Here is an R package I put together to help with making my project work more accessible and standardized. The main function, create_site(), will allow the user to create an Rmarkdown file to use as a notebook that will be rendered into an html file as well. This will be placed in the docs directory of your project. This in turn makes it able to used by Github Pages for the repository. To enable this, under your repository go to the Settings tab, scroll down to the Github Pages section and under the source option change the drop down menu from 'None' to 'master branch/docs folder'. Just like that you have a basis for a website to share your project analysis!
There are several directories created including docs, src, results and references. I have found these are broad enough to encompasses most of what is necessary for a project. You could easily argue for others or consolidation of these directories. An Rmarkdown file is created for a Notebook and placed in the src directory. It will also created a sub directory of src/chunks. Each Notebook chunk will be separated out and converted into its own .R file and placed in the chunks directory while maintaining the full Notebook.Rmd file as well. Each .R file will be named after the chunk it came from, making it important that you name your chunks in your notebook to find them later on. This helped keep me honest in documenting my code knowing this will be important later on. Also each chunk has #!/usr/bin/env Rscript in order to be run as a standalone Rscript file. Finally an index.html file and a Notebook.html file are created as well which can be edited later.
Every time updates are made to the notebook, there is another function to update_site() that will take the latest Notebook.Rmd file and save each chunk again to its own .R file and knit the Notebook.html file again to reflect the latest updates. There is also an index.Rmd file that was made and placed in the src directory that can be updated. This would be the main part of your new website and can be edited as desired.
This is the first iteration of this package and I wanted to get it out there. I would love to hear feedback and suggestions for the package. I had pieces of this in R scripts that I would use to do some of these tasks. i decided it would be easier just to create a package and make some easy to use functions. One draw back may be it assumes too much and people might not like that. It is something that I have worked out for my workflow. I might try to add more flexibility to it. I could see adding options and modules to pick and choose things for your specific project. I didn't want it to be too complicated but at the same time want to be broad enough to be useful. I would gladly take pull requests or suggestions. Hope this helps some people!
