Re: [R-pkg-devel] NOTEs in packages using Rust

2024-06-20 Thread Hiroaki Yutani
> savvy is opposed to supporting CRAN publication Let me correct a bit. While it's true **I** don't recommend CRAN for Rust-powered R packages unless you are an expert, savvy is not that opinionated. It does nothing to prevent you from submitting to CRAN. Best, Yutani 2024年6月21日(金) 1:28 Josiah

Re: [R-pkg-devel] NOTEs in packages using Rust

2024-06-20 Thread Balasubramanian Narasimhan
@JeffNewmiller, indeed, deleting the vendor directory after cargo does its job fixes things. (Don't know how I missed that!) @JosiahParry,  the documentation was helpful. As to why my rextendr version still suffered from the same problem: the version of clarabel on CRAN predated the current ve

Re: [R-pkg-devel] NOTEs in packages using Rust

2024-06-20 Thread Josiah Parry
I personally wrote the rextendr functions to handle this. If you follow the instructions below you most likely shouldn't have an issue. I have quite a number of extendr based R package on CRAN. https://extendr.github.io/rextendr/dev/articles/cran-compliance.html On Thu, Jun 20, 2024 at 12:40 PM B

Re: [R-pkg-devel] NOTEs in packages using Rust

2024-06-20 Thread Balasubramanian Narasimhan
I think your answer suggests the nub of solution: to delete the offending vendor directory at the right time. Will try. @JosiahParry, I just remembered that I also have an extendr version that suffers from a similar affliction. Thanks. -Naras Balasubramanian Narasimhan _

Re: [R-pkg-devel] NOTEs in packages using Rust

2024-06-20 Thread Josiah Parry
To be clear, you need to delete the untarred folders after the build process. You only need the dependencies at build time. After the build time it has been compiled into a .so/.dll and then you must delete the vendored sources. There's no reason that installing an R package should keep the *source

Re: [R-pkg-devel] NOTEs in packages using Rust

2024-06-20 Thread Jeff Newmiller via R-package-devel
I am not connected to CRAN (by a long shot), but if I were I would not accept any of the arguments you are making here. I don't know what you are talking about with respect to excluding files in tarring... you should be using the package build processes as documented in Writing R Extensions. A

Re: [R-pkg-devel] NOTEs in packages using Rust

2024-06-20 Thread Balasubramanian Narasimhan
I’m using savvy but the version on CRAN uses extendr. Could you clarify? I’m not sure I know at which point to intervene in Makevars to fix things myself… Thanks. -Naras Balasubramanian Narasimhan From: Josiah Parry Sent: Thursday, June 20, 2024 9:08:10 AM To:

Re: [R-pkg-devel] NOTEs in packages using Rust

2024-06-20 Thread Josiah Parry
Hi Naras, then I can't help you there! Last I checked, savvy is opposed to supporting CRAN publication On Thu, Jun 20, 2024 at 12:19 PM Balasubramanian Narasimhan < na...@stanford.edu> wrote: > I’m using savvy but the version on CRAN uses extendr. > > Could you clarify? > > I’m not sure I know at

Re: [R-pkg-devel] NOTEs in packages using Rust

2024-06-20 Thread Josiah Parry
Hi Naras, this should be handled by the Makevars itself. Are you using extendr to do this? Using `rextendr::use_cran_defaults()` should handle it for you. Then you can use `rextendr::vendor_pkgs()` to handle vendoring. On Thu, Jun 20, 2024 at 11:57 AM Balasubramanian Narasimhan < na...@stanford.

[R-pkg-devel] NOTEs in packages using Rust

2024-06-20 Thread Balasubramanian Narasimhan
I am preparing an update of Clarabel to synchronize with the upstream repo. When checking the package, I get NOTEs such as: Found the following CITATION file in a non-standard place:   src/rust/vendor/clarabel/CITATION.bib Most likely ‘inst/CITATION’ should be used instead. Foun

Re: [R-pkg-devel] CRAN check servers terminating on stanExports_*.o files

2024-06-20 Thread Ivan Krylov via R-package-devel
В Sat, 15 Jun 2024 15:11:36 + "Slager, Dave" пишет: > The r-patched-linux-x86_64 and/or r-release-linux-x86_64 CRAN check > servers seem to be erroring out on multiple packages, including > these, for similar reasons: > > bakR > baldur > bayes4psy > BayesGmed > ctsem > eggCounts > FlexReg >