Commit 8dd82672 authored by Marco Cavalli's avatar Marco Cavalli
Browse files

fix: tags all references when th more occurrences in the same paragraph

parent cfd1e5d3

md_to_docx_converter/src/to_html/postprocessing.py

+33 −26
Original line number Diff line number Diff line
@@ -319,23 +319,34 @@ def add_links_to_references_in_text(soup): 
    REF_REGEX_N = r"(?<!\[)\[(n\.[A-Za-z0-9]+)\]"



    def insert_link_with_reference(

        content, tag_contents, internal_match, is_informative

        content, is_informative

    ):

        if content.parent is None:

            return



        before_text = str(content).split(tag_contents)[0]

        after_text = str(content).split(tag_contents)[1]

        tag_contents = tag_contents.replace("[n.", "[") # normative can't have a prefix, this won't affex informative

        opening_bracket_index = content.find("[")

        closing_bracket_index = content.find("]") + 1



        if opening_bracket_index > 0:

            before_text = content[:opening_bracket_index]

        else:

            before_text = ""

        

        if closing_bracket_index < len(content):

            after_text = content[closing_bracket_index:]

        else:

            after_text = ""

        

        internal_text = content[opening_bracket_index+1:closing_bracket_index-1]



        # prepare the new <a> tag

        link = (

            f"{informative_file}.html#{internal_match}"

            f"{informative_file}.html#{internal_text}"

            if is_informative

            else f"{normative_file}.html#{internal_match}"

            else f"{normative_file}.html#{internal_text}"

        )

        a = soup.new_tag("a", attrs={"href": link})

        a.append(f"{tag_contents}")

        a.append(f"[{internal_text.replace('n.', '')}]")

        content.replace_with(a)



        # Add any remaining text after the <a> tag
@@ -345,25 +356,21 @@ def add_links_to_references_in_text(soup): 
    def process_text_nodes(element):

        for content in list(element.contents):

            if isinstance(content, NavigableString):

                informative_match = re.search(REF_REGEX_I, str(content))

                normative_match = re.search(REF_REGEX_N, str(content))

                if informative_match:

                    # get the content

                    tag_contents = informative_match.group(0)

                    internal_match = informative_match.group(1)

                split_content = content.split(" ")

                for part in split_content:

                    element = NavigableString(part + " ")

                    content.insert_before(element)

                    if re.match(REF_REGEX_I, part):

                        insert_link_with_reference(

                        content, tag_contents, internal_match, is_informative=True

                            element, is_informative=True

                        )



                if normative_match:

                    # get the content

                    tag_contents = normative_match.group(0)

                    internal_match = normative_match.group(1)

                    elif re.match(REF_REGEX_N, part):

                        insert_link_with_reference(

                        content, tag_contents, internal_match, is_informative=False

                            element, is_informative=False

                        )

                content.extract()



            elif isinstance(content, Tag):

            elif isinstance(content, Tag) and not content.name in ["a", "code"]:

                process_text_nodes(content)



    for element in soup.find_all(["p", "div"]):