elazar / scribing
A simple static blog generator
Requires
- cocur/slugify: ^2.5
- elazar/auryn-configuration: ^1.0
- league/commonmark: ^0.15.3
- league/plates: ^3.3
- rdlowrey/auryn: ^1.4
- symfony/console: ^3.2
- zendframework/zend-feed: ^2.8
README
Scribing is a simple static blog generator, or static site generator designed for generating a blog, written in PHP.
Installation
Use Composer.
composer require elazar/scribing
Example
If you'd like to see a working example of a site that uses Scribing, check out this repository.
Content
Content for Scribing is written in Common Markdown.
Scribing supports two document types: posts and pages.
Posts
Posts include a top level heading containing a title followed by a date formatted with emphasis.
# Post Title *September 28, 2017* Content goes here
Scribing both builds an individual file for each post and also includes a link to it on an archive page. Its path is derived from the date and title included in the Markdown content file.
Pages
Pages include a top level heading, but unlike posts, they are not date-specific. Often, their content rarely changes.
# Page Title
Content goes here
Scribing builds a file for each page at a path based on the name of the Markdown content file.
Design
Scribing uses the Plates template engine.
When it is run, Scribing requires a template path to be specified. This path must reference a directory containing two Plates template files: the layout template and the archive page template.
Layout Template
layout.php
is used as the layout template for both posts and pages.
HTML generated from Markdown content is passed into the $content
variable.
The top level heading is passed into the $title
variable. It can be used in
the document's <title>
tag.
Archive Template
archive.php
is used to generate the content for the archive page. Unlike the
content for other pages, archive page content is generated from data contained
in post content files rather than being created manually in Markdown.
An associative array of posts keyed by year is passed into $posts
, where the
value referenced by each year is an enumerated array of posts for that year.
Each post is represented by an associative array containing 'date'
, 'url'
,
and 'title'
keys.
Common Data
For any common data that needs to be shared between templates, such as for navigation, create a PHP file that returns an array of data.
return [ 'nav' => [ '/' => 'Home', '/archive' => 'Archive', // ... ], ];
Generating the Site
To generate the static files for a blog based on the post and page content
files and Plates template files you've created, you use the scribing
console
command.
Posts and Pages
scribing
supports multiple subcommands, two of which are: build:posts
and
build:pages
. Both use the same parameters.
build:posts [--templateData=path/to/data.php] <sourcePath> <destinationPath> <templatePath>
--templateData
is optional and used to specify a path to a common data file,
if one is needed.
<sourcePath>
is a directory that contains the source files for posts or pages.
<destinationPath>
is a directory into which scribing
will store the files
it generates.
<templatePath>
is a directory containing the template files.
To build pages instead of posts, simply substitute build:pages
for
build:posts
in the example above.
Feed
To build a feed from posts, scribing
supports a build:feed
subcommand.
build:feed --feedTitle="Feed Title" --feedLink="https://example.com/feed.xml" <sourcePath> <destinationPath>
--feedTitle
is a title for the feed required by the Atom format
--feedLink
is an absolute URL to the feed required by the Atom format
<sourcePath>
is a directory that contains the source files for posts to include in the feed
<destinationPath>
is a directory into which scribing
will store the feed file
Using GitHub Pages
If you are using Scribing to generate a site to be hosted on GitHub Pages, set up a repository to store your content, templates, stylesheets, images, etc. that is separate from the repository that GitHub uses to display the site.
Here's a recommended directory structure:
.
├── content
│ ├── pages
│ └── posts
├── images
├── scripts
├── styles
└── templates
The purpose of most of the directories above should be self-explanatory.
The scripts
directory is intended to house build and utility scripts used to
tweak or deploy the files generated by Scribing.
For example, Scribing doesn't handle stylesheet or content license files, so these have to be manually copied to the build directory.
If you want to create a custom 404 page, you can author the content as a normal page and then move the generated file to an appropriate location in your build script like so:
mv build/404/index.html build/404.html ; rm -fR build/404
If you'd like to locate the latest post and use that as the landing page for the site, you can do it like so:
LATEST=`find build/ -type f | grep -E '[0-9]+' | grep -v '404' | sort -nr | head -n 1` cp "$LATEST" build/index.html
Here are some related resources for further reading:
- Securing your GitHub Pages site with HTTPS
- Quick start: Setting up a custom domain
- GitHub Pages and Single-Page Apps
- GitHub Pages Deployment - Travis CI
License
The source code for Scribing is licensed under the MIT License.