The goal of grateful is to make it very easy to cite R and the R packages used in any analyses, so that package authors receive their deserved credit. By calling a single function, grateful will scan the project for R packages used and generate a BibTeX file containing all citations for those packages.
grateful can then generate a new document with citations in the desired output format (Word, PDF, LaTeX, HTML, Markdown). These references can be formatted for a specific journal, so that we can just paste them directly into our manuscript or report.
Alternatively, we can use grateful directly within an Rmarkdown or Quarto document. In this case, a paragraph containing in-text citations of all used R packages will (optionally) be inserted into the Rmarkdown/Quarto document, and these packages will be included in the reference list when rendering.
You can install {grateful} from CRAN:
install.packages("grateful")Or from GitHub:
# install.packages("remotes")
remotes::install_github("Pakillo/grateful")grateful can be used in one of two ways:
to generate a ‘citation report’ listing each package and their citations
to build citation keys to incorporate into an existing R Markdown or Quarto document.
Imagine a project where we are using the packages: dplyr, ggplot2, vegan and lme4. We want to collect all the citations listed for these packages, as well as a citation for base R (and for RStudio, if applicable).
Calling cite_packages() will scan the project, find
these packages, and generate a document with formatted citations.
library(grateful)cite_packages(out.dir = ".") # save report to working directory
This document can also be a Word or LaTeX document, PDF file,
markdown file, or left as the source Rmarkdown file using
out.format:
cite_packages(out.format = "docx", out.dir = ".")We can specify the citation style for a particular journal using
citation.style.
cite_packages(citation.style = "peerj", out.dir = ".")In all cases a BibTeX (.bib) file with all package citations will be saved to disk.
If you are building a document in RMarkdown or Quarto and want to cite R packages, grateful can automatically generate a BibTeX file and ensure these packages are cited in the appropriate format (see template Rmarkdown and Quarto documents).
First, include a reference to the BibTeX file in your YAML header.
bibliography: grateful-refs.bib(Note: You can reference multiple BibTeX files, if needed)
bibliography:
- document_citations.bib
- grateful-refs.bibThen call cite_packages(output = "paragraph") within a
code chunk (block or inline) to automatically include a paragraph
mentioning all the used packages, and include their references in the
bibliography list.
```{r}
cite_packages(output = "paragraph", out.dir = ".")
```We used R version 4.2.3 [@base] and the following R packages: lme4 v. 1.1.32 [@lme4], tidyverse v. 2.0.0 [@tidyverse], vegan v. 2.6.4 [@vegan].
Alternatively, you can get a table with package name, version, and
citations, using output = 'table':
```{r }
pkgs <- cite_packages(output = "table", out.dir = ".")
knitr::kable(pkgs)
```
If you want the references to appear in a particular format, you can specify the citation style in the YAML header:
bibliography: grateful-refs.bib
csl: peerj.cslAlternatively, you can cite particular packages using the citation
keys generated by grateful, as with any other BibTeX
reference, or just include citations in the References section, using
the function nocite_references(). See the package help and
the RMarkdown
cookbook for more details.
Use scan_packages
scan_packages()
pkg version
1 badger 0.2.4
2 base 4.4.1
3 knitr 1.48
4 pkgdown 2.1.0
5 remotes 2.5.0
6 renv 1.0.7
7 rmarkdown 2.28
8 testthat 3.2.1.1
9 tidyverse 2.0.0
10 visreg 2.7.0If you just want to get all package references in a BibTeX file, you
can call get_pkgs_info(). Besides printing a table with
package info, it will also save a BibTeX file with references. By
default, the file will be called grateful-refs.bib, but you
can change that (see function help).
