[Rd] R CMD check and CRAN's Rust policy

Ben Bolker bbo|ker @end|ng |rom gm@||@com
Mon Mar 31 19:48:05 CEST 2025



On 2025-03-31 1:04 p.m., Duncan Murdoch wrote:
> On 2025-03-31 12:41 p.m., Josiah Parry wrote:
>> Duncan, the changes to symbols checking was introduced March 22nd see 
>> https://bugs.r-project.org/show_bug.cgi?id=18789 <https://bugs.r- 
>> project.org/show_bug.cgi?id=18789> and https://developer.r- 
>> project.org/blosxom.cgi/R-devel/NEWS/2025/03/22#n2025-03-22 <https:// 
>> developer.r-project.org/blosxom.cgi/R-devel/ 
>> NEWS/2025/03/22#n2025-03-22>. But that is unrelated.
> 
> Sorry, I missed that.
> 
>>
>> To Tim's comment—the check is a simple grep of the installation log 
>> for "Downloading crates." This could be easily circumvented on CRAN 
>> and locally by suppressing stdout/err. But that would be adversarial 
>> and I would like to adhere to the intent of the check.
> 
> I think Tim was suggesting that you modify your Github action to ignore 
> this particular warning.  The warning would still appear, but it 
> wouldn't cause a check failure.
> 
> Duncan Murdoch

   At a very quick look, I don't see an easy way to do that (but I am 
admittedly really bad at doing stuff with Github actions). Maybe longer 
term, but it feels like the best way to do this would be to request a 
feature in `rcmdcheck` that allowed the user to request ignoring 
specific warnings (e.g. specified by regexp?), then expose that feature 
in the r-check-package action (or whatever it's called ...)

> 
> 
> 
>>
>>
>>
>> On Mon, Mar 31, 2025 at 9:23 AM Duncan Murdoch 
>> <murdoch.duncan using gmail.com <mailto:murdoch.duncan using gmail.com>> wrote:
>>
>>     On 2025-03-31 11:50 a.m., Josiah Parry wrote:
>>      > Following up with this as I address the new R-devel "Compiled code
>>      > should not call entry points which might terminate R" WARNING and
>>     this
>>      > issue has reared its head again.
>>      >
>>      > Would a path forward be an environment variable similar
>>      > to _R_CHECK_CRAN_INCOMING_ to skip this check primarily for GitHub
>>      > Actions and CI?
>>
>>     The "Compiled code should not call entry points which might
>>     terminate R"
>>     isn't a new warning.  I think the last change to it was made in 2022.
>>
>>     Maybe your code, or code in one of the libraries you use, has 
>> changed?
>>
>>     Duncan Murdoch
>>
>>
>>
>>
>>      >
>>      > Or, alternatively, if this could be a NOTE when the `--as-cran` 
>> flag
>>      > isn't set but a WARNING when it is?
>>      >
>>      > Re-vendoring dependencies each time they are changed during the
>>      > development lifecycle is quite a bit. However, vendoring once
>>     prior to
>>      > publishing makes good sense.
>>      >
>>      > Is there a balance we can strike here to lower development
>>     friction but
>>      > also ensure the robust package installation requirements when
>>     expected?
>>      >
>>      > Using
>>      >
>>      >
>>      >
>>      >
>>      > On Sun, Mar 2, 2025 at 11:42 AM Duncan Murdoch
>>     <murdoch.duncan using gmail.com <mailto:murdoch.duncan using gmail.com>
>>      > <mailto:murdoch.duncan using gmail.com
>>     <mailto:murdoch.duncan using gmail.com>>> wrote:
>>      >
>>      >     On 2025-03-02 1:09 p.m., Ben Bolker wrote:
>>      >      >    I, like Duncan, am just following along here. I think 
>> there
>>      >     might be
>>      >      > two distinct questions which it would be useful to keep
>>     distinct:
>>      >      >
>>      >      >    * how to silence the rust-check if desired?
>>      >      >
>>      >      >     rather than debating whether the rust-check should be
>>     always-on,
>>      >      > on-for-CRAN-only, etc., would it provide for useful
>>     flexibility
>>      >     to add
>>      >      > an environment variable that enables/disables this
>>      >     functionality?  There
>>      >      > are already 168 of these environment variables, how much 
>> would
>>      >     one more
>>      >      > cost?
