MathCAT: Math Capable Assistive Technology

is a library that supports conversion of MathML to:

• Speech strings (in several languages) with embedded speech engine commands
• Braille (Nemeth, UEB Technical, CMU, and many others)
• Navigation of math (in multiple ways including overviews)

There are four related projects that make use of MathCAT:

• MathCATDemo -- an online demonstration of some of what can be done with MathCAT
• A python interface for MathCAT -- used by a MathCAT NVDA add-on.
• A C/C++ interface for MathCAT
• A Java interface for MathCAT (thanks to Michael Whapples for working on that)

MathCAT is used in many assistive technologies including NVDA and JAWS.

For more information, see the full documentation.

Test coverage

Line coverage for the Rust test suite is generated weekly with cargo-llvm-cov.

• Run manually on GitHub: open Actions → Test coverage, press Run workflow, run on main. This uploads the llvm-cov-html artifact and commits the same HTML into llvm-cov/ on main (for GitHub Pages; same as the weekly schedule). If the push step fails, check branch protection: allow GitHub Actions to push to main, or use a PAT secret the workflow is allowed to use.
• Run locally: install the tool (cargo install cargo-llvm-cov and add llvm-tools-preview via rustup component add llvm-tools-preview), then from the repo root run cargo llvm-cov --workspace --html --output-dir llvm-cov-html and open llvm-cov-html/index.html.
• Scheduled runs: open the same workflow link, pick the latest successful run, and download the llvm-cov-html artifact; open index.html inside the archive.
• Browsable URL: in Settings → Pages, use branch main and /(root) (same as the docs site). The workflow then keeps coverage at https://daisy.github.io/MathCAT/llvm-cov/ (only files under llvm-cov/ on main are added or replaced).