updated to work with pandoc chunked HTML option (Spezzatino)

--THIS LUA FILTER MUST BE APPLIED TO PANDOC IN THE CORRECT ORDER

--global table for references link

references = {}

annex_headers = {}

        -- fix annex headers

        if el.level == 8 then

            annex_headers[el.content[3].text] =  "#"..el.attr.identifier -- annex letter / link correspondance

            el.level = 1

            return el

        end

        -- fix header level of sub-clauses of annexes 

        if el.content[1].t == "Str" and el.content[1].text:find("^%u%.%d") then

            el.level = el.level + 1

            return el

        end

        --search for style

        return Style(el) 

    function Image(el)

        -- retrieve aspect ratio of image

        local width = tonumber(el.attr.attributes.width:sub(1,-3))

        local height = tonumber(el.attr.attributes.height:sub(1,-3))

        local ratio = height/width

        -- set height

        el.attr = { style = "width: 100%; height: calc(100%*"..ratio..");"}

        local filePath, extension = el.src:match("(.*)%.(.*)$") -- image.png, jpeg or emf

        --fixes extensions

        if extension == "emf" then 

        end

        --adds yellow overlay on top of the image

        if extension == "png" then

            local overlay = pandoc.Span({}, { style = "position: absolute; top: 0; right: 0; bottom: 0; left: 0; background-color: rgba(255, 255, 0, 0.5); pointer-events: none; z-index: 1;", class = "image_overlay"})

            el.attr = { style = "width: 100%; height: auto; position: relative;"}

            local overlay = pandoc.Span({}, { 

                style = "position: absolute; top: 0; right: 0; bottom: 0; left: 0; background-color: rgba(255, 255, 0, 0.5); pointer-events: none; z-index: 1;",

                class = "image_overlay"

            })

            return pandoc.Span({el, overlay}, { style = "position: relative; display: inline-flex;" })

        end

        return el;

    end

    function Pandoc(el)

    function Emph(el)

        el = Linking(el)

        return el        

        return pandoc.RawInline('html', '<i>' .. pandoc.utils.stringify(el) .. '</i>')      

    end

    function Underline(el)

import os

import re

import sys

from dataclasses import dataclass

from docx.enum.text import WD_ALIGN_PARAGRAPH

from docx.shared import Emu, Pt, RGBColor

DEBUG = False #print debug info switch

DEBUG = True #print debug info switch

@dataclass

    return ' '+inline_text

#use new document to try word diff

doc = Document("ETSI_GS_skeleton.docx")

cssrules = {} #collection of parsed css rules as Style objects

def handle_tag(HTML_tag : bs4.element.Tag):

    global para, run, cell, style, list_level, para_style, prev_inline, curr_inline

    ##THIS HANDLES TEXT LITERALS

    if isinstance(HTML_tag, bs4.element.NavigableString):

        if DEBUG:

            print(HTML_tag)

        return

    curr_inline = None     

    #resolve blockquotes

    ##THIS HANDLES BLOCKQUOTES

    if HTML_tag.name == "blockquote":

        for child in HTML_tag.children:

            handle_tag(child)

    #resolve table

    ##THIS HANDLES TABLES

    elif HTML_tag.name == "table":

        handle_table(HTML_tag)

    #resolve header

    ##THIS HANDLES HEADERS

    elif header := re.match(r"h(\d)", HTML_tag.name):

        if  HTML_tag.string:

            heading = doc.add_paragraph(style= "Heading"+header.group(1))

            if DEBUG:

                print("ERROR: header string is not available", list(HTML_tag.children))

    #resolve lists

    ##THIS HANDLES LISTS

    elif HTML_tag.name in ["ol", "ul"]:

        list_level += 1 #start of the list

        list_level -= 1 #end of the list

    #start a new paragraph

    ##THIS HANDLES DIVS AND PARAGRAPHS (BLOCKS)

    elif HTML_tag.name in blocks: 

        #reset default

            handle_tag(child)

    #handle links

    ##THIS HANDLES LINKS

    elif HTML_tag.name == "a":

        old_run = None

        if run: #we wont write here, but we want to save the style info, if present

    # add a new run

    ##THIS HANDLES INLINES

    elif HTML_tag.name in inlines:

        curr_inline = HTML_tag

def handle_table(table: bs4.element.Tag):

    global cell, para

    col_widths = [int(col.attrs["style"][-3:-1]) for col in table.find_all(lambda tag: tag.name == "col")] #list of each column width in percentages