>>      >
>>      >     I may have misunderstood Josiah.  I thought his message said
>>     that it is
>>      >     already easy to silence the check, by stopping the code from
>>     issuing
>>      >     the
>>      >     message the check is looking for.
>>      >
>>      >     Presumably the package shouldn't do that, but if there's an
>>     environment
>>      >     variable that can be set to do it, then the repository or
>>     user can
>>      >     choose to do it, so there's no need for R to add another
>>     environment
>>      >     variable.
>>      >
>>      >     BTW, as far as I can see current R-devel doesn't issue an
>>     error, it
>>      >     just
>>      >     issues warnings about two issues:
>>      >
>>      >        - the package is downloading crates
>>      >        - the rustc compiler doesn't report a version number
>>      >
>>      >     Duncan Murdoch
>>      >
>>      >      >
>>      >      >     I'm not sure how adding an environment variable to
>>     allow easier
>>      >      > user/alternate-repository control of the check is 
>> "against the
>>      >     spirit of
>>      >      > the check" ...
>>      >      >
>>      >      >     All the existing check-regulating env variables ...
>>      >      >
>>      >      > cd src/library/tools/R
>>      >      > grep 'Sys.getenv("_R_CHECK' * | sed -e
>>     's/^.*Sys.getenv(//' | sed -e
>>      >      > 's/[,)].*//' | sort | uniq | wc
>>      >      >
>>      >      >
>>      >      >     * should CRAN allow Rust crates to be downloaded?
>>      >      >
>>      >      >     This is a much more fundamental policy decision, which
>>     I have no
>>      >      > opinion about.
>>      >      >
>>      >      >     cheers
>>      >      >      Ben Bolker
>>      >      >
>>      >      >
>>      >      >
>>      >      >
>>      >      > On 2025-03-02 12:21 p.m., Duncan Murdoch wrote:
>>      >      >> On 2025-03-02 11:03 a.m., Josiah Parry wrote:
>>      >      >>> Well this has surely veered off course!
>>      >      >>>
>>      >      >>> As the one who filed the BugZilla report, I'd like to
>>     redirect the
>>      >      >>> conversation and provide further context.
>>      >      >>>
>>      >      >>> The question should be /"how do we get a dialogue
>>     started on this
>>      >      >>> bugzilla issue before the next minor /
>>      >      >>> /release of R?"/
>>      >      >>
>>      >      >> Isn't this exactly that dialogue?
>>      >      >>
>>      >      >>>
>>      >      >>> The current check for Rust-based R package's downloading
>>     external
>>      >      >>> dependencies works by looking at
>>      >      >>> the output logs for the presence of  "Downloading
>>     crates." This
>>      >     can is
>>      >      >>> an entirely fine requirement for
>>      >      >>> CRAN—however, due to the fact that it is an error, 
>> packages
>>      >      >>> distributed through other repositories
>>      >      >>> fail the R-CMD check.
>>      >      >>
>>      >      >> I think you misunderstood me.  CRAN shares the view I
>>     gave that you
>>      >      >> should be able to run old code to reproduce old 
>> results, but
>>      >     they aren't
>>      >      >> the only ones.  That's always been a goal of R.
>>      >      >>
>>      >      >>> Folks who use R-universe or PPM or some mysterious third
>>     thing
>>      >     may not
>>      >      >>> share the same philosophy as
>>      >      >>> CRAN and prefer the convenience of fetching the
>>     dependencies at
>>      >      >>> compile time and not vendoring them.
>>      >      >>> An alternative would be for the check to be optionally
>>     skipped or
>>      >      >>> become a NOTE when the CRAN
>>      >      >>> flag is not set and an ERROR otherwise. Skipping this 
>> CRAN
>>      >     check is as
>>      >      >>> easy as adding `--quiet`
>>      >      >>> or setting an environment variable—but that is against 
>> the
>>      >     spirit of
>>      >      >>> the check.
>>      >      >>
>>      >      >> If it is that easy to skip the check, then I really 
>> don't see
>>      >     the issue.
