Software Design by Example in Python 12: A Template Expander

When I was outlining the content for Software Design by Example there were two natural follow-ons to the HTML checker. One was a template expander of the sort used in static site generators (like the one that created the page you’r reading right now). I chose to tackle it first because it calls back to the simple interpreter built earlier with its variables, loops, and conditionals.

As Chapter 12: A Template Expander says, there are basically three approaches to HTML templating:

1. Mix commands in an existing language such as JavaScript with the HTML or Markdown using some kind of marker to indicate which parts are commands and which parts are to be taken as-is.

2. Create a mini-language with its own commands like Jekyll. Mini-languages are appealing because they are smaller and safer than general-purpose languages, but eventually they acquire most of the features of a general-purpose language.

3. Put directives in specially-named attributes in the HTML. This approach is the least popular, but it eliminates the need for a special parser.

I chose #3 to avoid having to write, test, debug, and explain another parser. The resulting syntax isn’t something I’d want to use:

<ul z-loop="item:names">
  <li><span z-var="item"/></li>
</ul>

but it works well enough to illustrate the key ideas.

If there is ever a sequel to this book, I would really like to take this a step further and show how to build a computational notebook that re-executes code and captures its output. I did make a start at that, but realized that concurrency and process management are hard enough that they need an entire book of their own.

Terms defined: abstract class, abstract method, Application Programming Interface, Boolean expression, static site generator, truthy.