Feature request for proper figure formatting during conversion with Calibre about safaribooks HOT 4 OPEN

import pathlib
import shutil
import io
import xml.dom.minidom
import xml.etree.ElementTree as ET
import zipfile
import re
import difflib
from pprint import pprint

# https://medium.com/dev-bits/ultimate-guide-for-working-with-i-o-streams-and-zip-archives-in-python-3-6f3cf96dca50

# #############################  Change this  #############################

paths = r"path/to/your/calibre/library"

# #########################################################################

paths = list(pathlib.Path(paths).rglob("*.epub"))
pprint(paths)

# path_backup = pathlib.Path(r"")
# shutil.copyfile(path_backup, path)

for path in paths:
    print("\n"*0 + f'---------------    Processing book: {path.name}    ---------------')

    zip_updated = io.BytesIO()

    with zipfile.ZipFile(path, "a") as zip:
        with zipfile.ZipFile(zip_updated, "w", compression = zipfile.ZIP_DEFLATED) as zip_u:
            files = zip.infolist()

            f_types = ['.xml', '.html', '.xhtml']

            for f in files:
                # print("." * 50, f'  {f.filename}  ', "." * 50)

                # copy not modified files
                if pathlib.Path(f.filename).suffix not in f_types:
                    zip_u.writestr(f, zip.read(f.filename))

                # modify data in archive
                else:
                    data = zip.read(f.filename).decode()
                    data_temp = data

                    # use xml to modify all img tags
                    if False:
                        # data = ET.canonicalize(data, rewrite_prefixes = True)
                        # data_xml = ET.fromstring(data)
                        data_xml = ET.ElementTree(ET.fromstring(data)).getroot()
                        # namespace = data_xml.tag[1:data_xml.tag.index("}")]

                        imgs = data_xml.findall(".//{*}img")  # use XPATH and use wildcard for any namespace
                        for img in imgs:
                            del img.attrib["width"]  # img.attrib.pop("width")
                            del img.attrib["height"]  # img.attrib.pop("height")
                            # img.tag = img.tag[img.tag.index("}") + 1:]  # remove namespace from tag

                        ET.register_namespace("", xml.dom.XHTML_NAMESPACE)  # prevent namespace before tags
                        # ET.register_namespace("", namespace)
                        data = ET.tostring(data_xml, encoding = "unicode")

                    # use simple regex to remove the attributes
                    else:
                        data = re.sub(r" width=\"(.*?)\" height=\"(.*?)\"", "", data)

                    zip_u.writestr(f, bytearray(data, "utf-8"))

                    if False:  # for debugging purposes
                        TMP_data = xml.dom.minidom.parseString(data).toprettyxml()
                        TMP_data_temp = xml.dom.minidom.parseString(data_temp).toprettyxml()
                        print(*[x for x in difflib.Differ().compare(TMP_data_temp.splitlines(), TMP_data.splitlines())
                                if x.find("<img ") >= 0 and not x.startswith("-")], sep = "\n")
                        print(*list(difflib.context_diff(TMP_data_temp.splitlines(keepends = True),
                                                         TMP_data.splitlines(keepends = True), n = 0)), sep = "")

    with open(path, "wb") as zip:
        zip.write(zip_updated.getbuffer())

    zip_updated.close()

# remove width / height attributes from img tags
imgs = book_content.findall(".//{*}img")  # use XPATH and use wildcard for any namespace
for img in imgs:
    del img.attrib["width"]
    del img.attrib["height"]