I created this small script to hopefully improve my quality of life while typesetting things with LaTeX. I found writing TeX to be annoying in many ways that modern markdown-style languages try to address - Once you're used to TeX, it's easy to write, but it never becomes easy to read, especially when writing formulas. See this example, from a graph theory course I took recently:
\[
\sum_{v \in V}(\text{deg$v$})^{2} \geq \frac{\left(
\sum_{v\in V}\text{deg}v\right)^{2}}{n} = \frac{(2m)^{2}}{n}
\]
This produces nice output:
But it's quite difficult to parse what the code does without looking at the result. The end goal of this project is simpler LaTeX code, but so far only a few minor changes have been implemented.
Currently, the script texstream.py takes in a header .tex file, a .td input file and outputs a .tex file, processing with the following rules:
- Backticks are used as math mode delimiters, both inline and in blocks
- The title and author are specified in the first two lines of the .td file
- Imports and other metadata are specified in a header .tex file, which will commonly be shared between a few .tex files
To convert example.td into example.tex using header.tex's declarations,
python texstream.py example.td header.tex example.tex.
A couple other things I'd like to implement:
- Unicode characters like ≥ and ∈ converted into corresponding commands
- Packages imported based on whether or not their commands are used