[Rd] Bad URLs in DESCRIPTION files

[Kurt Hornik]

>>>>> Colin Gillespie writes:

>> I'm surprised this wasn't captured by the url checker on CRAN.
> As was I

>> Are you suggesting a new check on package descriptions for the urls present?
> Yes - but more fixing edge cases than a new check

>> In that case I guess it could be part of the standard R CMD check.

> Yep. Kurt's message suggests that these were edge cases that missed
> the check.

Actually, the check code finds all these problems, but it is only run
upon submission.  Running 

x <- tools::CRAN_package_db()
y <- parallel::mclapply(x$URL,
                        tools:::.bad_DESCRIPTION_URL_field_parts,
                        mc.cores = 6L)
names(y) <- x$Package
y <- Filter(length, y)
y

currently finds problems in 50 CRAN packages.

Best
-k

> Thanks

>> 
>> Lluís
>> 
>> On Thu, 3 Apr 2025 at 17:17, Colin Gillespie <csgillespie using gmail.com> wrote:
>>> 
>>> Hi All,
>>> 
>>> Just want to flag a few packages:
>>> 
>>> MatrixLDA: https://cran.r-project.org/web/packages/MatrixLDA/index.html
>>> The URL is technically valid - ajmolstad using github.io/ - but returns a
>>> 404. Almost certainly a typo.
>>> 
>>> redistmetrics: https://cran.r-project.org/web/packages/redistmetrics/index.html
>>> Not a valid URL: https::/github.com/alarm-redist/redistmetrics/
>>> 
>>> yarrr: https://cran.r-project.org/web/packages/yarrr/
>>> It looks like a lapsed domain, and Cypro folks have grabbed it.
>>> 
>>> Thanks
>>> 
>>> Colin
>>> 
>>> Ps I only found the last package because the name caught my attention
>>> 
>>> ______________________________________________
>>> R-devel using r-project.org mailing list
>>> https://stat.ethz.ch/mailman/listinfo/r-devel

> ______________________________________________
> R-devel using r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-devel