>>      >      >>    Just ask the repository where you want to put your
>>     package to
>>      >     put that
>>      >      >> option or environment variable in place, and there's no
>>     longer a
>>      >     problem.
>>      >      >>
>>      >      >> Duncan Murdoch
>>      >      >>
>>      >      >>> Ideally, the check can remain, but scoped appropriately.
>>      >      >>>
>>      >      >>>
>>      >      >>> On Sun, Mar 2, 2025 at 6:49 AM Duncan Murdoch
>>      >      >>> <murdoch.duncan using gmail.com
>>     <mailto:murdoch.duncan using gmail.com> <mailto:murdoch.duncan using gmail.com
>>     <mailto:murdoch.duncan using gmail.com>>
>>      >     <mailto:murdoch.duncan using gmail.com
>>     <mailto:murdoch.duncan using gmail.com> <mailto:murdoch.duncan using gmail.com
>>     <mailto:murdoch.duncan using gmail.com>>>>
>>      >     wrote:
>>      >      >>>
>>      >      >>>      You seem to be taking a confontational tone, which
>>     isn't
>>      >     likely to
>>      >      >>>      encourage a reasonable dialogue.
>>      >      >>>
>>      >      >>>      I've looked for other messages on this, and didn't
>>     see any
>>      >     besides
>>      >      >>> this
>>      >      >>>      one explaining why including check_rust() in the
>>     checks is
>>      >     a problem.
>>      >      >>>      The problem you talk about here is that it 
>> encourages
>>      >     vendoring,
>>      >      >>> which
>>      >      >>>      makes it harder for package authors to count 
>> downloads.
>>      >      >>>
>>      >      >>>      To be honest, that doesn't seem like a very serious
>>      >     problem.  I
>>      >      >>> assume
>>      >      >>>      the packages ("crates") we are talking about are 
>> open
>>      >     source, so
>>      >      >>>      this is
>>      >      >>>      entirely in the spirit of how they are allowed to be
>>      >     distributed.
>>      >      >>>
>>      >      >>>      If they aren't open source, then users of those
>>     packages
>>      >     should be
>>      >      >>>      warned about that, and a check failure is a good
>>     way to do
>>      >     that.
>>      >      >>>
>>      >      >>>      So you need to explain why it is important to be
>>     able to
>>      >     download and
>>      >      >>>      install software and not be warned about it.
>>      >      >>>
>>      >      >>>      I am not in R Core or CRAN, but I can suggest why 
>> it is
>>      >     better to
>>      >      >>>      include source in the package:  it makes the use of
>>     that
>>      >     package more
>>      >      >>>      reliable in the future.  It's not uncommon to run 
>> an R
>>      >     computation
>>      >      >>> that
>>      >      >>>      was written a few years ago.  Sometimes libraries 
>> or R
>>      >     have changed,
>>      >      >>>      and
>>      >      >>>      a user will need to go back to a previous version to
>>      >     reproduce the
>>      >      >>>      calculation.  Being able to able to rebuild a
>>     system as it
>>      >     would have
>>      >      >>>      been back then is important.
>>      >      >>>
>>      >      >>>      Is that possible if the package needs to make a
>>     download?  The
>>      >      >>> download
>>      >      >>>      site that worked a few years ago may no longer
>>     exist.  If
>>      >     the site
>>      >      >>>      exists, the code versions there may be different.
>>      >      >>>
>>      >      >>>      Those are some of the issues that Simon was
>>     alluding to.
>>      >      >>>
>>      >      >>>      Duncan Murdoch
>>      >      >>>
>>      >      >>>
>>      >      >>>
>>      >      >>>      On 2025-03-02 5:45 a.m., Mossa Merhi Reimert via
>>     R-devel
>>      >     wrote:
>>      >      >>>       > Dear Simon Urbanek,
>>      >      >>>       >
>>      >      >>>       > There has been very little engagement with the
>>     issue I
>>      >     referred
>>      >      >>>      to. If it was decided that this “check” ought to be
>>     part
>>      >     of the
>>      >      >>>      default checks for R,
>>      >      >>>       > then that could have been written to us. Either
>>     on the
>>      >      >>> bugs.r-project.org <http://bugs.r-project.org>
>>     <http://bugs.r-project.org <http://bugs.r-project.org>>
>>      >     <http://bugs.r-project.org <http://bugs.r-project.org>
>>     <http://bugs.r-project.org <http://bugs.r-project.org>>> or the 
>> proposed
>>      >      >>>      patch. Before we talk about anything else,
>>      >      >>>       > it does seem very strange that we cannot get a
>>     reasonable
>>      >      >>>      dialogue going.
