Commit c3a058de authored by Marco Cavalli's avatar Marco Cavalli
Browse files

fix: prevent custom styles to be applied twice

parent d2b22cba

md_to_docx_converter/src/to_docx/postprocessing.py

+51 −78
Original line number Diff line number Diff line
@@ -277,10 +277,12 @@ def format_examples_and_notes(doc: Doc): 
                    paragraph.text = f"\t{paragraph.text}"

                elif paragraph.style.name in STYLES_FOR_EXAMPLES_NOTES:

                    if paragraph.style.name == "Source Code":

                        paragraph.style = "EX Source Code"

                        # paragraph.style = "EX Source Code"

                        paragraph.paragraph_format.left_indent = Pt(85.04)  # 3 cm in points

                    

                    if paragraph.style.name == "Compact":

                        paragraph.style = "EX Compact"

                        # paragraph.style = "EX Compact"

                        paragraph.paragraph_format.left_indent = Pt(103.68)  # 3.66 cm in points

                else:

                    # No longer in example or note

                    in_example_or_note = False
@@ -465,33 +467,8 @@ def add_tagged_styles_and_formatting(doc: Doc): 
                and start_match.group(1) == end_match.group(1)

            )



            # if has_both_tags:

            #     style_name = start_match.group(1)



            #     # Separate text portions according to whether they should be styled or not

            #     text_before_start = text[: start_match.start()]

            #     text_to_style = text[start_match.end() : end_match.start()]

            #     text_after_end = text[end_match.end() :]



            #     run.text = text_before_start  # Nothing should happen to the text before the tag



            #     styled_run = paragraph.add_run(text_to_style)

            #     styled_run = apply_formatting_and_styling(styled_run, style_name)



            #     if text_after_end.strip():  # Add any remaining text as another run

            #         paragraph.add_run(text_after_end)

            #         runs = (

            #             paragraph.runs

            #         )  # Refresh runs now that a new run has been added



            #     i += 1

            #     continue



            if has_both_tags:

                style_name = start_match.group(1)

                start_tag = "{{{" + style_name + "}}}"

                end_tag = "{{{/" + style_name + "}}}"

                text = text.replace(start_tag + start_tag, start_tag).replace(end_tag + end_tag, end_tag)



                # Separate text portions according to whether they should be styled or not

                text_before_start = text[: start_match.start()]
@@ -502,75 +479,69 @@ def add_tagged_styles_and_formatting(doc: Doc): 


                styled_run = paragraph.add_run(text_to_style)

                styled_run = apply_formatting_and_styling(styled_run, style_name)

                i += 1

                paragraph._p.insert(

                    i+1, styled_run._element

                )  # Insert the styled run after the current run



                if text_after_end.strip():  # Add any remaining text as another run

                    paragraph.add_run(text_after_end)

                    i += 1

                    paragraph._p.insert(

                        i+1, paragraph.runs[-1]._element

                    )

                    runs = (

                        paragraph.runs

                    )  # Refresh runs now that a new run has been added



                i += 1

                continue



            if not has_start_tag and has_end_tag:

                # Separate text portions according to whether they should be styled or not

                text_before_end = text[: end_match.start()]

                text_after_end = text[end_match.end() :]

            # if not has_start_tag and has_end_tag:

            #     # Separate text portions according to whether they should be styled or not

            #     text_before_end = text[: end_match.start()]

            #     text_after_end = text[end_match.end() :]



                # Keep track of the text to style and this run

                text_to_style += text_before_end

                style_runs.append(run)

            #     # Keep track of the text to style and this run

            #     text_to_style += text_before_end

            #     style_runs.append(run)



                # Reset the old runs and create a new styled run with the accumulated text to style

                for style_run in style_runs:

                    style_run.text = ""

            #     # Reset the old runs and create a new styled run with the accumulated text to style

            #     for style_run in style_runs:

            #         style_run.text = ""



                styled_run = paragraph.add_run(text_to_style)

                styled_run = apply_formatting_and_styling(styled_run, active_style)

            #     styled_run = paragraph.add_run(text_to_style)

            #     styled_run = apply_formatting_and_styling(styled_run, active_style)



                if text_after_end.strip():  # Add any remaining text as another run

                    paragraph.add_run(text_after_end)

                    runs = (

                        paragraph.runs

                    )  # Refresh runs now that a new run has been added

            #     if text_after_end.strip():  # Add any remaining text as another run

            #         paragraph.add_run(text_after_end)

            #         runs = (

            #             paragraph.runs

            #         )  # Refresh runs now that a new run has been added



                # Prepare for other tags

                active_style = None

                style_runs = []

                text_to_style = ""

            #     # Prepare for other tags

            #     active_style = None

            #     style_runs = []

            #     text_to_style = ""



                i += 1

                continue

            #     i += 1

            #     continue



            if not active_style and has_start_tag and not has_end_tag:

                active_style = start_match.group(1)

            # if not active_style and has_start_tag and not has_end_tag:

            #     active_style = start_match.group(1)



                # Separate text portions according to whether they should be styled or not

                text_before_start = text[: start_match.start()]

                text_after_start = text[start_match.end() :]

            #     # Separate text portions according to whether they should be styled or not

            #     text_before_start = text[: start_match.start()]

            #     text_after_start = text[start_match.end() :]



                # Keep just the text before the the tag and start keeping track of the text to style

                run.text = text_before_start

                text_to_style += text_after_start

            #     # Keep just the text before the the tag and start keeping track of the text to style

            #     run.text = text_before_start

            #     text_to_style += text_after_start



                style_runs.append(run)

            #     style_runs.append(run)



                i += 1

                continue

            #     i += 1

            #     continue



            if active_style:  # Inside a tag

                # Simply keep track of this run and its text

                text_to_style += text

                style_runs.append(run)

            # if active_style:  # Inside a tag

            #     # Simply keep track of this run and its text

            #     text_to_style += text

            #     style_runs.append(run)



                i += 1

                continue

            #     i += 1

            #     continue



            i += 1  # No tag here, so just go on to the next run
@@ -582,6 +553,8 @@ def set_keep_with_next_false(doc: Doc): 


    for paragraph in iter_paragraphs(doc):

        paragraph.paragraph_format.keep_with_next = False

        paragraph.paragraph_format.space_after = Pt(12)





    return doc

md_to_docx_converter/src/to_docx/preprocessing.py

+1 −0
Original line number Diff line number Diff line
@@ -432,6 +432,7 @@ def handle_examples_and_notes(soup: BeautifulSoup): 


                    grandchild.clear()

                    grandchild.append(tagged_text)

                    grandchild.unwrap()



        return soup

md_to_docx_converter/src/to_html/postprocessing.py

+13 −0
Original line number Diff line number Diff line
@@ -638,6 +638,18 @@ def fix_custom_tags(soup: BeautifulSoup): 
            a.string = a.string.replace("root", new_id_prefix)

    return soup



def fix_lists(soup: BeautifulSoup):

    """

    Fix lists that have been improperly nested due to markdown conversion.

    """

    listitems = soup.find_all(["li"])



    for item in listitems:

        children = list(item.children)

        if len(children) == 1 and children[0].name in ["p"]:

            children[0].unwrap()

    

    return soup



def extract_images_from_html(soup: BeautifulSoup) -> dict:

    """
@@ -818,6 +830,7 @@ def postprocess(html_dir: str): 
        soup = replace_dash_characters(soup)

        soup = move_figure_id_to_FL_elements(soup)

        soup = fix_custom_tags(soup)

        soup = fix_lists(soup)

        

        images, soup = extract_images_from_html(soup)

        for image_id, image_src in images.items():