How to install voldemort
- Download and install ActivePython
- Open Command Prompt
- Type
pypm install voldemort
Lastest release
Voldemort
Voldemort is a simple static site generator using Jinja2 and markdown templates.
Installation
sudo python setup.py install
or
sudo easy_install -U voldemort
Usage Options
Usage: voldemort [options] Options: -h, --help show this help message and exit -w WORK_DIR, --work_dir=WORK_DIR Working Directory -s, --serve Start the HTTP Server -p PORT, --port=PORT Port inwhich the HTTPServer should run -d, --deploy Deploy this website -u USER, --user=USER Login name for server -a AT, --at=AT Server address to deploy the site -t TO, --to=TO Deployment directory --skip-blog Skip blog posts generation --skip-pages Skip pages generation --skip-tags Skip tags generation --skip-feeds Skip Atom feed generation --skip-sitemap Skip sitemap generation
Usage Example
Go to the example directory
cd example
and run
voldemort
start the HTTPServer
voldemort --serve -p 8080
Open your browser and see the website in action.
Deploy the website
voldemort --deploy -u foobarnb -a foobarnbaz.com -t /home/foobarnbaz/public_html
Writing posts
Posts mainly contain 2 sections. Config section and the Template section. All data inside two --- defines the config and are validated as YAML data. You can set your post related attributes here. In template section you can use Jinja2 templates or Markdown in {% markdown %} and {% endmarkdown %} blocks (You could ignore these blocks if layout is defined in the metadata section).
As per Voldemort's default configuration, all base templates should be in layout and include directories. This is not a hard limitation, but kept for preserving the meaning. Posts are written in a directory named posts. For example, we have a directory structure as shown below
layout/ listing.html post.html include/ navigation.html posts/ voldemort-is-awesome.markdown index.html css/ screen.css pygments.css
And we have the following data in layout/listing.html
<!DOCTYPE html> <html lang="en-US"> <head> <title>foobarnbaz.com - {{ page.title }}</title> {% include "head-common.html" %} </head> <body> <section class="page-content"> {% block content %}{% endblock %} </section> </body> </html>
and include/header.html contains
<meta charset="UTF-8" /> <meta http-equiv="content-type" content="text/html; charset=utf-8" /> <meta name="author" content="Sreejith K" /> <link rel="alternate" href="http://feeds2.feedburner.com/foobarnbaz" title="foobarnbaz.com" type="application/atom+xml" /> <link rel="stylesheet" href="/css/screen.css" type="text/css" /> <link rel="stylesheet" href="/css/pygments.css" type="text/css" /> <link href='/images/layout/favicon.ico' rel='shortcut icon' type='image/ico' />
We will be able to write the following index.html which generates the front page of your blog with all the posts, paginated with the value provided in settings.yaml (defaults to 5).
--- paginate: true --- {% extends "listing.html" %} {% block content %} {% for post in paginator.posts %} <article class="excerpt"> <header> <h1><a href="{{ post.url }}">{{ post.title }}</a></h1> <time datetime="{{ post.date | date_to_string }}" pubdate="pubdate"> {{ post.date.strftime("%b %d, %Y") }} </time> </header> {% if loop.first %} {{ post.content }} <p class="full-post"><a href="{{ post.url }}#comments">comments...</a></p> {% else %} <p>{{ post.content }}</p> <p class="full-post"><a href="{{ post.url }}">full post...</a></p> {% endif %} </article> {% endfor %} {% endblock %}
And our sample post posts/voldemort-is-awesome.markdown,
--- title: Voldemort date: '02-10-2011' time: '10:45' layout: 'post.html' --- [Voldemort](https://github.com/semk/voldemort) is an awesome static site generator based in Jinja2 and Markdown templates.
For more information about templating read the following documentations.
- Read Jinja2 Documentation
- Read Markdown Documentation
Configuration
You can change the default settings by editing the settings.yaml.
layout_dirs : - layout # directory inwhich base tempaltes reside - include # html code that can be included goes here posts_dir : posts # directory where you write posts post_url : "%Y/%m/%d" # url to posts. You can alter the order site_dir : _site # generated site will be in this directory paginate : 5 # number of pages to be paginated at once
User defined data should only be added under site as shown below
site : name : "Pythoned!" address : "http://foobarnbaz.com" author_name : "Sreejith Kesavan" author_email: "sreejithemk@gmail.com"
and you may deploy your website to a preferred location or GitHub itself.
deploy : user : semk at : github.com to : semk.github.com
Global variables
site: User defined variables from settings.yaml. Also includes site.time Eg: site.name, site.address, site.time posts: A list of all your posts. All attributes in the YAML section can be accessed either using . or []. eg. post['date'], post.date paginator: You can paginate your posts using this object. eg: {% for post in paginator.posts %} Attributes: posts: list of posts in this paginator current_page : current page number (None if not) next_page : next page number (None if not) previous_page : previous page number (None if not) post: Info about the post. Only accessible in posts. Attributes: content : html content of the post url : url to this post id : identifier for the post (url) next : points to the next post previous : points to the previous post tags : points to the tags and you can access all the attributes in the config section (eg: post.date) page: Info about a page. Only available in pages other than posts. Attributes: content : html content of the post and you can access all the attributes in the config section (eg: page.title) tags: Tags for the blog posts. Globally available. Eg: You can loop like {% for tag in tags %} and access tag.name, tag.url and tag.posts tag: Available only to the tag template (Default `tag.html`) Usage: {% for post in tag.posts %}
Filters
Apart from built-in filters provided by Jinja2, Voldemort provides the following filters to use inside HTML pages.
date: Format datetime objects. eg. post.date | date("%d-%m-%Y") date_to_string: Convert date to string. eg. "27 Jan 2011" date_to_long_string: Format a date in long format. eg. "27 January 2011" date_to_xmlschema: Format a date for use in XML. eg. "2011-04-24T20:34:46+05:30" xml_escape: Replace special characters "&", "<" and ">" to HTML-safe sequences. cgi_escape: CGI escape a string for use in a URL. Replaces any special characters with appropriate %XX replacements. uri_escape: Escape special characters in url. number_of_words: Return number of words in a string. excerpt: Split the html data. Eg: {{ post.content | excerpt(70) }}