>>      >      >>>       >
>>      >      >>>       > I would like to say that the R/Rust community
>>     has grown
>>      >      >>>      substantially. From my end, there are 3 bindings
>>     project,
>>      >     extendr,
>>      >      >>>      savvy, and roxido.
>>      >      >>>       > Then, there are now many rust-based packages on
>>     CRAN,
>>      >     see this
>>      >      >>>      most recent compiled list
>>      > https://github.com/nanxstats/r-rust-pkgs
>>     <https://github.com/nanxstats/r-rust-pkgs>
>>      >     <https://github.com/nanxstats/r-rust-pkgs
>>     <https://github.com/nanxstats/r-rust-pkgs>>
>>      >      >>>      <https://github.com/nanxstats/r-rust-pkgs
>>     <https://github.com/nanxstats/r-rust-pkgs>
>>      >     <https://github.com/nanxstats/r-rust-pkgs
>>     <https://github.com/nanxstats/r-rust-pkgs>>>.
>>      >      >>>       > There is also proof-of-concept
>>      >      >>> https://github.com/r-rust/hellorust
>>     <https://github.com/r-rust/hellorust>
>>      >     <https://github.com/r-rust/hellorust
>>     <https://github.com/r-rust/hellorust>>
>>      >      >>>      <https://github.com/r-rust/hellorust
>>     <https://github.com/r-rust/hellorust>
>>      >     <https://github.com/r-rust/hellorust
>>     <https://github.com/r-rust/hellorust>>> that integrates `cargo`,
>>      >      >>>      rust’s official build system, with R’s package
>>     build system,
>>      >      >>>       > and https://github.com/extendr/hellorustc
>>     <https://github.com/extendr/hellorustc>
>>      >     <https://github.com/extendr/hellorustc
>>     <https://github.com/extendr/hellorustc>>
>>      >      >>>      <https://github.com/extendr/hellorustc
>>     <https://github.com/extendr/hellorustc>
>>      >     <https://github.com/extendr/hellorustc
>>     <https://github.com/extendr/hellorustc>>>, which showcases how Rust
>>      >      >>>      compiler could be directly linked with R’s package
>>     system.
>>      >      >>>       >
>>      >      >>>       >   Let me say, that the current R CMD check is
>>     not meant
>>      >     to be
>>      >      >>>      “helpful”. When a package is built, `cargo` tells
>>     the user
>>      >      >>>      “Downloading crates”.
>>      >      >>>       > Thus, this information is already conveyed to
>>     the user.
>>      >      >>>       >
>>      >      >>>       > Personally, I do wish we could debate this
>>     requirement
>>      >     further. I
>>      >      >>>      do not believe that having R-packages on CRAN
>>     vendor rust
>>      >      >>> dependencies
>>      >      >>>       > as a good policy. Download statistics is a 
>> success
>>      >     metric of a
>>      >      >>>      given r-package and rust packages. By insisting on
>>      >     vendoring, and
>>      >      >>> thus
>>      >      >>>       > side-stepping `cargo` / crates.io
>>     <http://crates.io> <http://crates.io <http://crates.io>>
>>      >     <http://crates.io <http://crates.io> <http://crates.io
>>     <http://crates.io>>>, we are
>>      >      >>>      robbing upstream authors of their download-numbers.
>>     I do
>>      >     not think
>>      >      >>>      such policy is honourable.
>>      >      >>>       >
>>      >      >>>       > While C/C++ do not have official package
>>     repositories,
>>      >     it could
>>      >      >>>      be thought of, as fair game, to have CRAN act as a
>>     pseudo
>>      >     package
>>      >      >>>      manager for C/C++ libraries.
>>      >      >>>       > I’m not going to argue for or against this part.
>>      >      >>>       >
>>      >      >>>       > There are many objections from the CRAN side to
>>     all things
>>      >      >>>      related to Rust. I don’t want to open multiple
>>     topics in
>>      >     the same
>>      >      >>>      thread.
