blacksmith
Advanced tools
A generic static site generator built using flatiron, plates, and marked.
blacksmithBlacksmith sites have a specific directory structure for storing the various parts of your site: settings, layout, partials, pages, and content. Content exists in two forms:
blacksmith will render.All content will be rendered into /public. To render a blacksmith site:
Install Blacksmith
$ npm install blacksmith -g
Render a Site
#
# Defaults to `cwd`
#
$ blacksmith /path/to/your/site
Here's an example of a simple blog that blacksmith would render.
/site-name
#
# Settings for this blacksmith site.
#
.blacksmith
/content
#
# Actual Markdown content to render.
#
/posts
a-post.md
another-post.md
/layouts
#
# Layouts to use for pages. You can specify
# multiple layouts, but default.html is ... the default.
#
default.html
/metadata
#
# Metadata entities which can be reference in content metdata
#
/authors
author-name.json
/partials
#
# HTML for partials inside of pages
#
post.html
sidebar.html
/pages
#
# Metadata for rendering specific pages
#
index.json
post.json
/public
#
# Any additional files for viewing the site. All
# rendered HTML will be placed here. e.g.:
#
/css
styles.css
/img
favicon.png
blacksmith siteEach blacksmith site defines a hierarchical set of components which can be composed to create any type of site you want! A couple of examples:
Lets examine each of these components and where they are stored in your filesystem:
The settings for a given blacksmith site are stored in the .blacksmith file located in the root directory of your site.
{
//
// Default Layout. Specifying a name here will cause Blacksmith to attempt to read
// /layouts/layout-name.json for rendering info. If no layout-name.json file is found
// blacksmith will use `/layouts/layout-name.html`
//
// Default: default
//
"layout": "layout-name"
"pages": {
//
// These settings will be used for all content in
// /content/posts
//
"post": {
"layout": "custom-layout-for-posts",
"partials": {
"html-id": "additional-partials",
"another-id": "for-this-page-only"
}
}
}
}
You can define settings for all components of your site in the .blacksmith file or break them out into individual files within component directories. For example, the above example is equivalent to:
/.blacksmith
{
"layout": "layout-name"
}
/pages/post.json
{
"layout": "custom-layout-for-posts",
"partials": {
"html-id": "additional-partials",
"another-id": "for-this-page-only"
}
}
Layouts are fully-formed HTML files; they are the top of the rendering hierarchy. All content will be inserted into a layout inside of the HTML element with the id="content". Lets look at the worlds simplest layout:
<html>
<head>
<title>Simple Layout</title>
<link rel="stylesheet" href="css/styles.css">
</head>
<body>
<div id="content">
<!-- All rendered content place here-->
</div>
</body>
</html>
Settings for a layout may specified in your .blacksmith file or in /layouts/layout-name.json. Lets look at an example:
/layouts/layout-name.json NOTE: These properties is not yet supported. See Roadmap for more details.
{
//
// The "template" property specifies which HTML file within the /layouts directory
// to use when rendering this layout. In this way a single HTML file can be reused with
// different partials.
//
// If no template is specified then blacksmith will look for /layouts/layout-name.html
// when rendering.
//
"template": "shared-layout.html",
//
// The "partials" property specifies which HTML partial (or partials) to insert at a given
// id within the layout. These partials will be rendered with the metadata for a given
// individual page with content.
//
"partials": {
"sidebar-container": "sidebar",
//
// Multiple partials can be concatentated into a single HTML id.
//
"multi-container": ["partial1", "partial2"]
}
}
Pages allowed you to specify what type of content you wish to render and where to render it within a given layout. As with layouts, pages may specified in your .blacksmith file or in /pages/page-name.json. Lets look at two examples from our blog:
/pages/index.json
This example of particular interest: index.json will always be used to render index.html. In the below example specifies that blacksmith should render a list of post content, which should be truncated, limited to a maximum of 20 posts. Note: All lists are sorted descending by the date the content was created.
{
//
// The "content" property specifies what blacksmith should render
// inside the HTML element with id="content"
//
"content": {
"list": "post",
"truncate": true,
"limit": 20
}
}
/pages/post.json
The rendering information for an individual post is much simpler than our index.json. By creating this file, blacksmith will:
/content/posts using the partial found at /partials/post.html./partials/sidebar.html with the metadata for each Markdown file in /content/posts and append it to the HTML element with id="content" {
"partials": {
"content": "sidebar"
}
}
It is important to take note of the convention:
Convention: By default, page content will be rendered in a partial of the same name by default.
Alternatively we could have specified a specific partial to use. If no partial is specified and the default partial is not found then no metadata would be rendered (in this case author name, date of post, etc).
A list of all options that can be specified in a page.json file are listed below:
{
//
// Specifies the layout for the page.
// Default: Layout for the site.