If you want to get the BibTeX references for a few specific packages:
get_pkgs_info(pkgs = c("remotes", "renv"), out.dir = getwd())
#> pkg version citekeys
#> 1 remotes 2.5.0 remotes
#> 2 renv 1.0.7 renvIf you use one or several packages from the tidyverse, you can choose to cite the ‘tidyverse’ rather than the individual packages:
cite_packages(cite.tidyverse = TRUE)Most R packages also depend on other packages. To include those
package dependencies in your citations, rather than just the packages
you called directly, use dependencies = TRUE:
cite_packages(dependencies = TRUE)Some R packages wrap core external software that should perhaps be
cited too. For example, rjags
is an R wrapper to the JAGS software written in
C++. Ideally, R packages wrapping core external software will include
them in their CITATION file. But otherwise, we can investigate external
software requirements of our used packages, e.g. using
remotes:
remotes::system_requirements(package = c("rjags"), os = "ubuntu-20.04")
#> [1] "apt-get install -y jags"Citing software is pretty much like citing papers. Authors have to decide what to cite in each case, which depends on research context.
As written in the Software Citation Principles paper (Smith et al. 2016):
The software citation principles do not define what software should be cited, but rather how software should be cited. What software should be cited is the decision of the author(s) of the research work in the context of community norms and practices, and in most research communities, these are currently in flux. In general, we believe that software should be cited on the same basis as any other research product such as a paper or book; that is, authors should cite the appropriate set of software products just as they cite the appropriate set of papers, perhaps following the FORCE11 Data Citation Working Group principles, which state, “In scholarly literature, whenever and wherever a claim relies upon data, the corresponding data should be cited”
And these are the guidelines from the Software Citation Checklist:
You should cite software that has a significant impact on the research outcome presented in your work, or on the way the research has been conducted. If the research you are presenting is not repeatable without a piece of software, then you should cite the software. Note that the license or copyright of the software has no bearing on whether you should cite it.
This might include:
Software (including scripts) you have written yourself to conduct the research presented. A software framework / platform upon which the software you wrote to conduct the research relies. Software packages, plugins, modules and libraries you used to conduct your research and that perform a critical role in your results. Software you have used to simulate or model phenomena/systems. Specialist software (which is not considered commonplace in your field) used to prepare, manage, analyse or visualise data. Software being evaluated or compared as part of the research presented Software that has produced analytic results or other output, especially if used through an interface.
In general, you do not need to cite:
Software packages or libraries that are not fundamental to your work and that are a normal part of the computational and scientific environment used. These dependencies do not need to be cited outright but should be documented as part of the computational workflow for complete reproducibility. Software that was used during the course of the research but had no impact on research results, e.g. word processing software, backup software.
Apart from citing the software most relevant to the particular
research/analysis performed, I think it is good idea to record the
entire computational environment elsewhere, e.g. using
sessionInfo() or
sessioninfo::session_info().
Before running grateful you might want to run funchir::stale_package_check
or annotater
to check for unused packages before citing them.
If getting an error like “Error in (function (pkg, lib.loc = NULL):
there is no package called…”, that means that some of your scripts is
loading a package that is no longer available in your computer, so
{grateful} cannot grab its citation. To fix this, there are several
options. First, you could omit that package (or those packages, if more
than one) from {grateful} citations using
cite_packages(omit = c("package1", "package2").
Alternatively, try checking if that package is still needed for your
project and you want to cite it, otherwise remove or comment that line
where the package is loaded. If you still use and want to cite that
package, install it, and then run cite_packages again.
When a project includes many used packages (or files),
renv may issue a warning. Use
options(renv.config.dependencies.limit = 10000) to overcome
the warning and scan the project for all packages used. Alternatively,
use .renvignore to ignore certain files or folders (see
renv help).
citation("grateful")
To cite package 'grateful' in publications use:
Rodriguez-Sanchez F, Jackson C (2023). _grateful: Facilitate citation
of R packages_. <https://pakillo.github.io/grateful/>.
A BibTeX entry for LaTeX users is
@Manual{,
title = {grateful: Facilitate citation of R packages},
author = {Francisco Rodriguez-Sanchez and Connor P. Jackson},
year = {2023},
url = {https://pakillo.github.io/grateful/},
}Citation keys are not guaranteed to be preserved when regenerated, particularly when packages are updated. This instability is not an issue when citations are used programmatically, as in the example above. But if references are put into the text manually, they may need to be updated periodically.