The pySynTeX Package

With its beginner-friendly syntax and extensive package ecosystem, the Python programming language has proved useful to many scientists across a number of fields.

The pySynTeX package is intended as a documentation tool for these scientists, by providing function decorators that allow for symbolic binding between a textual representation in LaTeX and the parameters of a function. This is best illustrated by an example, which also exemplifies how simple using pySynTeX is.

@equation(r"\sum\limits_{i = 0}^{n} a \cdot r^i", n="n", init="a", ratio="r")
def geometric_progression(n, init, ratio):
    total = 0

    for i in range(n + 1):
        total += ratio ** i

    return init * total

There are a number of observations we can make straight away. Firstly, the first argument of the decorator is a string written in LaTeX syntax which represents in mathematical notation the equation of a geometric progression. Secondly, observe that the subsequent arguments bind a paramater of the function to the corresponding symbol in the LaTeX string. This binding allows for the substitution of symbols with numbers. For example,

print(geometric_progression.syntex(9, 1, 1))

results in the output

"\sum\limits_{i = 0}^{9} 1 \cdot 1^i"

Observe how geometric_progression is no longer simply a function, but is rather an extended callable object, by virtue of the @equation decorator.

We can even include the result of the function for a given set of parameter values, by using Python-style formatted strings. For example,

print(geometric_progression.syntex_eval(" = {}", 9, 1, 1))

results in the output

"\sum\limits_{i = 0}^{9} 1 \cdot 1^i = 10"

Indeed, syntex and syntex_eval calls returns an object defining multiple different representations. For example, the package has support for displaying the result as typeset LaTeX if run in a Jupyter notebook environment. More so, we can concatenate different symbolic expressions, as demonstrated below:

In contrast to existing solutions, pySynTeX requires no modification to existing code, besides the addition of a simple decorator.

The long term goal of pySynTeX is to develop into a debugging tool for logical errors - namely by introducing the capability of traversing a function's abstract syntax tree and generating a human-readable expression of the function body in LaTeX. This would allow one to better interpret whether a function is doing what is desired or not.