Structure and Interpretation of Computer Programs (SICP) Adaptations

This repository provides three editions of Structure and Interpretation of Computer Programs (SICP):

• SICP JS: Adapted for JavaScript. Includes a PDF, an interactive web version, and a comparison edition (Scheme vs. JS).
• SICPy: Adapted for Python. Includes a PDF, an interactive web version, a comparison edition (Scheme vs. Python), and an AI chatbot knowledge base.
• Scheme: The original edition. Includes a PDF and structured JSON content (not yet interactive).

Each edition provides source code archives containing program snippets (e.g., programs_js, programs_py, programs_scm) and single-file Markdown exports.

The chatbot_notes_py/ directory contains a knowledge base designed for Retrieval-Augmented-Generation (RAG) Q&A bots. It is not autogenerated by the XML pipeline but is AI-generated and checked into git.

For each chapter, it provides:

• sicpy_notes_chapter<N>.ex: An Elixir module containing bullet-point summaries and key terms.
• sicpy_index_terms_chapter<N>.json: A lookup table mapping terms to section numbers for keyword-based retrieval.

You can generate the JSON textbook content and extract program snippets from XML sources using Node.js.

Platform Requirements:

• Node.js: Required for yarn json and yarn programs.
• Unix/macOS: Required for generating PDF editions and running the full build. Windows users should use WSL.

Testing the original Scheme edition requires MIT/GNU Scheme to be installed and available on your PATH.

Usage:

# Generate the programs
SICP_EDITION=scm npx tsx ./javascript/index.js programs_scm

# Run all tests
SICP_EDITION=scm yarn test:scm

# Run tests for a specific chapter
SICP_EDITION=scm yarn test:scm -- programs_scm/chapter1

Note: This infrastructure is currently informational; many failures are due to missing cross-snippet prerequisites (REQUIRES) in the Scheme edition.

To install a custom favicon in the project, copy your generated .ico file to the Rails assets directory. The Ruby on Rails asset pipeline will automatically detect and process it for use in the application.

cp lambda1.ico ../../app/assets/images/favicon.ico

You can test the JavaScript example programs (programs_js) against expected outputs using scripts/test.js.

By default, tests run through js-slang. To run tests natively on Node using the sicp npm package instead, set the JS_SLANG environment variable to 0.

Usage:

# Run all tests
yarn test

# Run tests scoped to a specific chapter
yarn test -- programs_js/chapter1

# Run tests natively on Node (using 'sicp' package)
JS_SLANG=0 yarn test:native

To create a favicon, design a simple, bold icon and export it as a PNG. If you require a transparent background, use a tool like Adobe Photoshop with the 'Magic Eraser Tool' to remove white areas before saving.

To transform the PNG into a 32x32 pixel ICO format, use the ImageMagick convert command.

convert -resize x32 -gravity center -crop 32x32+0+0 lambda1_transparent.png lambda1.ico

The addName function extracts a name from a node by looking for the first child element with the tag name NAME. It processes the text content of that element using recursiveProcessTextLatex, joins the resulting parts, trims them, and pushes the name followed by }\n\n into the writeTo array. It returns the extracted name string.

// node: the parent node containing the <NAME> tag
// writeTo: an array to which the formatted name is pushed
const name = addName(node, writeTo);

The processList function iterates through a linked list of DOM nodes. If a node is an LI element, it appends the LaTeX list item marker \item to the writeTo array and then recursively processes the node's first child using recursiveProcessTextLatex. It then continues to the next sibling in the list.

// node: the starting DOM node
// writeTo: an array to which processed strings are pushed
processList(node, writeTo);