This is a basic example of setting up an rOpenSci package review project:
Create the review project, using pkgreview_create
. The
function takes arguments:
pkg_repo
: the GitHub
repo details of the package under review in
the form username/repo
review_parent
:, the local
directory in which the review project (and folder) will
be created and package source code will be cloned
into.library(pkgreviewr)
pkgreview_create(pkg_repo = "ropensci/rdflib",
review_parent = "~/Documents/workflows/rOpenSci/reviews/")
The function creates a new review project in the
review_parent
directory following project naming convention
{pkgname}-review
and populates the review templates to
create all required documents.
The review project directory will contain all the files you’ll need to complete the review and will be initialised with git.
rdflib-review
├── README.md
├── index.Rmd
├── pkgreview.md
└── rdflib-review.Rproj
index.Rmd
The most important file it creates is the
index.Rmd html_notebook
file. This workbook is prepopulated
with all the major steps required to complete the review in an
interactive document to perform and record it in. It also extracts
useful links, information and parameter values.
See example here.
Once rendered to index.nb.html
(*.nb.html
is the notebook file format), this report can be
pushed to GitHub for publication.
pkgreview.md
Template response form to submit to the package rOpenSci onboarding review issue.
See template here.
README.md
Prepopulated README for the review repo that will present the repo to people navigating to it.
See example here:.
To enable local testing of the package, review creation also clones
the review package source code into review_parent
from the
github repository defned in pkg_repo
. This also makes it
available for local review and perhaps even a pull request. Correcting
typos in documentation can be a great review contribution, but first you
might want to check the contributing guidelines or ask the author if
they are open to such pull requests.
The resulting files from a successful review project will look like this:
reviews
├── rdflib
│ ├── DESCRIPTION
│ ├── LICENSE
│ ├── NAMESPACE
│ ├── NEWS.md
│ ├── R
│ │ └── rdf.R
│ ├── README.Rmd
│ ├── README.md
│ ├── appveyor.yml
│ ├── codecov.yml
│ ├── codemeta.json
│ ├── docs
│ │ ├── LICENSE.html
│ │ ├── articles
│ │ │ ├── index.html
│ │ │ ├── rdflib.html
│ │ │ └── rdflib_files
│ │ │ ├── datatables-binding-0.2
│ │ │ │ └── datatables.js
│ │ │ ├── dt-core-1.10.12
│ │ │ │ ├── css
│ │ │ │ │ ├── jquery.dataTables.extra.css
│ │ │ │ │ └── jquery.dataTables.min.css
│ │ │ │ └── js
│ │ │ │ └── jquery.dataTables.min.js
│ │ │ ├── htmlwidgets-0.9
│ │ │ │ └── htmlwidgets.js
│ │ │ └── jquery-1.12.4
│ │ │ ├── LICENSE.txt
│ │ │ └── jquery.min.js
│ │ ├── authors.html
│ │ ├── index.html
│ │ ├── jquery.sticky-kit.min.js
│ │ ├── link.svg
│ │ ├── news
│ │ │ └── index.html
│ │ ├── pkgdown.css
│ │ ├── pkgdown.js
│ │ └── reference
│ │ ├── index.html
│ │ ├── rdf.html
│ │ ├── rdf_add.html
│ │ ├── rdf_parse.html
│ │ ├── rdf_query.html
│ │ ├── rdf_serialize.html
│ │ └── rdflib-package.html
│ ├── inst
│ │ ├── examples
│ │ │ └── rdf_table.R
│ │ └── extdata
│ │ ├── ex.xml
│ │ └── vita.json
│ ├── man
│ │ ├── rdf.Rd
│ │ ├── rdf_add.Rd
│ │ ├── rdf_parse.Rd
│ │ ├── rdf_query.Rd
│ │ ├── rdf_serialize.Rd
│ │ └── rdflib-package.Rd
│ ├── paper.bib
│ ├── paper.md
│ ├── rdflib.Rproj
│ ├── tests
│ │ ├── testthat
│ │ │ └── test-rdf.R
│ │ └── testthat.R
│ └── vignettes
│ └── rdflib.Rmd
└── rdflib-review
├── README.md
├── index.Rmd
└── rdflib-review.Rproj
Use the index.Rmd notebook to work through the review interactively. The document is designed to guide the process in a logical fashion and bring your attention to relevant aspects and information at different stages of the review. You can make notes and record comments within index.Rmd or directly in the review submission form.
Currently the workflow is just set up for you to just copy your
response from your completed pkgreview.md
and paste it into
the review issue but we’re exploring programmatic submission also.
Because the response is currently submitted as .md
, package
reprex
might be useful for inserting reproducible demos of
any issues encountered.
Optional. Have a look at the Publish pkgreview on GitHub vignette.