Skip to content

purajit/YASS

Use this GitHub action with your project
Add this Action to an existing workflow or create a new one
View on Marketplace

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

11 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

YASS

Yet Another Static Site generator

As bare-bones as a static site generator can be, taking less than 50% of the time as something like Jekyll. YASS structures your entire site as a tree, where every page is either a table of contents of its children, or an end page. The defaults are meant to work nicely with GitHub Pages, but could be used in a number of scenarios where you just want to serve a directory tree as a website.

All templating is done via Jinja2.

Overall structure

├── data  # see data_dir below
│   ├── page1
│   ├── page2
│   ├── sublevel1
│   │   ├── page1.1
│   │   └── page1.2
│   ├── sublevel2
│   │   ├── page2.1
│   │   ├── sublevel2.1
│   │   │   └── page2.1.1
│   │   └── page2.2
│   .
│   .
├── docs  # see site_dir below
│   ├── CNAME
│   └── static
├── sitemap.json  # see sitemap_file below
├── templates  # see templates_dir below
│   ├── template_blog.html
│   ├── template_gallery.html
│   .
│   .

sitemap structure

The structure of the sitemap json essentially reflects the structure of the data_dir, and tells YASS how to interpret the data files and how to arrange the children of a page. Each element in the json either gives information about how to get rendered (if it's an end page), or has a list of its children (if not). The structure of the sitemap is identical at every level - the config options available at each level are:

  • children:
    Optional (the presence of this defines whether it will an end page or a contents page).
    Specify the children of the page (if any) as a list of sitemaps.
  • data_type:
    Optional. Only for end pages. One of json, markdown, text. Defaults to text.
    Makes the data in the corresponding file available as the variable data within your templates.
    • text: data will be the entire text blob as-is
    • json: data will be a dict of those values
    • markdown: data will be the marko-produced conversion of the given Markdown to HTML
  • route:
    Required at all levels.
    Defines the web route for the level, as well as the name of the data file/directory within data_dir that will be looked up if this is an end page. For the top level, this should be the empty string "".
  • template:
    Required at all levels.
    Defines the Jinja template to be used at each of these levels, which will be looked up in templates_dir. This allows you to define your own contents page template, and in fact allows you to define several contents page templates where you could use a different one in each place. If it is a contents page, a parameter called contents will be made available in your template, containing a list of maps of {"title": <title>, "link": <link>} of the child pages that can then be displayed however you wish.
  • title:
    Required at all levels.
    Defines how the parent of this page will refer to this page, and also is made available in your template directly as title that can be used however you wish.
    • tab_title:
      Optional. Defaults to title.
      An additional, admittedly odd parameter - makes the tab_title variable available in your templates, in cases where you might want the actual page title to be different that the title. Only matters if you actually use it in your templates, and the generator does not control any behavior with it.

Example

{
  "title": "YASS - Yet Another Static Site Generator",
  "tab_title": "YASS",
  "template": "template_index.html",
  "route": "",
  "children": [
    {
      "title": "Documentation",
      "route": "documentation",
      "template": "template_contents.html",
      "children": [
        {
          "title": "Installation",
          "route": "installation",
          "template": "template_documentation.html",
          "data_type": "text"
        },
        {
          "title": "Usage",
          "route": "usage",
          "template": "template_documentation.html",
          "data_type": "text"
        },
      ]
    },
    {
      "title": "Testimonials",
      "route": "testimonials",
      "template": "template_testimonials.html",
      "data_type": "json"
    }
  ]
}

Variables available in your templates

The above section went over some of this, but to summarize:

  • contents: present if this is not an end-page, a list of maps of {"title": <title>, "link": <link>} of the child pages
  • data: the actual data of the page, either a json or text blob based on the sitemap definition
  • title: this is what the parent of a page will be given as part of its children list
  • tab_title: additional parameter that defaults to title that you can use as the actual HTML title if you wish
  • parent_page: a link to the parent page
  • static_url: defaults to the empty string, but can be used to switch out the source of your statics
  • All parameters in global_template_parameters (see below)

Configuration

All configurations are optional, and have default values:

  • data_dir:
    The location of the raw data of your site, structured in a tree. Each data file should either be raw text or json; if it is json, it will be used as parameters to the template.
    Example: ./data
  • global_template_parameters:
    Dictionary of parameters that should be provided to every single page. Could be used for generating sites for different environments, or to create your own opinionated template set that you then use to generate multiple similar sites.
  • permanent_paths:
    List of files with the generated site path that should never be cleared out, since the hosting service you use might require certain files in the directory that's being hosted; you may also have static files that shouldn't be impacted.
    Example: ["CNAME", "static"]
  • site_dir:
    The directory that your site will be served out of, and where all the generated files will be populated.
    Example: ./docs
  • sitemap_file:
    Path to your sitemap file, as defined above.
    Example: ./sitemap.json
  • static_url:
    URL for all your static assets. If your templates use the form {{ static_url }}/path/to/static, then the generator will make sure to populate it correctly. This way, you can generate sites for multiple environments that use statics in different locations - even localhost.
    Example: https://assets.example.com, or /static/assets for local use
  • templates_dir:
    Directory with all the templates mentioned in your sitemap.
    Example: ./templates
  • template_functions_file:
    Path to a file containing any custom functions you'd like to use in your templates. Each function you want available must begin with yass_; this prefix will be stripped out while injecting it. So, yass_foo will be available in your templates as foo.
    Example: yass_template_functions.py

GitHub Action usage

YASS comes with a GitHub action that can generate the statics for you (either for testing or deployment)

See action.yml

Example

jobs:
  deploy_site:
    steps:
      - name: Checkout
        uses: actions/checkout@v3

      - name: Generate statics
        uses: purajit/[email protected]
        with:
          yass-config-path: yass_config_prod.json

      # Futher deploy steps

Notes

Automatic sitemap generation from data_dir

I considered a structure where the files in data_dir automatically tell the generator how to deal with them and how to arrange them - essentially as a way to completed remove the sitemap json, since that does essentially copy the structure over.

However, I prefer the sitemap being explicit and all in one place; this also allows you to have "staging" data in your data_dir that doesn't get hosted until you want it to be. It also makes the code and readability much, much simpler.

Future work

  • Allow for templating of static files

Used by

Site | GitHub

purajit.com

Site | GitHub

mythmancer.com