00, Oleksii Nikolaienko wrote:
> Hi all,
> it seems that the devel version of R CMD check has raised the level of
> messages in "checking compiled code" from NOTE to WARNING. My package
> contains
> 
> entry points for abort, exit, stdout, stderr and sprintf, while I use only
> snprintf and don't terminate or write out anything in a non recommended
> way. I can see similar warnings in checks of Rsamtools and rtracklayer as
> well.
> Is there anything one needs to do about this? Would it become a problem in
> the future?
>
> Best,
> Oleksii
>
>   [[alternative HTML version deleted]]
>
> ___
> Bioc-devel@r-project.org  mailing list
> https://stat.ethz.ch/mailman/listinfo/bioc-devel

-- 
Hervé Pagès

Bioconductor Core Team
hpages.on.git...@gmail.com

[[alternative HTML version deleted]]

___
Bioc-devel@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/bioc-devel