hubeiqiao/tim-project-skills/python-docx-style-id-mismatch

python-docx Style ID Mismatch (OxmlElement)

What This Skill Helps With

Use this skill when your low-level python-docx XML edits look correct in code but Word still renders the paragraph as Normal instead of the heading or list style you expected.

Ask for this skill with prompts like

• Use $python-docx-style-id-mismatch to fix why my Heading 2 paragraphs render as Normal.
• Use $python-docx-style-id-mismatch to explain the difference between style names and style IDs in python-docx.
• Use $python-docx-style-id-mismatch to review this OxmlElement paragraph insertion code.

Problem

When creating Word document paragraphs using python-docx's low-level OxmlElement API, setting w:pStyle with the style's display name (e.g., "Heading 2") causes the paragraph to silently fall back to "Normal" style. No error is raised. The XML requires the style ID (e.g., "Heading2") which differs from the display name.

Context / Trigger Conditions

• Creating paragraphs via OxmlElement('w:p') instead of doc.add_paragraph()
• Setting style with pStyle.set(qn('w:val'), 'Heading 2') (WITH space)
• Paragraphs render as Normal/body text instead of the intended style
• paragraph.style.name shows "Normal" even though you set a heading style
• No error or warning is raised -- complete silent failure
• Affects all styles with spaces in their display names

Root Cause

python-docx has two name systems:

• Display name (style.name): Human-readable, e.g., "Heading 2", "List Bullet", "Body Text"
• XML style ID (style.style_id): Used in the XML, e.g., "Heading2", "ListBullet", "BodyText"

When you use doc.add_paragraph(style='Heading 2'), python-docx handles the mapping internally. But when using OxmlElement directly to set w:pStyle, you must provide the XML style ID.

Solution

Create a mapping dictionary and helper function:

STYLE_ID_MAP = {
    'Normal': 'Normal',
    'Heading 1': 'Heading1',
    'Heading 2': 'Heading2',
    'Heading 3': 'Heading3',
    'Heading 4': 'Heading4',
    'List Bullet': 'ListBullet',
    'List Number': 'ListNumber',
    'Caption': 'Caption',
    'table of figures': 'TableofFigures',
    'Body Text': 'BodyText',
    'First Paragraph': 'FirstParagraph',
    'Compact': 'Compact',
}
def get_style_id(style_name):
    """Convert a style display name to its XML style ID."""
    return STYLE_ID_MAP.get(style_name, style_name.replace(' ', ''))

Then use it when setting styles via OxmlElement:

# WRONG - silent failure, falls back to Normal
pStyle.set(qn('w:val'), 'Heading 2')
# CORRECT - uses XML style ID
pStyle.set(qn('w:val'), get_style_id('Heading 2'))  # -> 'Heading2'

The fallback style_name.replace(' ', '') handles most cases since Word typically just removes spaces, but some styles have non-obvious IDs (e.g., "table of figures" -> "TableofFigures" with different casing).

Verification

After applying the fix, verify styles are correctly assigned:

for i, p in enumerate(doc.paragraphs):
    if p.style.name.startswith('Heading'):
        print(f"[{i}] style.name='{p.style.name}' text='{p.text[:60]}'")

If headings appear as "Normal" in this output, the style ID is wrong.

To inspect what XML style ID a document actually uses:

for style in doc.styles:
    if style.name.startswith('Heading'):
        print(f"Display: '{style.name}' -> ID: '{style.style_id}'")

Related Issue: Section Breaks Lost During Paragraph Removal

When removing paragraphs between chapter boundaries using lxml's body.remove(element), any w:sectPr elements embedded in paragraph properties (w:pPr) are removed along with the paragraphs. This silently destroys section breaks (used for Roman/Arabic page numbering).

Fix: After content replacement operations, verify section count and re-add section breaks if needed:

# Check sections
print(f"Sections: {len(doc.sections)}")
# Re-add section break before a target paragraph
prev_para = doc.paragraphs[target_idx - 1]
prev_pPr = prev_para._element.get_or_add_pPr()
sectPr = OxmlElement('w:sectPr')
# ... configure sectPr properties ...
prev_pPr.append(sectPr)

Example

Full pattern for creating a heading paragraph via OxmlElement:

from docx.oxml import OxmlElement
from docx.oxml.ns import qn
def create_heading(doc, ref_element, text, level=2):
    p = OxmlElement('w:p')
    pPr = OxmlElement('w:pPr')
    pStyle = OxmlElement('w:pStyle')
    # Key line: use get_style_id, not raw display name
    pStyle.set(qn('w:val'), get_style_id(f'Heading {level}'))
    pPr.append(pStyle)
    p.append(pPr)
    r = OxmlElement('w:r')
    t = OxmlElement('w:t')
    t.set(qn('xml:space'), 'preserve')
    t.text = text
    r.append(t)
    p.append(r)
    ref_element.addprevious(p)
    return p

Notes

• This only affects the low-level OxmlElement API. Using

doc.add_paragraph(style='Heading 2') works correctly because python-docx resolves the name internally.

• The OxmlElement approach is necessary when inserting paragraphs at

specific positions (before a reference element) rather than appending to the end.

• Custom styles defined in a document may have arbitrary IDs. Always

check style.style_id for the actual XML ID if unsure.

• The style_name.replace(' ', '') heuristic works for ~95% of built-in

Word styles but won't catch casing differences.