Commit 7ae0a0ea authored by Miguel Angel Reina Ortega's avatar Miguel Angel Reina Ortega
Browse files

Add fix for editor's note containing a reference

parent 81ed0daa

generateBaseline/postprocessing.py

+30 −7
Original line number Diff line number Diff line
@@ -1897,9 +1897,27 @@ def update_references(docx_input, docx_output): 
    new_style = "EX"

    counter = 0

    

    def _paragraph_text_from_runs(para):

        # Build paragraph text from runs (including runs inside hyperlinks).

        parts = []

        for run in para.xpath('.//w:r', namespaces=ns):

            for child in run:

                if child.tag == f"{{{ns['w']}}}t" and child.text:

                    parts.append(child.text)

                elif child.tag in (f"{{{ns['w']}}}tab", f"{{{ns['w']}}}br"):

                    parts.append(" ")

        return ''.join(parts).strip()



    # Loop over all paragraphs with style "BodyText" or "FirstParagraph" and change to "EX"

    for para in root.xpath('.//w:p[w:pPr/w:pStyle[@w:val="BodyText" or @w:val="FirstParagraph"]]', namespaces=ns):

        old_val = para.get(f"{{{ns['w']}}}val")

        

        # Keep Editor's Notes untouched here; they are formatted later in

        # update_body_text_style (including highlighting).

        para_text = _paragraph_text_from_runs(para)

        if re.search(r'Editor[\'’]s\s+Note:', para_text, re.IGNORECASE):

            continue



        # A reference is a paragraph that contains:

        # - A pPr with pStyle "BodyText" or "FirstParagraph" (already checked in loop)

        # - A bookmarkStart and a bookmarkEnd
@@ -2001,6 +2019,17 @@ def update_body_text_style(docx_input, docx_output): 
    counter = 0

    h6_counter = 0

    

    def _paragraph_text_from_runs(para):

        # Build paragraph text from runs (including runs inside hyperlinks).

        parts = []

        for run in para.xpath('.//w:r', namespaces=ns):

            for child in run:

                if child.tag == f"{{{ns['w']}}}t" and child.text:

                    parts.append(child.text)

                elif child.tag in (f"{{{ns['w']}}}tab", f"{{{ns['w']}}}br"):

                    parts.append(" ")

        return ''.join(parts).strip()



    # Loop over all elements to find "BodyText" and "FirstParagraph" and change to "Normal"

    # Combine both XPath queries

    body_text_elems = root.xpath('.//w:pStyle[@w:val="BodyText"]', namespaces=ns)
@@ -2008,7 +2037,6 @@ def update_body_text_style(docx_input, docx_output): 
    all_elems = body_text_elems + first_para_elems

    

    for elem in all_elems:

        full_text = ""

        # elem is w:pStyle, its parent is w:pPr, and w:pPr's parent is w:p

        pPr = elem.getparent()

        if pPr is None:
@@ -2016,12 +2044,7 @@ def update_body_text_style(docx_input, docx_output): 
        para = pPr.getparent()

        if para is None:

            continue

        runs = para.xpath('./w:r', namespaces=ns)

        for run in runs:

            text_elems = run.xpath('.//w:t', namespaces=ns)

            for text_elem in text_elems:

                if text_elem.text:

                    full_text += text_elem.text

        full_text = _paragraph_text_from_runs(para)

        

        # Strip whitespace and check if text matches "Essential patents" or "Trademarks"

        full_text_stripped = full_text.strip()