--- title: "Manipulating Citations with cffr" description: > Learn how to modify `cff` objects. output: rmarkdown::html_vignette: toc: true bibliography: REFERENCES.bib link-citations: yes vignette: > %\VignetteIndexEntry{Manipulating Citations with cffr} %\VignetteEngine{knitr::rmarkdown} %\VignetteEncoding{UTF-8} --- ```{r, include = FALSE} knitr::opts_chunk$set( collapse = TRUE, comment = "#>", warning = FALSE, message = TRUE ) library(cffr) ``` **cffr** is a tool whose target audience are **R** package developers. The main goal of **cffr** is to create a `CITATION.cff` file using the metadata information of the following files: - Your `DESCRIPTION` file. - If available, the citation information located in `inst/CITATION`. ## What is a `CITATION.cff` file? [Citation File Format (CFF](https://citation-file-format.github.io/)) [@druskat_citation_2021] (v1.2.0) are plain text files with human- and machine-readable citation information for software (and datasets). Code developers can include them in their repositories to let others know how to correctly cite their software. This format is becoming popular within the software citation ecosystem. Recently [GitHub](https://docs.github.com/en/repositories/managing-your-repositorys-settings-and-features/customizing-your-repository/about-citation-files), [Zenodo](https://citation-file-format.github.io/#/supported-by-zenodo-) and [Zotero](https://citation-file-format.github.io/#/supported-by-zotero-) have included full support of this citation format [@druskat_stephan_making_2021]. GitHub support is of special interest: ```{r echo=FALSE, out.width="400", fig.align='center', fig.alt="GitHub-link"} knitr::include_graphics("tweet-1.png") ``` *--- Nat Friedman (\@natfriedman) July 27, 2021* See [Customize your repository/About CITATION files](https://docs.github.com/en/repositories/managing-your-repositorys-settings-and-features/customizing-your-repository/about-citation-files) for more info. ## Creating a `CITATION.cff` file for my R package With **cffr** creating a `CITATION.cff` file is quite straightforward. You just need to run `cff_write()`: ```{r setup, eval=FALSE} library(cffr) cff_write() # You are done! ``` Under the hood, `cff_write()` performs the following tasks: - It extracts the metadata using `cff_create()`. - Writes a `CITATION.cff` file using `yaml::write_yaml()`. - Validates the result using `cff_validate()`. Congratulations! Now you have a full `CITATION.cff` file for your **R** package. ## Modifying your `CITATION.cff` file You can easily customize the `cff` object (a custom class of **cffr**) using the parsers provided in the package, as well as making use of the `keys` parameter. We would create a `cff` object using `cff()` (for example purposes only) and we would add or modify contents of it. ### Adding new fields ```{r newfields} newobject <- cff_create(cff()) # For modifying your auto-generated object, run this line instead: # newoobject <- cff_create() newobject ``` The valid keys of the [Citation File Format schema version 1.2.0](https://github.com/citation-file-format/citation-file-format/blob/main/schema-guide.md) can be displayed with `cff_schema_keys()`: ```{r validkeys} cff_schema_keys() ``` In this case, we are going to add `url`, `version` and `repository`. We would also overwrite the `title` key. We just need to create a list and pass it to the `keys` argument of `cff_create()`: ```{r modify} newkeys <- list( "url" = "https://ropensci.org/", "version" = "0.0.1", "repository" = "https://github.com/ropensci/cffr", # If the field is already present, it would be overridden title = "Modifying a 'cff' object" ) modobject <- cff_create(newobject, keys = newkeys) modobject # Validate against the schema cff_validate(modobject) ``` ### Parsing persons and citations **cffr** provides two functions that parse `person` objects and `bibentry` objects (See `?person` and `?bibentry`). These objects are included in the **utils** package and are a core part of the metadata of any **R** package. Following the previous example, we are going to add a new author first. For doing that, we need first to extract the current author of the package and append the parsed person: ```{r includeauthor} # Valid person keys cff_schema_definitions_person() # Create the person chiquito <- person("Gregorio", "Sánchez Fernández", email = "fake@email2.com", comment = c( alias = "Chiquito de la Calzada", city = "Malaga", country = "ES", ORCID = "0000-0000-0000-0001" ) ) chiquito # Parse it chiquito_parsed <- cff_parse_person(chiquito) chiquito_parsed # Append to previous authors # Needs to be append as a list newauthors <- c(modobject$authors, list(chiquito_parsed)) newauthors newauthorobject <- cff_create(modobject, keys = list(authors = newauthors)) newauthorobject cff_validate(newauthorobject) ``` Now, we may want to add `references` to our data. **cffr** supports two types of references: - References created with `bibentry()` - References extracted from packages using `citation()` On the following example, we would add two references, one of each type: ```{r parsingcits} # Valid reference keys cff_schema_definitions_refs() # Auto parsed from another R package base_r <- cff_parse_citation(citation("base")) base_r # Create with bibentry bib <- bibentry("Book", title = "This is a book", author = "Lisa Lee", year = 1980, publisher = "McGraw Hill", volume = 2 ) bib # Now parse it bookparsed <- cff_parse_citation(bib) bookparsed ``` Now the process is similar to the example with `person`: we append both references (as lists) and add them to our object: ```{r references} refkeys <- list(references = c(list(base_r), list(bookparsed))) refkeys finalobject <- cff_create(newauthorobject, keys = refkeys) finalobject cff_validate(finalobject) ``` ### Create your modified `CITATION.cff` file The results can be written with `cff_write()`: ```{r write} # For example tmp <- tempfile(fileext = ".cff") see_res <- cff_write(finalobject, outfile = tmp) see_res ``` And finally we can read our created `CITATION.cff` file using `cff()`: ```{r read} reading <- cff(tmp) reading ``` Note that `cff_write()` also has the `keys` param, so the workflow can be simplified as: ```{r} allkeys <- list( "url" = "https://ropensci.org/", "version" = "0.0.1", "repository" = "https://github.com/user/repo", # If the field is already present, it would be overridden title = "Modifying a 'cff' object", authors = newauthors, references = c(list(base_r), list(bookparsed)) ) tmp2 <- tempfile(fileext = ".cff") res <- cff_write(cff(), outfile = tmp2, keys = allkeys) res ``` ## References