This is the source code of the icatproject.github.io GitHub Pages site. It uses GatsbyJS to generate the site using React for the structure of the site and Markdown files for the site contents.
The src folder contains all the React code used to create the site. Look here if you need to edit the layout of the site or add new functionality. See developer notes
The content folder contains all the Markdown files and images used to fill in the site contents. If you need to edit site content or add a new page then look here. See content notes
The static folder contains other resources such as PDFs. These are copied automatically to the public folder, retaining any file structure.
To develop locally, install GatsbyJS and clone this branch (and change into its directory) and then run npm install to install all the needed dependencies and then run gatsby develop. This will build the site at http://localhost:8000
To deploy manually, run npm run deploy. This will automatically build and publish the build to the master branch. However, this repository has TravisCI set up so that when changes are detected on the gatsbyjs branch the site is deployed after it is built, so manual deployment would rarely be needed.
The src directory contains all of the source code for the application. It contains three subfolders.
src/components contains any React components that in and of themselves are not standalone pages - they are instead used to construct the layout. There is no index.js file or App.js file, instead the "base" file is src/components/layout.jsx. This determines the overall structure of the site as all pages are built using <Layout>. It also pulls in all the other components in src/components to create the static areas of the site such as the header. So starting from layout.jsx should allow one to discover how the site is constructed. The GatsbyJS tutorial is a good resource to learn how to to develop, but most of the components are plain React. The only Gatsby-specific information to know is to use GatsbyJS <Link> components when navigating to a page internal to the site as this uses react-router underneath and allows for faster page navigation, and the usage of StaticQuery throughout the components.
src/pages contains standalone pages that are not generated from Markdown. For example, 404.jsx is a special page that Gatsby looks for to navigate to when the user specifies a route that does not exist. Meanwhile search.jsx is a regular page at /search.
src/templates contains pages that take Markdown and generate pages in the website. Currently, every Markdown page is processed exactly the same through the default.jsx template, that merely extracts the title from the Markdown frontmatter and uses it to create a <h1> element and otherwise dumps the rest of the Markdown html into the DOM.
There are also some configuation files that may be of interest. gatsby-config.js contains the gatsby configuration - this is mostly used to specify plugins and modify the plugin settings. gatsby-node.js is where we process all the so called nodes - these are used to create pages. This is where we convert our Markdown content into Gatsby MarkdownRemark nodes (using onCreateNode), giving them a slug or URL path. Then createPages is used to convert MarkdownRemark nodes into html pages using our default.jsx template.
This project has Prettier and ESLint set up. See .prettierrc and .eslintrc.json for details. Use npm run format to format files according to Prettier and use npm run lint to lint files using ESLint. There are tutorials online for setting these up with any text editors you use for a better workflow. They will also run automatically on staged files before commit - any ESLint errors will cancel the commit.
- All top level folders of the
contentdirectory are converted into navbar headers and their children into dropdown items. Seenavbar.jsanddropdown.jsfor more info on this - When creating a new Markdown file, ensure to give it a frontmatter section with a title - and if it is a set of meeting minutes a date. The format is as shown below:
or
--- title: My page title ------ title: My Meeting Minutes date: 2019-01-30 --- - To automatically give an image a caption, use the below format:
[image title](image_url "Your caption here")