//
"layout": "custom-layout-for-page",
//
// Specifies a mapping of HTML ids to partial(s). If "content" is specified
// then it is appended after the rendered markdown content.
//
"partials": {
"html-id": "partial-name",
"another-id": ["multiple-partials"]
},
//
// Case 1: Rendering content with a partial
//
"content": "custom-partial"
//
// Case 2: Consolidating multiple content sources in a list.
//
"content": {
"list": "post",
"truncate": true,
"limit": 20
}
}
In blacksmith, "content" is raw Markdown and supporting files (e.g. images) located within the /content directory. Content for individual pages should be placed under /content/page-name.
In our example all content for the post page should be placed under /content/posts. It is important to take note of the convention:
Convention: Page names are always singular, but their content folder will always be plural.
The content for an individual page may also be a directory where supporting files (such as images can be placed). For example if we wanted to create a post with images the directory structure would be:
/site-name
/content
/posts
/post-with-supporting-files
content.md
an-image.png
another-image.png
some-code.js
The directory structure will be respected, but the /content prefix will be dropped. So the full-url for an-image.png would be http://your-site.com/post-with-supporting-files/an-image.png.
All metadata associated with content is stored within the individual Markdown files as link definitions prefixed with 'meta:'. Because of a small limitation in the Markdown format you must use the following syntax to specify metadata:
[meta:author] <> (Author Name)
[meta:title] <> (A Really Long Title Different from the Filename)
[meta:nested:values] <> (Are also supported)
In large content pages it is often useful to have examples or references to other files to be inserted later in rendering. A classic example is code samples in a blog post. This is easy in blacksmith using Content Snippets. Consider the content:
/content
/dir-post
index.md
whatever.js
This is a piece of markdown content to be rendered later. It has a reference to the file "whatever.js"
<whatever.js>
In this example <whatever.js> in index.md would be replaced with the contents of /content/dir-post/whatever.js. Note: Only files with text extensions (.js, .rb, .txt, etc) less than 1MB will be inserted.
Because blacksmith uses marked we get the benefit of Github Flavored Markdown through highlight.js. By default code highlighting is enabled. All you need to do is add some CSS for your site. For example:
pre {
border: 1px solid #e0ded3;
border-radius: 4px;
margin: 10px 10px 40px 10px;
padding: 10px;
background-color: #f0efe8;
overflow: auto;
font-size: 14px;
font-family: Consolas, "Liberation Mono", Courier, monospace;
}
code {
white-space: pre;
color: rgba(0,0,0, 1);
}
code .keyword { font-weight: bold; color: #6089d4; }
code .string, code .regexp { color: green }
code .class, code .special { color: #6089d4 }
code .number { color: red }
code .comment { color: grey }
It is necessary to truncate large content pages when rendering them within compilations (i.e. lists). This is supported by blacksmith with a special identifier in your Markdown files. For example:
/content/posts/dir-post/index.markdown
A simple post that is truncated with content snippets and metadata that is not used in the post.
##!!truncate
<file1.js>
A second file reference with spaces
< file2.js >
[meta:author]: <> (Charlie index)
When rendered in a page like index.json with { truncate: true } only the content above ##!!truncate will be rendered.
Partials are HTML fragments which are inserted into a layout, a page, or another partial. All partials are rendered with plates.
{
"page-details": {
"author": {
//
// Contents of /metdata/authors/author-name.json
//
},
"href": "/full/path/to/page",
"date": "Fully qualified date page was published",
"files": {
"js": [{ "filename": "file1.js", "url": "/full/path/to/file1.js" }],
"img": [{ "filename": "img1.png", "url": "/full/path/to/img1.png" }]
}
}
}
All metadata is placed into partials using a set of simple plates conventions. In the future it may be possible to customize these conventions but that is not currently planned.
map.where('href').is(key).use(key).as('href');map.class('content').use('content').as('value');map.class(key).use(key)map.class(key).use(key);exports.map(metadata[key][0], map);exports.map(metadata[key], map);Some content should only exist in partials when a given key is present in the metadata. This is supported by blacksmith in the following way: The keys in remove must be specified at the fully qualified path into an Object.
/partials/partial-name.json
{
"remove": {
"page-details": {
"author": ["github", "twitter"]
}
}
}
In the corresponding HTML file for the partial any elements with "class=if-[keyname]" will be removed if the is no value for keyname. In the above example using the following template:
/partials/partial-name.html
<div class="page-details">
<div class="author">
<h3>About the author</h3>
<div class="name">Author Name</div>
<div class="if-github">
<a href="github-url" class="github">Author Github</a>
</div>
<div class="if-twitter">
<a href="twitter-url" class="twitter">Author Twitter</a>
</div>
</div>
</div>
and this set of metadata:
{
"page-details": {
"author": {
"name": "Charlie Robbins",
"github": "indexzero",
"github-url": "https://github.com/indexzero"
}
}
}
The output would be the following. Notice how everything inside if-twitter has been removed.
