[R-pkg-devel] renaming a package

[Jonathan Callahan]

We are developing a suite of three packages related to seismology. As I was
dealing with 64 bit compilation issues that became apparent from a previous
submission attempt, another package with the name 'seismic' was accepted to
CRAN.

This brings up two issues:

1) CRAN might wish to have better package naming guidelines than "short and
sweet, first come, first served".

As the number of packages continues to rise, it might be useful to have
java style *OrganizationPackage* naming. In our case, we are now going to
release the core seismic data package as 'IRISSeismic'.

2) What is the best practice for renaming a package that has already been
accepted to CRAN?

We had previously released a 'seismicRoll' package that we may now wish to
rename 'IRISSeismicRoll'.

Thanks,

Jon

-- 
Jonathan Callahan, PhD
Mazama Science
206-708-5028
mazamascience.com

[[alternative HTML version deleted]]

__
R-package-devel@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-package-devel

Re: [R-pkg-devel] renaming a package

[Duncan Murdoch]

On 19/07/2015 6:19 PM, Jonathan Callahan wrote:
> We are developing a suite of three packages related to seismology. As I was
> dealing with 64 bit compilation issues that became apparent from a previous
> submission attempt, another package with the name 'seismic' was accepted to
> CRAN.
> 
> This brings up two issues:
> 
> 1) CRAN might wish to have better package naming guidelines than "short and
> sweet, first come, first served".
> 
> As the number of packages continues to rise, it might be useful to have
> java style *OrganizationPackage* naming. In our case, we are now going to
> release the core seismic data package as 'IRISSeismic'.

I don't see a need for this.  If you want to do it for your packages, go
ahead.

> 
> 2) What is the best practice for renaming a package that has already been
> accepted to CRAN?
> 
> We had previously released a 'seismicRoll' package that we may now wish to
> rename 'IRISSeismicRoll'.

Rename it and submit it.  In your message to CRAN, mention that you are
renaming it from seismicRoll, and say why the name seismicRoll is no
longer appropriate.

Don't do this too often, but there's no problem doing it occasionally,
when there's a good reason to do so.

Duncan Murdoch

__
R-package-devel@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-package-devel

Re: [R-pkg-devel] renaming a package

[Uwe Ligges]

On 20.07.2015 00:53, Duncan Murdoch wrote:

On 19/07/2015 6:19 PM, Jonathan Callahan wrote:

We are developing a suite of three packages related to seismology. As I was
dealing with 64 bit compilation issues that became apparent from a previous
submission attempt, another package with the name 'seismic' was accepted to
CRAN.

This brings up two issues:

1) CRAN might wish to have better package naming guidelines than "short and
sweet, first come, first served".

As the number of packages continues to rise, it might be useful to have
java style *OrganizationPackage* naming. In our case, we are now going to
release the core seismic data package as 'IRISSeismic'.


I don't see a need for this.  If you want to do it for your packages, go
ahead.



2) What is the best practice for renaming a package that has already been
accepted to CRAN?

We had previously released a 'seismicRoll' package that we may now wish to
rename 'IRISSeismicRoll'.


Rename it and submit it.  In your message to CRAN, mention that you are
renaming it from seismicRoll, and say why the name seismicRoll is no
longer appropriate.

Don't do this too often, but there's no problem doing it occasionally,
when there's a good reason to do so.



Thanks, and yes, CRAN will allow renaming in case there is a really good 
reason.


Note that package names, once a package got renamed, cannot be used by 
others any more, hence we are allowing such a change only in rare cases 
with very good reasons...

In the past not many maintainers succeeded to get us convinced...

Best,
Uwe Ligges





Duncan Murdoch

__
R-package-devel@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-package-devel



__
R-package-devel@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-package-devel

[R-pkg-devel] Note about "No repository set, so cyclic dependency check skipped"

[Antonio José Saez Castillo]

When checking a package I am getting

* checking package dependencies ... NOTE
   No repository set, so cyclic dependency check skipped

This is an old source of discussion and I've found a lot of suggestions,
but none of them work for me (.Rprofile and/or Rprofile.site edition and
so). What is more worrying is that I'm also getting this note when I check
in R-devel on win-builder site (win-builder.r-project.org). There is
another similar topic in relation to that (
http://r.789695.n4.nabble.com/No-repository-set-so-cyclic-dependency-check-skipped-td4684184.html),
but I don't know if it is the same problem.

How can I avoid this note?







*Dr. Antonio José Sáez-Castillo*Department of Statistics and
Operational Research
Escuela Politécnica Superior de Linares
UNIVERSIDAD DE JAÉN
C/ Alfonso X el Sabio, 28
23700 Linares (Jaén)
Teléfono: +34 953 64 85 78
e-mail: ajs...@ujaen.es
 


[[alternative HTML version deleted]]

__
R-package-devel@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-package-devel