sunainapai / makesite

Simple, lightweight, and magic-free static site/blog generator for Python coders

Take full control of your static website/blog generation by writing your own simple, lightweight, and magic-free static site generator in Python. That's right! Reinvent the wheel, fellas!

View Source View Demo MIT License



This repository contains the source code of an example website containing two static blogs and a few static pages. The website can be generated by running The output looks like this. That's it!

So go ahead, fork this repository, replace the content with your own, and generate your static website. It's that simple!

You are free to copy, use, and modify this project for your blog or website, so go ahead and fork this repository and make it your own project. Change the layout if you wish to, improve the stylesheet to suit your taste, enhance if you need to, and develop your website/blog just the way you want it.

But Why?

For fun and profit! Okay, maybe not for profit, but hopefully for fun.

Have you used a popular static site generator like Jekyll to generate your blog? I have too. It is simple and great. But then did you yearn to use something even simpler to generate your blog? Do you like Python? Perhaps the thought of writing your own static site generator crossed your mind but you thought it would be too much work? If you answered "yes" to these questions, then this project is for you.

With this project, you are in full control. There is no hidden magic! Everything is laid out in as plain and simple Python code. It is just 125 lines of code (excluding blank lines and comments). It gets you off the ground pretty quickly. You can have a decent website/blog generated within a few minutes and then you can begin tinkering with the source code, the layout, and the css to customize the look and feel of your website to your satisfaction.

Get Started

This section provides some quick steps to get you off the ground as quickly as possible.

  1. For a quick demo on your local system, just enter this command:

    make serve

    If you don't have make but have Python 3.x, enter this command:

    cd _site
    python3 -m http.server

    Note: In some environments, you may need to use python instead of python3 to invoke Python 3.x.

    If you only have Python 2.7, enter this command:

    cd _site
    python -m SimpleHTTPServer

    Then visit http://localhost:8000/. It should look like this.

    Note: You can run with Python 2.7 or Python 3.x.

  2. You may see a few Cannot render Markdown warning messages in the output of the previous command. This is due to the fact that an example blog in this project has a few posts written in Markdown. To render them correctly, install the commonmark package with this command:

    pip install commonmark

    Then try the previous step again.

  3. For an Internet-facing website, you would be hosting the static website/blog on a hosting service and/or with a web server such as Apache HTTP Server, Nginx, etc. You probably only need to generate the static files and know where the static files are and move them to your hosting location.

    If you have the make command, enter this command to generate your website:

    make site

    If you don't have make but have python3, enter this command:


    Note: In some environments, you may need to use python instead of python3 to invoke Python 3.x.

    If you only have python, enter this command:


    The _site directory contains the entire generated website. The content of this directory may be copied to your website hosting location.

The Code

Now that you know how to generate the static website that comes with this project, it is time to see what does. You probably don't really need to read the entire section. The source code is pretty self-explanatory but just in case, you need a detailed overview of what it does, here are the details:

  1. The main() function is the starting point of website generation. It calls the other functions necessary to get the website generation done.

  2. First it creates a fresh new _site directory from scratch. All files in the static directory are copied to this directory. Later the static website is generated and written to this directory.

  3. Then it creates a params dictionary with some default parameters. This dictionary is passed around to other functions. These other functions would pick values from this dictionary to populate placeholders in the layout template files.

    Let us take the subtitle parameter for example. It is set to our example website's fictitious brand name: "Lorem Ipsum". We want each page to include this brand name as a suffix in the title. For example, the about page has "About - Lorem Ipsum" in its title. Now take a look at the page layout template that is used as the layout for all pages in the static website. This layout file uses the {{ subtitle }} syntax to denote that it is a placeholder that should be populated while rendering the template.

    Another interesting thing to note is that a content file can override these parameters by defining its own parameters in the content header. For example, take a look at the content file for the home page. In its content header, i.e., the HTML comments at the top with key-value pairs, it defines a new parameter named title and overrides the subtitle parameter.

    We will discuss the syntax for placeholders and content headers later. It is quite simple.

  4. It then loads all the layout templates. There are 6 of them in this project.

    • layout/page.html: It contains the base template that applies to all pages. It begins with <!DOCTYPE html> and <html>, and ends with </html>. The {{ content }} placeholder in this template is replaced with the actual content of the page. For example, for the about page, the {{ content }} placeholder is replaced with the the entire content from content/about.html. This is done with the make_pages() calls further down in the code.

    • layout/post.html: It contains the template for the blog posts. Note that it does not begin with <!DOCTYPE html> and does not contain the <html> and </html> tags. This is not a complete standalone template. This template defines only a small portion of the blog post pages that are specific to blog posts. It contains the HTML code and the placeholders to display the title, publication date, and author of blog posts.