>>      >      >>>       > But there is plenty to bring up. And I had 
>> hoped we
>>      >     could talk
>>      >      >>>      this little issue through, before embarking on a 
>> larger
>>      >     discussion.
>>      >      >>>       > I do not appreciate the “random demands” 
>> comment, as
>>      >     this is not
>>      >      >>>      a demand, nor is it random.
>>      >      >>>       > I have inquired my end of the community for
>>     suggestions
>>      >      >>>       > to compile a larger proposal, but then I was 
>> afraid
>>      >     that this
>>      >      >>>      would be perceived as a big, bulky demand.
>>      >      >>>       >
>>      >      >>>       > Rust is not C/C++/Java, and the support for Rust
>>     cannot
>>      >     look like
>>      >      >>>      the support for these languages.
>>      >      >>>       >
>>      >      >>>       >
>>      >      >>>       >
>>      >      >>>       > From: Simon Urbanek <simon.urbanek using R-project.org>
>>      >      >>>       > Date: Sunday, 2 March 2025 at 00.39
>>      >      >>>       > To: Mossa Merhi Reimert <mossa using sund.ku.dk
>>     <mailto:mossa using sund.ku.dk>
>>      >     <mailto:mossa using sund.ku.dk <mailto:mossa using sund.ku.dk>>
>>      >      >>> <mailto:mossa using sund.ku.dk <mailto:mossa using sund.ku.dk>
>>     <mailto:mossa using sund.ku.dk <mailto:mossa using sund.ku.dk>>>>
>>      >      >>>       > Cc: r-devel using r-project.org
>>     <mailto:r-devel using r-project.org>
>>      >     <mailto:r-devel using r-project.org <mailto:r-devel using r-project.org>>
>>     <mailto:r-devel using r-project.org <mailto:r-devel using r-project.org>
>>      >     <mailto:r-devel using r-project.org <mailto:r-devel using r-project.org>>>
>>      >      >>>      <r-devel using r-project.org
>>     <mailto:r-devel using r-project.org> <mailto:r-devel using r-project.org
>>     <mailto:r-devel using r-project.org>>
>>      >     <mailto:r-devel using r-project.org <mailto:r-devel using r-project.org>
>>     <mailto:r-devel using r-project.org <mailto:r-devel using r-project.org>>>>
>>      >      >>>       > Subject: Re: [Rd] R CMD check and CRAN's Rust 
>> policy
>>      >      >>>       > [Du får ikke ofte mails fra
>>     simon.urbanek using r-project.org <mailto:simon.urbanek using r-project.org>
>>      >     <mailto:simon.urbanek using r-project.org
>>     <mailto:simon.urbanek using r-project.org>>
>>      >      >>>      <mailto:simon.urbanek using r-project.org
>>     <mailto:simon.urbanek using r-project.org>
>>      >     <mailto:simon.urbanek using r-project.org
>>     <mailto:simon.urbanek using r-project.org>>>. Få mere at vide om, hvorfor
>>      >      >>>      dette er vigtigt, på
>>      > https://aka.ms/LearnAboutSenderIdentification
>>     <https://aka.ms/LearnAboutSenderIdentification>
>>      >     <https://aka.ms/LearnAboutSenderIdentification
>>     <https://aka.ms/LearnAboutSenderIdentification>>
>>      >      >>>      <https://aka.ms/LearnAboutSenderIdentification
>>     <https://aka.ms/LearnAboutSenderIdentification>
>>      >     <https://aka.ms/LearnAboutSenderIdentification
>>     <https://aka.ms/LearnAboutSenderIdentification>>> ]
>>      >      >>>       >
>>      >      >>>       > Mossa,
>>      >      >>>       >
>>      >      >>>       > the issue you cite is lacking any pertinent
>>     information
>>      >     and it's
>>      >      >>>      not even clear why it should be an issue. The 
>> check is
>>      >     perfectly
>>      >      >>>      justified, it just reports whether a package 
>> using rust
>>      >     declares
>>      >      >>>      this correctly and where it downloads 3rd party
>>     content -
>>      >     something
>>      >      >>>      that is important to R users in general and not
>>     related to
>>      >     CRAN. I
>>      >      >>>      don't see how any of this is "prohibitive" it just
>>     calls
>>      >     out what
>>      >      >>>      the package is already doing.
