!/usr/bin/env python # convert pdf slides, into markdown. # extract correct images from each page # and then the text as well.

import sys import tempfile import subprocess import os import xml.etree.ElementTree as ET

print(“”“

layout: tutorial_slides

title: “” questions: objectives: key_points: contributions:

authorship:
editing:

“”“)

PDF = os.path.abspath(sys.argv)

# pdftohtml -xml -q -stdout CAN_Module1_Lecture.pdf xml = subprocess.check_output([‘pdftohtml’, ‘-xml’, ‘-q’, ‘-stdout’, PDF]).decode(‘utf-8’) tree = ET.fromstring(xml) for kid in tree:

sys.stdout.write(f"\n\n--- # {kid.attrib['number']}\n\n")
lasttop = 0
for e in kid:
    if e.tag == 'fontspec': continue

    if e.tag == "image":
        print(f"![](images/{os.path.basename(e.attrib['src'])})")
    elif e.tag == 'text':
        if e.text is not None:
            if abs(int(e.attrib['top']) - lasttop) < 5:
                sys.stdout.write(" " + e.text)
            else:
                sys.stdout.write("\n" + e.text)
        else:
            if len(e) == 1:
                if abs(int(e.attrib['top']) - lasttop) < 5:
                    sys.stdout.write(" " + ET.tostring(e[0]).decode('utf-8'))
                else:
                    sys.stdout.write("\n" + ET.tostring(e[0]).decode('utf-8'))
            else:
                raise Exception("Unsupported")
    else:
        raise Exception(f"Unsupported tag {e.tag}")
    lasttop = int(e.attrib['top'])