[GSoC 2026] Introduction & Proposal: Improving the Lark-Based LaTeX Parse

[ILYAS Rahmouni]

Hi SymPy Community and Mentors,

My name is Ilyes Rahmouni, and I am a third-year Applied Mathematics student. I am writing to express my strong interest in contributing to SymPy under Google Summer of Code 2026, specifically on the "Improve the Lark-based LaTeX parser" project.

Given my academic background and my previous experience working with Python to parse and generate LaTeX code (including building automated tools for math students), this project aligns perfectly with my skills and interests.

To understand the current state of the parser, I recently conducted a comparative analysis between the ANTLR and Lark parsers using a benchmark of 25 representative expressions. Through this, I identified a few specific areas that I plan to target:

1. Subscript arithmetic failures (e.g., missing grammar rules for $x_{i+1}$).

2. Unresolved ambiguity trees (e.g., implicit multiplication in $x(x-1)$).

3. Trigonometric parsing inconsistencies compared to the ANTLR output.

I have put together a draft of my GSoC proposal outlining a detailed timeline and methodology to address these issues. I would be very grateful if any of the mentors or community members could take a quick look and share their feedback so I can refine it before the final deadline:

https://drive.google.com/file/d/1lVFvy0fez-kUKxJEd-Moa1jU6tJgTSz2/view?usp=sharing

Currently, I have set up my local development environment and I am exploring the codebase to find a good first issue related to parsing to work on.

Thank you for your time, and I look forward to learning from this amazing community!

Best regards,

Ilyes Rahmouni