Skip to content
Lesson formatter
Branch: master
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Type Name Latest commit message Commit time
Failed to load latest commit information.

*** This repository is soon to be retired. Please read github issue 168 ( for more information. ***

Code Club Lesson Formatter

The lesson formatter transforms our markdown lessons into a static website.

Installation and Running

The site builder is a python script, which uses pandoc for rendering, as well as phantom.js for beautifying scratch code. For generating PDFs, it uses phantom.js.

If you have downloaded a zip file, you will need to set the script pandoc_scratchblocks/ to +x

These instructions are for installing on linux / OSX. You may be able to get it running on Windows, but unfortunately we’re not able to help with that. Sorry.

We recommend using virtualenv when working with any Python project, and have illustrated this as an optional step in the installation instructions below.


We have a dependency, at present, on Pandoc 1.13.2. If you use more recent versions you may see errors similar to those mentioned in Until we remove this dependency, please install 1.13.2, which can be found here:


# clone the repo
git clone --recursive && cd lesson_format

# [optional] create a virtual environment using virtualenv
virtualenv venv
source venv/bin/activate

# install python requirements
pip install -r requirements.txt

# install phantom.js etc
npm install


The script reads from locally checked out files, and writes to an output directory. We run it with the scratch-curriculum, webdev-curriculum, and python-curriculum as inputs, and write to a directory.

The first argument is the theme for the website, which must be cc (as other themes have recently been deprecated)

Compiling Sass

#install bundler
sudo gem install bundler

# install css compilation stuff
bundle install

# install bower dependencies
$(npm bin)/bower install

# compile the Sass
$(npm bin)/grunt sass:dev


./ cc lessons/scratch lessons/webdev lessons/python output/codeclub

To build the PDFs, specify the --pdf phantomjs option as follows:

./ --pdf phantomjs cc lessons/scratch lessons/webdev lessons/python output/codeclub

There is also a makefile which can be used to automate the process.

  • First run make clone to build a local copy of all of the source and target repositories
  • Next, run make pages_cc.

You only need to run make clone once, and not before every make pages_cc.

Lesson Formatting

See this guide to writing projects.

Term Manifests

A manifest file is a json file which is a list of all files needed for a term, and looks like the following:

    "id": "example",
    "title":"Example Term",
    "description":"This is for example purposes",
    "language": "en-GB",
    "number": 1,

    "projects": [
            "filename": "01 Felix and Herbert/01 Felix and",
            "number": 1
            "filename": "02 Ghostbusters/02",
            "number": 2
    "extras": [
            "name": "Volunteer notes",
            "note": "volunteer resources/*.md"
  • id is the identifier, and used for filenames and urls
  • title is the proper name of the term, used for lists and headers
  • description is used in the term page, with a list of projects
  • number is the term number. currently scratch is 1 & 2, webdev is 3, python is 4.

Then there is a list of projects, and a list of extra files to include.

  • Projects must contain filename and number, but optionally any project header can be included here too, including note, materials, title.

You can find the [manifest for the en-GB scratch term 1 here]( 1/en-GB_scratch_term1.manifest).


The lesson formatter can use different languages in the project listing.

    "code": "en-GB",
    "name": "English",
    "legal": {
        "cc": "Disclaimer for Code Club World, OUTSIDE UK"
    "translations": {
        "Download PDF": "…",
        "Download Project Materials": "…",
        "Extras": "…"

You can find a full list of translations in the en-GB.language.sample file.

Underneath the hood

Before starting, themes are loaded from assets/themes/* and language support from assets/languages/*.

Various directories and files from /assets are then copied directly to the output.

All of the input directories are scanned for manifest files, and an index is built for each language, containing all of the terms.

It then creates /<lang-code>/<term>-<num>/<project num>/<project files> for each project and ancillary data, creating indexes by language, term, too.


Run a webserver in the output directory, e.g.

$ cd <output directory>
$ python -m SimpleHTTPServer <port>
You can’t perform that action at this time.