Re: [R-pkg-devel] proper way to cite an invalid URL

2025-04-03 Thread Jason Cory Brunson

Ivan, thank you! That seems to be correct. I'll replace the link.

For future reference in answer to the original question, it seems that 
Duncan's suggestion works best, assuming that the 'curl' problem was 
unrelated.


Cory

On 4/3/25 01:54, Ivan Krylov wrote:

В Wed, 2 Apr 2025 21:20:19 -0400
Jason Cory Brunson  пишет:


in case you're curious, www.ibge.Goiásv.br/

I went looking for links to this domain on the Web, found the link in
the documentation for the 'ripserr' package, and then found what looks
like a domain where the desired page could be found. Could it be

https://www.ibge.gov.br/geociencias/organizacao-do-territorio/estrutura-territorial/15761-areas-dos-municipios.html?edicao=30133&t=acesso-ao-produto

instead of the original link?

https://www.ibge.Goiasv.br/geociencias/organizacao-do-territorio/estrutura-territorial/15761-areas-dos-municipios.html?edicao=30133&t=acesso-ao-produto⁠


But i got a new error from Win-Builder (the development version via
devtools::check_win_devel() but not the release version via
devtools::check_win_release()). It seems to apply to all URLs and
DOIs, though i did not get it on any previous check:

Checking URLs failed with message:
    there is no package called 'curl'

Checking DOIs failed with message:
    there is no package called 'curl'

Looks like Win-Builder for R-devel is having transient problems:
https://www.r-project.org/nosvn/R.check/r-devel-windows-x86_64/curl-00install.html

("ERROR: cannot remove earlier installation, is it in use?", and then
it fails to overwrite libs\x64\curl.dll with "Permission denied".)



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


Re: [R-pkg-devel] [My solution] How and where do I document sysdata.rda

2025-04-03 Thread Michael Dewey

I promised I would update on this.

I now realise that what I was tryig to do was logically impossible. I 
was trying to hide the data set (achieved) but also hide the 
documentation. That makes no sense as anyone wanting to change the 
dataset would not know where to find the documentation. The probabiliy 
anyone would want to edit it is almost indsitinguishable from zero but 
it is as Charles suggests a good principle.


So the answer is to document it in the usual way and use the keyword 
internal. The main difference from a standard Rd file for data is the 
usage statement. As georgi pointed out this just consists of the dataset 
name not data(datasetname).


Thanks for all your help on this and sorry it took so long to respond.

Michael

On 20/03/2025 14:39, Lists wrote:
I want to put a look-up table into sysdata.rda to avoid having to 
compute it everytime. At the moment I create it and save the resulting 
file and then manually copy it to sysdata.rda. This works but the 
package now contains no documentation about the look-up table.


Most of the information which I turned up online involves doing it all 
indirectly using various packages. I do not want to do that, I am quite 
happy editing files by hand and moving them to where I want them by 
hand. So what I need is to know where I should put the documentation and 
what format it should be assuming it is not an Rd file.


Of course what I want may be impossible and I just need to document it 
in the script which creates it but that seems messy.


Michael Dewey

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


--
Michael Dewey

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