Ruby CFF

Robert Haines and The Ruby Citation File Format Developers

A Ruby library for creating, editing, validating and converting CITATION.cff files.

Synopsis

This library provides a Ruby interface to create and edit Citation File Format (CFF) files. The resulting files can be validated against a formal schema to ensure correctness and can be output in a number of different citation-friendly formats.

The primary API entry points are the Index and File classes.

See the CITATION.cff documentation for more details about the Citation File Format.

See the full API documentation for more details about Ruby CFF.

Installation

Add this line to your application's Gemfile:

gem 'cff'

And then execute:

$ bundle

Or install it yourself with:

$ gem install cff

Quick start

You can quickly build and save a CFF index like this:

index = CFF::Index.new('Ruby CFF Library') do |cff|
  cff.version = CFF::VERSION
  cff.date_released = Date.today
  cff.authors << CFF::Person.new('Robert', 'Haines')
  cff.license = 'Apache-2.0'
  cff.keywords << 'ruby' << 'credit' << 'citation'
  cff.repository_artifact = 'https://rubygems.org/gems/cff'
  cff.repository_code = 'https://github.com/citation-file-format/ruby-cff'
end

CFF::File.write('CITATION.cff', index)

Which will produce a file that looks something like this:

cff-version: 1.2.0
message: If you use this software in your work, please cite it using the following metadata
title: Ruby CFF Library
authors:
- family-names: Haines
  given-names: Robert
keywords:
- ruby
- credit
- citation
version: 1.0.0
date-released: 2022-10-01
license: Apache-2.0
repository-artifact: https://rubygems.org/gems/cff
repository-code: https://github.com/citation-file-format/ruby-cff

CFF::File can be used to create a file directly, and it exposes the underlying CFF::Index directly. If using a block with CFF::File::open the file will get written on closing it:

CFF::File.open('CITATION.cff') do |cff|
  cff.version = CFF::VERSION
  cff.date_released = Date.today
  cff.authors << CFF::Person.new('Robert', 'Haines')
  cff.license = 'Apache-2.0'
  cff.keywords << 'ruby' << 'credit' << 'citation'
  cff.repository_artifact = 'https://rubygems.org/gems/cff'
  cff.repository_code = 'https://github.com/citation-file-format/ruby-cff'
end

You can read a CFF file quickly with CFF::File::read:

cff = CFF::File.read('CITATION.cff')

And you can read a CFF file from memory with CFF::Index::read or CFF::Index::open - as with CFF::File a block can be passed in to open:

cff_string = ::File.read('CITATION.cff')
cff = CFF::Index.read(cff_string)

CFF::Index.open(cff_string) do |cff|
  # Edit cff here...
end

To quickly reference other software from your own CFF file, you can use CFF::Reference.from_cff. This example uses the CFF file from the core CFF repository as a reference for the Ruby CFF repository:

require 'open-uri'

uri = 'https://raw.githubusercontent.com/citation-file-format/citation-file-format/main/CITATION.cff'
other_cff = URI(uri).open.read

ref = CFF::Reference.from_cff(CFF::Index.read(other_cff))

CFF::File.open('CITATION.cff') do |cff|
  cff.references = [ref]
end

Notes on CFF files with YAML anchors and aliases

Ruby CFF can read files that use YAML anchors and aliases. An anchor (&<label>) identifies a section of your file for reuse elsewhere. An alias (*<label>) is then used to mark where you want that section to be repeated. In this example, the &authors anchor marks an author list for reuse wherever the *authors alias is used:

cff-version: 1.2.0
title: Ruby CFF Library
authors: &authors
- family-names: Haines
  given-names: Robert
  affiliation: The University of Manchester, UK

...

references:
  - type: software
    title: Citation File Format
    authors: *authors

Ruby uses a single object to represent all aliases of an anchor. This means that once the above has been read in by Ruby CFF, if you add an author to either the top-level author list, or the author list in the reference, the new author will appear in both places. With this in mind, you should only use anchors and aliases where the relationship between sections is such that you are sure that exact repetition will always make sense.

When saving CFF files that use anchors and aliases the underlying YAML library will not preserve their names. For example, if the above is loaded into Ruby CFF and then immediately saved &authors/*authors will most likely become &1/*1.

Validating CFF files

To quickly validate a file and raise an error on failure, you can use CFF::File directly:

begin
  CFF::File.validate!('CITATION.cff')
rescue CFF::ValidationError => e
  # Handle validation errors here...
end

Both CFF::File and CFF::Index have instance methods to validate CFF files as well:

cff = CFF::File.read('CITATION.cff')
begin
  cff.validate!(fail_fast: true)
rescue CFF::ValidationError => e
  # Handle validation errors here...
end

Non-bang methods (validate) return an array, with true/false at index 0 to indicate pass/fail, and an array of errors at index 1 (if any).

Passing fail_fast: true (default: false) will cause the validator to abort on the first error it encounters and report just that. Only the instance methods on CFF::File and CFF::Index provide the fail_fast option.

The validation methods (both class and instance) on File also validate the filename of a CFF file; in normal circumstances a CFF file should be named 'CITATION.cff'. You can switch this behaviour off by passing fail_on_filename: false. The non-bang methods (validate) on File return an extra value in the result array: true/false at index 2 to indicate whether the filename passed/failed validation.

Outputting citation text

This library can use CFF data to output text suitable for use when citing software. Currently the output formats supported are:

• BibTeX; and
• an APA-like format.

You can use this feature as follows:

cff = CFF::File.read('CITATION.cff')

cff.to_bibtex
cff.to_apalike

These methods assume that the CFF data is valid - see the notes on validation above.

Assuming the same CFF data as above, the two formats will look something like this:

BibTeX format

@software{Haines_Ruby_CFF_Library_2022,
author = {Haines, Robert},
license = {Apache-2.0},
month = {10},
title = {{Ruby CFF Library}},
url = {https://github.com/citation-file-format/ruby-cff},
version = {1.0.0},
year = {2022}
}

APA-like format

Haines, R. (2022). Ruby CFF Library (Version 1.0.0) [Computer software]. https://github.com/citation-file-format/ruby-cff

Citing a paper rather than software

The CFF has been designed with direct citation of software in mind. We'd like software to be considered a first-class research output, like journal articles and conference papers. If you would rather that your citation text points to a paper that describes your software, rather than the software itself, you can use the preferred-citation field for that paper. When producing citation text this library will honour preferred-citation, if present, by default. If you would like to specify a preferred-citation and still produce a direct citation to the software then you can configure the formatter as follows:

cff = CFF::File.read('CITATION.cff')

cff.to_bibtex(preferred_citation: false)
cff.to_apalike(preferred_citation: false)

A note on citation formats

Due to the different expectations of different publication venues, the citation text may need minor tweaking to be used in specific situations. If you spot a major, or general, error in the output do let us know, but please check against the BibTeX and APA standards first.

Library versions

From version 1.0.0 onwards, the principles of semantic versioning are applied when numbering releases with new features or breaking changes.

Minor or stylistic changes to output formats are not considered "breaking" for the purposes of library versioning.

Developing Ruby CFF

Please see our Code of Conduct and our contributor guidelines.

Licence

Apache 2.0. See LICENCE for details.

Research notice

Please note that this repository is participating in a study into sustainability of open source projects. Data will be gathered about this repository for approximately the next 12 months, starting from June 2021.

Data collected will include number of contributors, number of PRs, time taken to close/merge these PRs, and issues closed.

For more information, please visit our informational page or download our participant information sheet.