>>      >      >>>       >
>>      >      >>>       > As discussed before, my hope was that the "R"ust
>>      >     community will
>>      >      >>>      mature enough to work on proper support. It is not
>>     clear
>>      >     that it
>>      >      >>>      happened yet, but once it does it would make sense
>>     to talk
>>      >     about
>>      >      >>>      support just as we have for C, C++ and Java, so
>>     certainly that
>>      >      >>>      should be the right discussion. However, it will
>>     have to
>>      >     start with
>>      >      >>>      some thinking and a proposal on how the associated
>>     issues
>>      >     (compiler
>>      >      >>>      support, versioning, dependency sources etc.) are 
>> to be
>>      >     addressed,
>>      >      >>>      as opposed to making random demands. All this has
>>     nothing
>>      >     to do with
>>      >      >>>      CRAN so the issue you mention seems irrelevant to 
>> the
>>      >     progress. Also
>>      >      >>>      I'd like to know what are the "challenges 
>> embedded in R
>>      >     itself".
>>      >      >>>       >
>>      >      >>>       > Cheers,
>>      >      >>>       > Simon
>>      >      >>>       >
>>      >      >>>       >
>>      >      >>>       >> On Mar 2, 2025, at 8:45 AM, Mossa Merhi 
>> Reimert via
>>      >     R-devel
>>      >      >>>      <r-devel using r-project.org
>>     <mailto:r-devel using r-project.org> <mailto:r-devel using r-project.org
>>     <mailto:r-devel using r-project.org>>
>>      >     <mailto:r-devel using r-project.org <mailto:r-devel using r-project.org>
>>     <mailto:r-devel using r-project.org <mailto:r-devel using r-project.org>>>> 
>> wrote:
>>      >      >>>       >>
>>      >      >>>       >> Hello everyone!
>>      >      >>>       >>
>>      >      >>>       >> I'm Mossa, I'm one of the maintainers of
>>     extendr, an
>>      >     automated
>>      >      >>>      generation of bindings project for
>>      >      >>>       >> Rust code, for use in R-packages.
>>      >      >>>       >>
>>      >      >>>       >> I'm writing to you, as R 4.4.3 was just
>>     released, and
>>      >     there have
>>      >      >>>      not been
>>      >      >>>       >> follow-up on an issue important to us. Link 
>> to the
>>      >     issue as
>>      >      >>>      discussed on r-devel
>>      >      >>>       >>
>>      > https://stat.ethz.ch/pipermail/r-devel/2024-October/083666.html
>>     <https://stat.ethz.ch/pipermail/r-devel/2024-October/083666.html>
>>      >       <https://stat.ethz.ch/pipermail/r-devel/2024- 
>> October/083666.html
>>     <https://stat.ethz.ch/pipermail/r-devel/2024-October/083666.html>>
>>      >      >>>
>>      >       <https://stat.ethz.ch/pipermail/r-devel/2024- 
>> October/083666.html
>>     <https://stat.ethz.ch/pipermail/r-devel/2024-October/083666.html>
>>      >       <https://stat.ethz.ch/pipermail/r-devel/2024- 
>> October/083666.html
>>     <https://stat.ethz.ch/pipermail/r-devel/2024-October/083666.html>>>
>>      >      >>>       >>
>>      >      >>>       >> A community member has provided a suggestion 
>> to a
>>      >     patch here
>>      >      >>> https://github.com/r-devel/r-svn/pull/182
>>     <https://github.com/r-devel/r-svn/pull/182>
>>      >     <https://github.com/r-devel/r-svn/pull/182
>>     <https://github.com/r-devel/r-svn/pull/182>>
>>      >      >>>      <https://github.com/r-devel/r-svn/pull/182
>>     <https://github.com/r-devel/r-svn/pull/182>
>>      >     <https://github.com/r-devel/r-svn/pull/182
>>     <https://github.com/r-devel/r-svn/pull/182>>>, and we have also
>>      >      >>>      attempted to bring it up on
>>      >      >>>       >> Bugzilla:
>>      > https://bugs.r-project.org/show_bug.cgi?id=18806
>>     <https://bugs.r-project.org/show_bug.cgi?id=18806>
>>      >     <https://bugs.r-project.org/show_bug.cgi?id=18806
>>     <https://bugs.r-project.org/show_bug.cgi?id=18806>>
>>      >      >>>      <https://bugs.r-project.org/show_bug.cgi?id=18806
>>     <https://bugs.r-project.org/show_bug.cgi?id=18806>
>>      >     <https://bugs.r-project.org/show_bug.cgi?id=18806
>>     <https://bugs.r-project.org/show_bug.cgi?id=18806>>>
>>      >      >>>       >>
>>      >      >>>       >> TLDR: Default `R CMD check` uses additional
>>      >     CRAN-specific checks
>>      >      >>>      for Rust,
>>      >      >>>       >> instead of keeping this behind the --as-cran 
>> flag.