      This template must be combined with the page layout template to create the final standalone template. To do so, we replace the {{ content }} placeholder in the page layout template with the HTML code in the post layout template to get a final standalone template. This is done with the render() calls further down in the code.

      The resulting standalone template still has a {{ content }} placeholder from the post layout template template. This {{ content }} placeholder is then replaced with the actual content from the blog posts.

    • layout/list.html: It contains the template for the blog listing page, the page that lists all the posts in a blog in reverse chronological order. This template does not do much except provide a title at the top and an RSS link at the bottom. The {{ content }} placeholder is populated with the list of blog posts in reverse chronological order.

      Just like the post layout template , this template must be combined with the page layout template to arrive at the final standalone template.

    • layout/item.html: It contains the template for each blog post item in the blog listing page. The make_list() function renders each blog post item with this template and inserts them into the list layout template to create the blog listing page.

    • layout/feed.xml: It contains the XML template for RSS feeds. The {{ content }} placeholder is populated with the list of feed items.

    • layout/item.xml: It contains the XML template for each blog post item to be included in the RSS feed. The make_list() function renders each blog post item with this template and inserts them into the layout/feed.xml template to create the complete RSS feed.

  5. After loading all the layout templates, it makes a render() call to combine the post layout template with the page layout template to form the final standalone post template.

    Similarly, it combines the list layout template template with the page layout template to form the final list template.

  6. Then it makes two make_pages() calls to render the home page and a couple of other site pages: the contact page and the about page.

  7. Then it makes two more make_pages() calls to render two blogs: one that is named simply blog and another that is named news.

    Note that the make_pages() call accepts three positional arguments:

    • Path to content source files provided as a glob pattern.
    • Output path template as a string.
    • Layout template code as a string.

    These three positional arguments are then followed by keyword arguments. These keyword arguments are used as template parameters in the output path template and the layout template to replace the placeholders with their corresponding values.

    As described in point 2 above, a content file can override these parameters in its content header.

  8. Then it makes two make_list() calls to render the blog listing pages for the two blogs. These calls are very similar to the make_pages() calls. There are only two things that are different about the make_list() calls:

    • There is no point in reading the same blog posts again that were read by make_pages(), so instead of passing the path to content source files, we feed a chronologically reverse-sorted index of blog posts returned by make_pages() to make_list().
    • There is an additional argument to pass the item layout template as a string.
  9. Finally it makes two more make_list() calls to generate the RSS feeds for the two blogs. There is nothing different about these calls than the previous ones except that we use the feed XML templates here to generate RSS feeds.

To recap quickly, we create a _site directory to write the static site generated, define some default parameters, load all the layout templates, and then call make_pages() to render pages and blog posts with these templates, call make_list() to render blog listing pages and RSS feeds. That's all!

Take a look at how the make_pages() and make_list() functions are implemented. They are very simple with less than 20 lines of code each. Once you are comfortable with this code, you can begin modifying it to add more blogs or reduce them. For example, you probably don't need a news blog, so you may delete the make_pages() and make_list() calls for 'news' along with its content at content/news.


In this project, the layout template files are located in the layout directory. But they don't necessarily have to be there. You can place the layout files wherever you want and update accordingly.

The source code of that comes with this project understands the notion of placeholders in the layout templates. The template placeholders have the following syntax:

{{ <key> }}

Any whitespace before {{, around <key>, and after }} is ignored. The <key> should be a valid Python identifier. Here is an example of template placeholder:

{{ title }}

This is a very simple template mechanism that is implemented already in the For a simple website or blog, this should be sufficient. If you need a more sophisticated template engine such as Jinja2 or Cheetah, you need to modify to add support for it.


In this project, the content files are located in the content directory. Most of the content files are written in HTML. However, the content files for the blog named blog are written in Markdown.

The notion of headers in the content files is supported by Each content file may begin with one or more consecutive HTML comments that contain headers. Each header has the following syntax:

<!-- <key>: <value> -->

Any whitespace before, after, and around the <!--, <key>, :, <value>, and --> tokens are ignored. Here are some example headers:

<!-- title: About -->
<!-- subtitle: Lorem Ipsum -->
<!-- author: Admin -->

It looks for the headers at the top of every content file. As soon as some non-header text is encountered, the rest of the content from that point is not checked for headers.


Thanks to Susam for writing the documentation and the unit tests.


This is free and open source software. You can use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of it, under the terms of the MIT License.

This software is provided "AS IS", WITHOUT WARRANTY OF ANY KIND, express or implied. See the MIT License for details.


To report bugs, suggest improvements, or ask questions, please visit

Ещё не оценен

Последние записи






RSS / Atom