bib field "urldate" converted into (invalid) links

[Nmoiroux]

Hello,
When the bibentry has an "urldate" field, this latter converted into an invalid link and displayed into the bibliography.
This leads to NOTES when testing an R package with win-builder such as:
Found the following (possibly) invalid file URIs: URI: 2025-01-29 From: man/function.Rd URI: 2026-02-03 From: man/function.Rd

[GeoBosh]

Thanks for the report. Could you give a bibentry that causes this?

[GeoBosh]

Without an example bibentry producing the warning, I can't even confirm that it is real. I don't doubt that you have observed it, but why not show the bibentry that raises that? It is not much to ask and if you look around github you will see that many repositories require it (and even more) for bug reports.

The field urldate is not converted to a link. For bibentry type online it is used to add a date when the link was accessed. For example:

rbibutils::charToBib('@online{n15,
  author = "Navarro, D.",
  title = "Learning statistics with {R}: A tutorial for psychology students and other beginners. (Version 0.5)",
  year = "2015",
  url = "https://learningstatisticswithr.com/",
  urldate = "7-3-2021"
}', direct = TRUE)

## Navarro D (2015). “Learning statistics with R: A tutorial for psychology students and other
## beginners. (Version 0.5).” <https://learningstatisticswithr.com/> (visited on 7-3-2021).

The date is treated as just pure text. Currently, the date is output 'as is' without parsing or validation. In the this example the date is in an ambiguous format.

[GeoBosh]

I observe the reported issue for 'online' bibentries that lack an 'url' field (or if it is present but empty.

With URL field we get:

rbibutils::charToBib('@online{n15,
  author = "Navarro, D.",
  title = "Learning statistics with {R}: A tutorial for psychology students and other beginners. (Version 0.5)",
  year = "2015",
  url = "https://learningstatisticswithr.com/",
  urldate = "7-3-2021"
}', direct = TRUE)

## Navarro D (2015). “Learning statistics with R: A tutorial for psychology students and other beginners. (Version 0.5).” 
## <https://learningstatisticswithr.com/> (visited on 7-3-2021).

Without URL:

rbibutils::charToBib('@online{n15,
  author = "Navarro, D.",
  title = "Learning statistics with {R}: A tutorial for psychology students and other beginners. (Version 0.5)",
  year = "2015",
  urldate = "7-3-2021"
}', direct = TRUE)

## Navarro D (2015). “Learning statistics with R: A tutorial for psychology students and other beginners. (Version 0.5).”
## <7-3-2021> (visited on 7-3-2021).

So, the URLDATE field is taken to be the missing URL (a technical explanation is at the end of this comment).

An 'online' entry must have an URL, so I have now fixed its processing to issue a warning if this is not the case or if the URL field is missing or empty:

## without url field
rbibutils::charToBib('@online{n15,
  author = "Navarro, D.",
  title = "Learning statistics with {R}: A tutorial for psychology students and other beginners. (Version 0.5)",
  year = "2015",
  urldate = "7-3-2021"
}', direct = TRUE)

##  Navarro D (2015). “Learning statistics with R: A tutorial for psychology students and other beginners. (Version 0.5).” 
## (visited on 7-3-2021).
Warning message:
In paste(strings, collapse = "\n") :
  missing 'url' field in bibentry of type 'online'


## with empty url field
rbibutils::charToBib('@online{n15,
  author = "Navarro, D.",
  title = "Learning statistics with {R}: A tutorial for psychology students and other beginners. (Version 0.5)",
  year = "2015",
  url = ,
  urldate = "7-3-2021"
}', direct = TRUE)

## Navarro D (2015). “Learning statistics with R: A tutorial for psychology students and other beginners. (Version 0.5).” 
## (visited on 7-3-2021).
Warning message:
In paste(strings, collapse = "\n") :
  missing 'url' field in bibentry of type 'online'

This fix is in rbibutils (not Rdpack) and I will update http://github.com/GeoBosh/rbibutils repository shortly.

The rest is some technical information for me.

The issue stems from a call paper$url in extraInfo() (a function in base R). If the field 'url' is missing R does partial matching and picks 'urldate'. Note that the '$' operator does not do partial matching for 'bibentryExra' objects created by rbibutils. But when rendered for printing, they are unclassed and become ordinary lists (for which partial matching is done).

[Nmoiroux]

Dear @GeoBosh,
First I apologyze for not having been able to answer shortly.
My bibentries were not of type @online.
I'm not sure how to produce a simple reproducible example. Nevertheless, the following release of my package ecoXCorr generates links for urldates fields in the documentation:

library(devtools)
install_github("Nmoiroux/ecoXCorr@v0.1.8")

library(ecoXCorr)
?ecoXCorr

below is bib entry used in the package documentation (from REFERENCES.bib):

@article{benjaminiControlFalseDiscovery2001,
  title = {The Control of the False Discovery Rate in Multiple Testing under Dependency},
  author = {Benjamini, Yoav and Yekutieli, Daniel},
  year = 2001,
  month = aug,
  journal = {The Annals of Statistics},
  volume = {29},
  number = {4},
  pages = {1165--1188},
  publisher = {Institute of Mathematical Statistics},
  issn = {0090-5364, 2168-8966},
  doi = {10.1214/aos/1013699998},
  urldate = {2025-01-29},
  abstract = {Benjamini and Hochberg suggest that the false discovery rate may be the appropriate error rate to control in many applied multiple testing problems. A simple procedure was given there as an FDR controlling procedure for independent test statistics and was shown to be much more powerful than comparable procedures which control the traditional familywise error rate. We prove that this same procedure also controls the false discovery rate when the test statistics have positive regression dependency on each of the test statistics corresponding to the true null hypotheses. This condition for positive dependency is general enough to cover many problems of practical interest, including the comparisons of many treatments with a single control, multivariate normal test statistics with positive correlation matrix and multivariate \$t\$. Furthermore, the test statistics may be discrete, and the tested hypotheses composite without posing special difficulties. For all other forms of dependency, a simple conservative modification of the procedure controls the false discovery rate. Thus the range of problems for which a procedure with proven FDR control can be offered is greatly increased.},
  keywords = {47N30,62G30,62J15,comparisons with control,discrete test statistics,FDR,Hochberg's procedure,MTP2 densities,Multiple comparisons procedures,multiple endpoints many-to-one comparisons,positive regression dependency,Simes'equality,unidimensional latent variables},
  file = {/home/nm/Zotero/storage/EG5F8PXP/Benjamini and Yekutieli - 2001 - The control of the false discovery rate in multipl.pdf}
}

Hope this will help and thank you for this useful package.
Dont know if I should reopen the issue...
N

[GeoBosh]

Thanks for the example, no worries about being late - it clarifies the issue (though I had already figured it out). Note that naming a GitHub repo of the package raising the issue is equivalent to a reproducible example here, as is the offending bibentry.

The issue is as I described it in my previous comment but I fixed it for 'online' type only (on GitHub for now). I am travelling now but will fix the issue more generally to avoid taking urldate for url when the latter is missing.

It looks like you generated the bibentry using zotero. Is urldate the date when zotero did this? Initially I thought it might be the date when the paper was put online but that doesn't look the case.

[Nmoiroux]

Thank you,
Yes, I confirm that the bibentries were generated using zotero. In Zotero, this field is displayed as "Accessed" and is filled with the current date when saving a reference using the "Zotero connector" plugin (for Firefox).