Docx

DOCX Skill

Quick Reference

Reading Content

python-docx

pythonCopy code
from docx import Document

doc = Document("document.docx")
for para in doc.paragraphs:
    print(f"[{para.style.name}] {para.text}")

for table in doc.tables:
    for row in table.rows:
        print([cell.text for cell in row.cells])

pandoc (with tracked changes)

bashCopy code
pandoc --track-changes=all document.docx -o output.md

Creating New Documents with python-docx

Full document example

pythonCopy code
from docx import Document
from docx.shared import Inches, Pt, Cm, RGBColor
from docx.enum.text import WD_ALIGN_PARAGRAPH
from docx.enum.table import WD_TABLE_ALIGNMENT
from docx.enum.section import WD_ORIENT

doc = Document()

# Page setup (US Letter)
section = doc.sections[0]
section.page_width = Inches(8.5)
section.page_height = Inches(11)
section.top_margin = Inches(1)
section.bottom_margin = Inches(1)
section.left_margin = Inches(1)
section.right_margin = Inches(1)

# Default font
style = doc.styles['Normal']
font = style.font
font.name = 'Arial'
font.size = Pt(11)

# Title
title = doc.add_heading('Document Title', level=0)
title.alignment = WD_ALIGN_PARAGRAPH.CENTER

# Body text
doc.add_paragraph('Executive summary paragraph with key findings and recommendations.')

# Heading hierarchy
doc.add_heading('Section 1: Overview', level=1)
doc.add_paragraph('Section content goes here.')

doc.add_heading('1.1 Subsection', level=2)
doc.add_paragraph('Subsection details.')

doc.save('document.docx')

Lists

pythonCopy code
# Bullet list
doc.add_paragraph('First item', style='List Bullet')
doc.add_paragraph('Second item', style='List Bullet')

# Numbered list
doc.add_paragraph('Step one', style='List Number')
doc.add_paragraph('Step two', style='List Number')

# Nested bullets (indent level)
p = doc.add_paragraph('Sub-item', style='List Bullet 2')

Tables

pythonCopy code
table = doc.add_table(rows=4, cols=3)
table.style = 'Table Grid'
table.alignment = WD_TABLE_ALIGNMENT.CENTER

# Set column widths
for row in table.rows:
    row.cells[0].width = Inches(2)
    row.cells[1].width = Inches(3)
    row.cells[2].width = Inches(1.5)

# Header row
headers = ['Name', 'Description', 'Status']
for i, h in enumerate(headers):
    cell = table.rows[0].cells[i]
    cell.text = h
    for p in cell.paragraphs:
        p.runs[0].bold = True
    shading = cell._element.get_or_add_tcPr()
    from docx.oxml.ns import qn
    from lxml import etree
    shd = etree.SubElement(shading, qn('w:shd'))
    shd.set(qn('w:fill'), '333333')
    shd.set(qn('w:val'), 'clear')
    for p in cell.paragraphs:
        p.runs[0].font.color.rgb = RGBColor(0xFF, 0xFF, 0xFF)

# Data
data = [['Project A', 'Core platform rewrite', 'Active'],
        ['Project B', 'API integration', 'Planning'],
        ['Project C', 'Documentation update', 'Complete']]
for r, row_data in enumerate(data, 1):
    for c, val in enumerate(row_data):
        table.rows[r].cells[c].text = val

Images

pythonCopy code
doc.add_picture('chart.png', width=Inches(5))
last_paragraph = doc.paragraphs[-1]
last_paragraph.alignment = WD_ALIGN_PARAGRAPH.CENTER

Headers and footers

pythonCopy code
section = doc.sections[0]
header = section.header
header_para = header.paragraphs[0]
header_para.text = "Company Name"
header_para.style.font.size = Pt(9)

footer = section.footer
footer_para = footer.paragraphs[0]
footer_para.text = "Confidential"
footer_para.alignment = WD_ALIGN_PARAGRAPH.CENTER

Page breaks

pythonCopy code
doc.add_page_break()

Landscape sections

pythonCopy code
new_section = doc.add_section(WD_ORIENT.LANDSCAPE)
new_section.orientation = WD_ORIENT.LANDSCAPE
new_section.page_width = Inches(11)
new_section.page_height = Inches(8.5)

Editing Existing Documents

pythonCopy code
from docx import Document

doc = Document("existing.docx")

# Find and replace text
for para in doc.paragraphs:
    for run in para.runs:
        if "OLD_TEXT" in run.text:
            run.text = run.text.replace("OLD_TEXT", "NEW_TEXT")

# Add content at the end
doc.add_heading('New Section', level=1)
doc.add_paragraph('Additional content appended to the document.')

# Modify styles
for para in doc.paragraphs:
    if para.style.name == 'Heading 1':
        for run in para.runs:
            run.font.color.rgb = RGBColor(0x1E, 0x27, 0x61)

doc.save("modified.docx")

Advanced: XML Editing Workflow

For complex modifications (TOC, footnotes, bookmarks, tracked changes) that python-docx can't handle, edit the raw XML:

bashCopy code
# Unpack
python scripts/office/unpack.py document.docx unpacked/

# Edit XML files in unpacked/word/ using the Edit tool
# Key files: document.xml, styles.xml, [Content_Types].xml

# Add comments
python scripts/comment.py unpacked/ 0 "Review this section"

# Repack
python scripts/office/pack.py unpacked/ output.docx

Key XML patterns

Tracked change (replace "30" with "60"):

xmlCopy code
<w:r><w:t>The term is </w:t></w:r>
<w:del w:id="1" w:author="Claude" w:date="2025-01-01T00:00:00Z">
  <w:r><w:delText>30</w:delText></w:r>
</w:del>
<w:ins w:id="2" w:author="Claude" w:date="2025-01-01T00:00:00Z">
  <w:r><w:t>60</w:t></w:r>
</w:ins>
<w:r><w:t> days.</w:t></w:r>

Smart quotes in XML:

xmlCopy code
<w:t>Here&#x2019;s a quote: &#x201C;Hello&#x201D;</w:t>

DXA unit reference

• 1440 DXA = 1 inch
• US Letter: 12240 x 15840 DXA
• A4: 11906 x 16838 DXA

Conversion

bashCopy code
# .doc to .docx
python scripts/office/soffice.py --headless --convert-to docx document.doc

# .docx to PDF
python scripts/office/soffice.py --headless --convert-to pdf document.docx

# .docx to images
python scripts/office/soffice.py --headless --convert-to pdf document.docx
pdftoppm -jpeg -r 150 document.pdf page

# Accept all tracked changes (produces clean document)
python scripts/accept_changes.py input.docx clean.docx