$ show source-academy/sicp
This repository is the source and build system for the multi-language edition of the classic computer science textbook *Structure and Interpretation of Computer Programs* (SICP). It contains the full text, exercises, and code snippets in XML, organized by chapter and section, for the original Scheme version as well as modern adaptations in JavaScript and Python. The build toolchain, implemented in TypeScript and JavaScript, parses the XML into multiple output formats: PDF (via LaTeX), a static HTML web edition, a JSON edition for the interactive Source Academy frontend, Markdown, and standalone program files. The system handles cross-references, code snippet extraction with dependency resolution, figure inclusion, index generation, and search index construction. The repository also includes CI/CD pipelines for testing generated programs against actual language runtimes (Node, CPython, MIT Scheme) and deploying the web edition to GitHub Pages. Additionally, AI-generated chatbot knowledge base modules for the Python edition are checked in. The codebase is well-structured with a clean edition abstraction and a modular processing pipeline, though it is tightly coupled to the SICP XML schema and the Source Academy ecosystem.
The LLM identified 1 files worth highlighting for reuse:
A generic trie data structure implementation supporting insert, search, and prefix-based autocomplete. The code is self-contained, well-typed (TypeScript), and exposes a clean API (TrieNode.insert, TrieNode.search, TrieNode.allWordsWithPrefix). It is used by the textbook's search index but can be extracted and reused in any project requiring efficient string prefix matching or autocomplete functionality.