>>      >      >>>       >>
>>      >      >>>       >> I would like to say, that there is a growing
>>     interest
>>      >     in Rust
>>      >      >>>      within the R community.
>>      >      >>>       >> And generally, Rust becoming a widely adopted
>>     language
>>      >     within
>>      >      >>>      the Python community (including the scientific part
>>     of that
>>      >      >>>      community). It is time to deal with the
>>      >      >>>       >> pain points with using Rust in R.
>>      >      >>>       >>
>>      >      >>>       >> Therefore, I would kindly ask that we have a
>>     dialogue
>>      >     on how to
>>      >      >>>      remedy the issue above first, and how we may deal 
>> with
>>      >     other issues
>>      >      >>>      going forward. There are both challenges embedded 
>> in R
>>      >     itself, and
>>      >      >>>      the current CRAN policy for Rust is prohibitive.
>>      >      >>>       >>
>>      >      >>>       >>
>>      >      >>>       >>
>>      >      >>>       >> Mossa Merhi Reimert
>>      >      >>>       >> Postdoctoral Researcher
>>      >      >>>       >>
>>      >      >>>       >> K�benhavns Universitet
>>      >      >>>       >> Department of Veterinary and Animal Sciences
>>      >      >>>       >> Animal Welfare and Disease Control
>>      >      >>>       >> Gr�nneg�rdsvej 8
>>      >      >>>       >> 1870 Frederiksberg C
>>      >      >>>       >> Denmark
>>      >      >>>       >>
>>      >      >>>       >> +45 35324135
>>      >      >>>       >> mossa using sund.ku.dk <mailto:mossa using sund.ku.dk>
>>     <mailto:mossa using sund.ku.dk <mailto:mossa using sund.ku.dk>>
>>      >      >>>      <mailto:mossa using sund.ku.dk <mailto:mossa using sund.ku.dk>
>>      >     <mailto:mossa using sund.ku.dk
>>     <mailto:mossa using sund.ku.dk>>><mailto:mossa using sund.ku.dk
>>     <mailto:mossa using sund.ku.dk>
>>      >     <mailto:mossa using sund.ku.dk <mailto:mossa using sund.ku.dk>>
>>      >      >>>      <mailto:mossa using sund.ku.dk <mailto:mossa using sund.ku.dk>
>>     <mailto:mossa using sund.ku.dk <mailto:mossa using sund.ku.dk>>>>
>>      >      >>>       >>
>>      >      >>>       >>
>>      >      >>>       >>        [[alternative HTML version deleted]]
>>      >      >>>       >>
>>      >      >>>       >> ______________________________________________
>>      >      >>>       >> R-devel using r-project.org
>>     <mailto:R-devel using r-project.org> <mailto:R-devel using r-project.org
>>     <mailto:R-devel using r-project.org>>
>>      >     <mailto:R-devel using r-project.org <mailto:R-devel using r-project.org>
>>     <mailto:R-devel using r-project.org <mailto:R-devel using r-project.org>>>
>>      >     mailing list
>>      >      >>>       >> https://stat.ethz.ch/mailman/listinfo/r-devel
>>     <https://stat.ethz.ch/mailman/listinfo/r-devel>
>>      >     <https://stat.ethz.ch/mailman/listinfo/r-devel
>>     <https://stat.ethz.ch/mailman/listinfo/r-devel>>
>>      >      >>>      <https://stat.ethz.ch/mailman/listinfo/r-devel
>>     <https://stat.ethz.ch/mailman/listinfo/r-devel>
>>      >     <https://stat.ethz.ch/mailman/listinfo/r-devel
>>     <https://stat.ethz.ch/mailman/listinfo/r-devel>>>
>>      >      >>>       >