<div class="page-details">
<div class="author">
<h3>About the author</h3>
<div class="name">Charlie Robbins</div>
<div class="if-github">
<a href="https://github.com/indexzero" class="github">indexzero</a>
</div>
</div>
</div>
It is useful when writing a content page to use a key to reference a larger section of metadata stored elsewhere. These other sections of metadata are called metadata references and are loaded by blacksmith from /metadata. For example:
/metdata/dogs/sparky-mcpherson.json
{
"name": "sparky-mcpherson",
"breed": "labridoodle",
"color": "brown",
"age": "3",
"temperment": "friendly"
}
/content/posts/a-post.md
(... Rest of Content ...)
[meta:dog] <> (Sparky McPherson)
When the content for this page finally gets rendered to a partial the metadata will not be { 'dog': 'sparky' }, it will be the entire object stored in /metadata/dogs/sparky.json. By convention:
Convention: For a given `key:value` blacksmith will lookup metadata references for the `value`
in the directory named with the plural of that `key`.
In the above example the key:value pair is dog:Sparky Mcpherson so we attempt to lookup sparky-mcpherson within /metadata/dogs/sparky-mcpherson.json.
Note: "Authors" is a special metadata reference which will always be within the metadata.page-details.
blacksmith render my Site?It's safe to skip this if you're not tinkering with blacksmith internals.
In order to render list pages with all options supported by blacksmith it is necessary to perform multiple rendering passes on a given site to fully render it. Lets examine that process start to finish:
Rendering starts when blacksmith('/full/path/to/your/site', function () { ... }) is invoked. This tells blacksmith where the root directory for a given site is.
Inside Site.prototype.load will load all HTML and rendering settings stored in:
We need to load all of these files before any rendering can begin because they store the necessary settings for rendering.
/contentBefore layout, page, and partial rendering can take place all content (i.e. Markdown and supporting files) must be known to blacksmith. In our example how can we render the page represented by /pages/index.json without all rendered content? Site.prototype.readContent recursively reads /content building this data structure:
/content
/posts
a-post.md
another-post.md
/dir-post
file1.js
file2.js
index.markdown
{
posts: {
"a-post": { _content: { html: "..", markdown: "..", metadata: { ... } } },
"another-post": { _content: { html: "..", markdown: "..", metadata: { ... } } }
"dir-post": {
_content: {
html: "HTML rendered from Markdown",
markdown: "Markdown in index.markdown",
metadata: {
"page-details": {
source: "/dir-post/index.markdown",
title: "Title of page in metadata" || "href of page",
date: "Saturday, November 10, 2012",
href: "/dir-post",
"files": {
"js": [{ "filename": "file1.js", "url": "/full/path/to/file1.js" }],
"img": [{ "filename": "img1.png", "url": "/full/path/to/img1.png" }]
}
}
}
},
_files: ['file1.js', 'file2.js']
}
}
}
As we discussed blacksmith uses hierarchical rendering components. Each component inherits the relevant properties from its parent entity. The full hierarchy is:
Notice that partials can be specified by both layouts and pages. In the event of a conflict the partials specified by the page will always be preferred.
Lets examine a fully-formed data structure for rendering each of our rendering data structures:
Page
{
//
// There are all inherited from Site and Layout that this
// page is associated with
//
"layout": "layout-name"
"html": {
"layouts": { ... },
"partials": { ... }
},
"partials": {
"html-id": "sidebar",
"another-id": "breadcrumbs",
"multiple-partials": ["that-can", "be-overridden"]
},
//
// Metadata is extracted from link definitions on the page prefixed,
// with "meta:{key}". The {key} is what is stored in the metadata property.
//
// e.g. [meta:author] <> (indexzero)
//
"metadata": {
"author": "indexzero"
}
}
All tests are written with vows and can be run with npm:
$ npm test
FAQs
A static site generator built with Node.js, Plates, and Marked.
The npm package blacksmith receives a total of 0 weekly downloads. As such, blacksmith popularity was classified as not popular.
We found that blacksmith demonstrated a not healthy version release cadence and project activity because the last version was released a year ago. It has 3 open source maintainers collaborating on the project.
Did you know?
Socket for GitHub automatically highlights issues in each pull request and monitors the health of all your open source dependencies. Discover the contents of your packages and block harmful activity before you install or update your dependencies.
Research
Security News
Socket researchers uncover a malicious npm package posing as a tool for detecting vulnerabilities in Etherium smart contracts.
Security News
Research
A supply chain attack on Rspack's npm packages injected cryptomining malware, potentially impacting thousands of developers.
Research
Security News
Socket researchers discovered a malware campaign on npm delivering the Skuld infostealer via typosquatted packages, exposing sensitive data.