Attempt to extract SVGs

[manuGil]

PDF miner has no native type for handling SVG elements. Instead, SVG elements in a document are collected using types like LTCurve, LTLine, LTBox, etc. Extracting these type of elements and convert them into SVG files might not be possible.

[manuGil]

This is an alternative using Inkspcape and Python, but it probably cannot distinguish between vectors and images:
https://gist.github.com/vitchyr/4894861de3fa8d4ffcba