>>      >      >>>       >       [[alternative HTML version deleted]]
>>      >      >>>       >
>>      >      >>>       > ______________________________________________
>>      >      >>>       > R-devel using r-project.org
>>     <mailto:R-devel using r-project.org> <mailto:R-devel using r-project.org
>>     <mailto:R-devel using r-project.org>>
>>      >     <mailto:R-devel using r-project.org <mailto:R-devel using r-project.org>
>>     <mailto:R-devel using r-project.org <mailto:R-devel using r-project.org>>>
>>      >     mailing list
>>      >      >>>       > https://stat.ethz.ch/mailman/listinfo/r-devel
>>     <https://stat.ethz.ch/mailman/listinfo/r-devel>
>>      >     <https://stat.ethz.ch/mailman/listinfo/r-devel
>>     <https://stat.ethz.ch/mailman/listinfo/r-devel>>
>>      >      >>>      <https://stat.ethz.ch/mailman/listinfo/r-devel
>>     <https://stat.ethz.ch/mailman/listinfo/r-devel>
>>      >     <https://stat.ethz.ch/mailman/listinfo/r-devel
>>     <https://stat.ethz.ch/mailman/listinfo/r-devel>>>
>>      >      >>>
>>      >      >>>      ______________________________________________
>>      >      >>> R-devel using r-project.org <mailto:R-devel using r-project.org>
>>     <mailto:R-devel using r-project.org <mailto:R-devel using r-project.org>>
>>      >     <mailto:R-devel using r-project.org <mailto:R-devel using r-project.org>
>>     <mailto:R-devel using r-project.org <mailto:R-devel using r-project.org>>>
>>      >     mailing list
>>      >      >>> https://stat.ethz.ch/mailman/listinfo/r-devel
>>     <https://stat.ethz.ch/mailman/listinfo/r-devel>
>>      >     <https://stat.ethz.ch/mailman/listinfo/r-devel
>>     <https://stat.ethz.ch/mailman/listinfo/r-devel>>
>>      >      >>>      <https://stat.ethz.ch/mailman/listinfo/r-devel
>>     <https://stat.ethz.ch/mailman/listinfo/r-devel>
>>      >     <https://stat.ethz.ch/mailman/listinfo/r-devel
>>     <https://stat.ethz.ch/mailman/listinfo/r-devel>>>
>>      >      >>>
>>      >      >>
>>      >      >> ______________________________________________
>>      >      >> R-devel using r-project.org <mailto:R-devel using r-project.org>
>>     <mailto:R-devel using r-project.org <mailto:R-devel using r-project.org>>
>>     mailing list
>>      >      >> https://stat.ethz.ch/mailman/listinfo/r-devel
>>     <https://stat.ethz.ch/mailman/listinfo/r-devel>
>>      >     <https://stat.ethz.ch/mailman/listinfo/r-devel
>>     <https://stat.ethz.ch/mailman/listinfo/r-devel>>
>>      >      >
>>      >
>>      >     ______________________________________________
>>      > R-devel using r-project.org <mailto:R-devel using r-project.org>
>>     <mailto:R-devel using r-project.org <mailto:R-devel using r-project.org>>
>>     mailing list
>>      > https://stat.ethz.ch/mailman/listinfo/r-devel
>>     <https://stat.ethz.ch/mailman/listinfo/r-devel>
>>      >     <https://stat.ethz.ch/mailman/listinfo/r-devel
>>     <https://stat.ethz.ch/mailman/listinfo/r-devel>>
>>      >
>>
> 

-- 
Dr. Benjamin Bolker
Professor, Mathematics & Statistics and Biology, McMaster University
Director, School of Computational Science and Engineering
 > E-mail is sent at my convenience; I don't expect replies outside of 
working hours.



More information about the R-devel mailing list