    #each col tag has a style element with something like "width: xx%", this extracts the xx as an intint(

    col_widths = [int(re.search(r'(\d+)%',col.attrs["style"]).group(1)) for col in table.find_all(lambda tag: tag.name == "col")] #list of each column width in percentages

    rows = [row for row in table.find_all(lambda tag: tag.name == "tr")]

    #create an empty table in docx document 

if __name__ == "__main__":

    if len(sys.argv) < 2 or len(sys.argv) > 3:

        print("Usage: html_to_docx.py file.html [style.css]")

    if len(sys.argv) != 2:

        print("Usage: html_to_docx.py <Diretory with html and css files>")

        sys.exit(1)

    # Define the directory path

    dir_path = sys.argv[1]

    directory = os.scandir(dir_path)

    # Get list of files (htmls are sorted based on their number)

    html_files = sorted([entry.path for entry in directory if entry.is_file() and entry.name.endswith('.html') if entry.name != "index.html" ], key = lambda x: int(x.split("/")[-1].split("-")[0]))

    css_files = [entry.path for entry in directory if entry.is_file() and entry.name.endswith('.css')]

    if len(sys.argv) == 3:

    #use first css file in alphabetical order (this should by default be API.css)

    if css_files:

        #keeps long color definition

        cssutils.ser.prefs.minimizeColorHash = False

        #parse css

        parser = cssutils.CSSParser()

        sheet = parser.parseFile(sys.argv[2])

        sheet = parser.parseFile(css_files[0])

        for rule in sheet:

            if DEBUG:

                print(rule.selectorText)

                cssrules[rule.selectorText[1:]] = _style # remove ".' form selector and uses it as key to add style to our rules dictionary

    #parse html

    with open(sys.argv[1], "r") as file:

    #parse htmls

    for html_file in html_files:

        if DEBUG:

            print(f"Parsing: {html_file}")

        with open(html_file, "r") as file:

            html_content = file.read()

            soup = bs4.BeautifulSoup(html_content, "html.parser")

    skip = True

        #construct doc

        for tag in soup.body.children:

        #skip to executive summary, everything before that is already in the skeleton

        if tag.string == "Executive summary":

            skip = False

        if not skip:

            handle_tag(tag)

$endif$

  $math$

$endif$

<script src="dist/bundle.js" defer></script>

  <!--[if lt IE 9]>

    <script src="//cdnjs.cloudflare.com/ajax/libs/html5shiv/3.7.3/html5shiv-printshiv.min.js"></script>

  <![endif]-->

</head>

<body>

  <button style="position: fixed; bottom: 10px; right: 10px;" onclick="toggleStyles()">Toggle Background highlighting</button>

  <button style="position: fixed; bottom: 50px; right: 10px;" id="download_btn">Download HTML</button>

  <label class="switch">Editing

    <input type="checkbox" id="editing">

    <span class="slider round"></span>

  </label>

  <style>

    /* The switch - the box around the slider */

    .switch {

      position: fixed;

      bottom: 90px;

      right: 10px;

      display: inline-block;

      width: 120px;

      height: 34px;

    }

    /* Hide default HTML checkbox */

    .switch input {

      opacity: 0;

      width: 0;

      height: 0;

    }

    /* The slider */

    .slider {

      position: absolute;

      cursor: pointer;

      top: 0;

      left: 60px;

      right: 0;

      bottom: 0;

      background-color: #ccc;

      -webkit-transition: .4s;

      transition: .4s;

    }

    .slider:before {

      position: absolute;

      content: "";

      height: 26px;

      width: 26px;

      left: 4px;

      bottom: 4px;

      background-color: white;

      -webkit-transition: .4s;

      transition: .4s;

    }

    input:checked+.slider {

      background-color: #2196F3;

    }

    input:focus+.slider {

      box-shadow: 0 0 1px #2196F3;

    }

    input:checked+.slider:before {

      -webkit-transform: translateX(26px);

      -ms-transform: translateX(26px);

      transform: translateX(26px);

    }

    /* Rounded sliders */

    .slider.round {

      border-radius: 34px;

    }

    .slider.round:before {

      border-radius: 50%;

    }

  </style>

  <script>

      let styleToggle = false;

$endif$

</header>

$endif$

$if(toc)$

<nav id="$idprefix$TOC" role="doc-toc">

$if(toc-title)$

<h2 id="$idprefix$toc-title">$toc-title$</h2>

$endif$

<ul><li><h1><a href="0-.html">ETSI title</a></h1></li></ul>

$table-of-contents$

</nav>

$endif$

<div id="editor">

$body$

$for(include-after)$

$include-after$

$endfor$

</div>

</body>

</html>