Skip to content

Latest commit

 

History

History
85 lines (59 loc) · 3.86 KB

CONTRIBUTING.md

File metadata and controls

85 lines (59 loc) · 3.86 KB
Raw

Contributing

Getting started

  • Clone Hyde
  • In the folder, run npm install to install all dependencies.

To run the build tool, the entry file is the executable bin/generate.js (that is named hyde after publishing). For example, in the test folder, run: ../bin/generate.js --watch

Publishing

If you have the rights to publish, run:

make publish

Hyde to replace Jekyll

The folder test/jekyll contains an experimental project that builds a Jekyll website using Hyde

  • Navigate to test/jekyll (this folder)

  • Run

    node ../../bin/generate.js --serve=_site

This is equivalent to hyde --serve=_site if Hyde was installed. This command opens Editor serving _site, and listen to changes made on index.html and the folder _layouts. Try to do the following to see that it works:

  • In index.html, replace in your favorite editor Hello with Hi. Reload the page on the browser to see the changes
  • In the browser, replace the ! by a ?. Since auto-save is on, Editor back-propagates the change automatically to _site/index.html, and then Hyde back-propagates the change automatically to the real source index.html, in a few seconds.

Contributing

Current state

The Hydefile that defines this Jekyll-like transformation is still a draft. I followed Jekyll's tutorial and implemented

  • Step 1: Basic website
  • Step 2: Object, filters and Liquid. I did not implement the tags 'if/then/else/for/while/do'.
  • Step 3: front matter
  • Step 4: Layouts, markdown, about page

There remains to implement

  • Step 5: Includes, current page highlighting
  • Step 6: Data, Data file usage
  • Step 7: Assets, Sass
  • Step 8: Blogging, posts, List Posts, More posts
  • Step 9: Collections, configurations, authors, staff page, output a page, front matter defaults, list author's posts, link to author's page
  • Step 10: Deployment

Learn Elm

The language Leo used in the hydefile is very similar to Elm. Elm has a standard library that has been reimplemented. The most useful data structures and methods for this tool are List, String and Regex. .

Structure of the current hydefile.elm for Jekyll's website

The entry point of this file is the variable all at the end of the file test/jekyll/hydefile.elm. It lists the file that are html and markdown and interpret them.

  • all returns a list of commands of type List (Write name content | Error String)
  • interpret
    • checks if a file is html or markdown (isMd), computes its final path (newName),
    • reads the file's content (which is a Maybe String)
    • Extract the front matter as an object (frontmattercode)
    • Removes the front matter (sourceWithoutFrontMatter)
    • If there is a layout, use this layout as the basic source
  • applyObjects takes an environment and a string and replaces the interpolated parts {{...}} by interpreting what's inside as code.
  • unjekyllify takes a piece of code in Jekyll's syntax and transforms it to a piece of code in Leo's syntax

Warm-up exercise

  • Implement the include procedure
    • Add a regular expression to extract {% include %} in the function applyObjects (duplicate the lines 68-74)
    • Use fs.read to read a filename. It returns a Maybe String
    • Append |> Maybe.map (\content -> ...) |> Maybe.withDefault "default" to the right of something of type Maybe String to further process the string and return a default value if it was Nothing