tex2image 1.0.0

A simple python library for converting latex snippets to images.

tex2image

This is a simple library to generate images from tex snippets.

It uses the standalone latex package to produce PDFs that are already set to an appropriate size, and then uses poppler to convert the PDF to an image. You must make sure that pdflatex (available from TeX Live) and pdftoppm (available in many package managers, and usually distributed as part of poppler-utils) are available for python to execute.

Quick start

Direct usage

Once you have installed pdflatex and pdftoppm, you can use the library like so:

from pathlib import Path
from tempfile import TemporaryDirectory

from tex2image import latex_to_png

with TemporaryDirectory() as temp_dir:
    image_file_path = latex_to_png("Pythagorean Theorem: $a^2 + b^2 = c^2$.", Path(temp_dir))
    # Do something with the image here, before temp_dir gets deleted...

Client / server method

If you would like to install the pdflatex and pdftoppm dependencies in a container, then you may wish to use the simple FastAPI server included in this library. To launch the server, run

pip install tex2image[server]
tex2image-server
# Or run `tex2image-server --help` to see available options.

Alternatively, this library distributes an official Docker image:

docker run --rm -it ghcr.io/olympiad-bot/tex2image

Both methods will serve a FastAPI server: navigate to [host]:[port]/docs (localhost:8000/docs by default) for OpenAPI documentation for the server.

The server can be easily interacted with from python by using the tex2image.client module. For example:

from tex2image import TexRenderingClient

c = TexRenderingClient()
image_bytes = c.latex_to_png("Pythagorean Theorem: $a^2 + b^2 = c^2$.")

For some more useful examples, see the examples folder; in particular, for examples of using the client, see the Discord bot example and the Flask app example.