blog-runner
v0.2.0
Published
A simple JavaScript static site generator built for bloggers
Downloads
22
Readme
A lightweight static site generator for bloggers built for the JavaScript environment.
Blog Runner is a simple static site generator that includes basic templating and Markdown support, built as a bare-bones replacement for Jekyll. Blog Runner is built on top of the mustache.js
templating library and the marked
Markdown parser. The latest build can be seen in action on the author's blog: alexpear.com.
Installation
Blog Runner is packaged as an npm
module. The package should be installed as a devDependency
in your project's package.json
file.
devDependency:
npm install --save-dev blog-runner
Usage
Directory Structure
blog-runner
requires a specific directory structure to run correctly. You may execute build
on any folder, as long as it is organized using a modified version of the Jekyll directory structure. The only modifications to this structure should be:
- No
_config.yml
file (configuration is intentionally limited in early versions) - No
index.html
file (one will be generated from thelanding.html
layout) - No
_site
directory (one will be generated for you) - No
_data
directory (YAML data is unsupported at this time) - In
_layouts
,default.html
should be calledlanding.html
- An additional directory called
_mixins
for your custom, JavaScript-built_includes
files
All files in _layouts
and _includes
should be *.html
files, all files in _mixins
should be *.js
files, and all files in _drafts
and _posts
should be *.md
(Markdown) files. No YAML font-matter is required, as all supported data is currently parsed from each post's filename.
All posts should be named according to the following structure:
YYYY-MM-DD-title-with-hyphens.md
For example:
2016-01-01-happy-new-year-everybody.md
When your _site
directory is built, the example post will have a file path of:
/2016/01/01/happy-new-year-everybody/index.html
Since there is no YAML front matter or config file, it is EXTREMELY IMPORTANT that you organize your blog directory and name your files according to these conventions! Otherwise, your site won't build correctly.
Execution
blog-runner
is meant to be used as a part of a build process. The blog structure is built into a _site
directory, which you will copy over to your live web server. The module currently exports four functions for your use: index
, roll
, mixin
, and build
.
require
You can access the main methods through require
:
const blog = require('blog-runner');
index(source)
The index()
method returns an array of objects containing all post data, including title, and the year, month, and day of publishing. All post objects are arranged in the array in chronological order by publishing date. This function is only a helper function, and does not make any changes to _site
.
You are required to pass in a source
parameter that points to the directory that you would like indexed.
blog.index('path/to/source/directory');
roll(source, [{config options}])
For those looking to include a standard index view of posts on one of their pages (probably a landing page), roll()
will build a new blogroll.html
include in the _includes
directory for your use.
roll()
requires a source directory, and takes an optional configuration object as an argument.
By default, roll()
will include formatted post titles, dates (in MM/DD/YYYY format), and a 200 character snippet for each post. Configuration options:
{
//includes a formatted title for each post.
//set to false to exclude titles.
title:true,
//includes a formatted date for each post.
//set to false to exclude dates.
date: true,
//sets date format. Currently accepts any combination of MM/DD/YYYY.
//More eloquent date parsing is a WIP.
dateFormat: 'MM/DD/YYYY',
//provides a snippet of length equal to snippetChars.
//set to false to exclude snippets.
snippet: true,
//default snippet length is 200 characters.
//set snippetChars to any integer value to modify snippet length.
snippetChars: 200
//adds a "read more..." link to the end of post snippets
//snippet must be set to true as well for the link to appear
readMore: true
}
Future configuration options will include more
dateFormat
options, apostNum
option for limiting the number of posts displayed, and apaginate
option to allow for multiple pages of posts equal topostNum
.
To use roll()
's output, be sure to include {{{ blogroll }}}
somewhere in your page's layout (usually landing.html
for most blogs), and include roll()
in your build process. Some example usages:
//with defaults options only
blog.roll('path/to/source/directory');
//without dates or snippets
blog.roll('path/to/source/directory', {
date: false,
snippet: false
});
//with 1000-character snippets
blog.roll('path/to/source/directory', {
snippetChars: 1000
});
mixin(source)
Just like the built-in roll()
method, you can create your own modules with JavaScript in the _mixins
directory. As an example, a sidebar.js
mixin is included in the example project. All mixins should be treated like CommonJS modules with a module.exports
function that outputs a function called output()
, which in turn returns the valid html that you would like included, e.g.:
// example.js
const plainText = "I'm some boring plain text";
module.exports = {
output(){
const html = `<p>${plainText}</p>`;
return html;
}
}
Then, by calling mixin()
on your top-level directory, you should see an include with the same name as your mixin (in this case, example.html
) with the HTML that you exposed ready to be included into your final posts with mustache syntax.
build(source)
The bread-and-butter method of blog-runner, build()
generates the _site
directory. No other function modifies the _site
directory, so make sure that if you're including custom mixins or generators (like roll()
) that you execute those before executing build()
.
A source is required for build()
to work:
blog.build('path/to/blog/directory');
Built and maintained by Alex Pearson