Skip to content

Conversation

eddelbuettel
Copy link
Contributor

Armadillo 15.0.* now makes C++14 the minimum compilation standard, which also means I can no-longer add the suppression of deprecation warnings (which used a macro before and now use a C++14 or later attribute). For most packages, adapting to it can be very simple, and yours is one of them -- in fact it is just two statements. In the patch below we simply update this as now required by Armadillo 15.0.x. Now, your repo seems to have more arma::is_finite() calls and you may need to consider changing those too---but they don't appear to be 'hit' when building the R package.

Please see issues #475 and below at the RcppArmadillo GitHub repo for context, and notably #491 for this second wave of PRs and patches (following one for moving away from C++11, something your package does not need). It would be terrific if you could make an upload to CRAN 'soon' to remove the deprecation warning. Please do not hesitate to reach out if I can assist in any way or clarify matters.

@config-i1
Copy link
Owner

Hi Dirk,

Thanks for the PR and bringing this to my attention!

I'll amend my C++ functions respectively and will try to submit soon (I need to make some additional tests first).

@config-i1 config-i1 merged commit 7ba08a5 into config-i1:master Sep 22, 2025
3 checks passed
@eddelbuettel
Copy link
Contributor Author

eddelbuettel commented Sep 22, 2025

Hi Ivan,

Looks like your scheduled CI runs fell victim to this yesterday given the CRAN update, and that the green light is back so fingers crossed!

And apologies for the intrusion but this comes from 'upstream' as I can longer avoid the warnings coming from Armadillo upstream. And, of course, "in the long run" it's all good even if this particular ones leaves me scracthing my chine -- I think I would have made arma::is_finite() a wrapper around std::isfinite.

@config-i1
Copy link
Owner

No worries.

I checked the C++ code I have, and I don't see arma::is_infinte() anywhere else. So, I think this it fixed now. So, I'm on the way to CRAN now (after some minor additional tests).

@eddelbuettel
Copy link
Contributor Author

I concur. I must have confused this with something else, or was sloppy otherwise. There are a number of member functions is_finite but as long as these are not deprecated all is good. Sorry for the false alert! Trying to help I did the opposite and created confusion. My bad.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

2 participants