Run2 python docx placeholders

python-docx Placeholder Replacement (Improved)

Installation

pip install python-docx

The Split-Run Problem

Word internally splits paragraph text across multiple XML "runs". {{CANDIDATE_NAME}} might be stored as Run1:{{CANDI + Run2:DATE_NAME}}. Always read para.text (which concatenates all runs), apply changes, then write back to runs.

Core Replacement Function

def replace_in_paragraph(para, data):
    """Replace all {{KEY}} placeholders in a paragraph, safe against split runs."""
    text = para.text
    if '{{' not in text:
        return

    new_text = text
    for key, value in data.items():
        new_text = new_text.replace('{{' + key + '}}', str(value))

    if new_text != text and para.runs:
        # Collapse all text into run[0], preserving its formatting
        para.runs[0].text = new_text
        for run in para.runs[1:]:
            run.text = ''

Formatting note: This preserves run[0]'s formatting (font, bold, size) but flattens any mid-paragraph formatting. Acceptable for offer letter templates.

Important: doc.paragraphs vs Table Cell Paragraphs

doc.paragraphs only returns paragraphs in the document body, NOT inside tables. You must separately iterate tables and recurse into cells.

def process_table(table, data):
    """Process all paragraphs in a table, including nested tables."""
    for row in table.rows:
        for cell in row.cells:
            for para in cell.paragraphs:
                replace_in_paragraph(para, data)
            for nested in cell.tables:   # Word allows tables inside table cells
                process_table(nested, data)

Full Template Fill Function

import json
from docx import Document

def fill_template(template_path, data, output_path):
    """
    Fill a Word template by replacing all {{KEY}} placeholders.

    Args:
        template_path: Path to the .docx template
        data: dict of {KEY: value} pairs (keys without {{ }})
        output_path: Where to save the filled document
    """
    doc = Document(template_path)

    # 1. Body paragraphs
    for para in doc.paragraphs:
        replace_in_paragraph(para, data)

    # 2. Tables (with nested table support)
    for table in doc.tables:
        process_table(table, data)

    # 3. Headers and footers — these are NOT in doc.paragraphs!
    for section in doc.sections:
        for para in section.header.paragraphs:
            replace_in_paragraph(para, data)
        for para in section.footer.paragraphs:
            replace_in_paragraph(para, data)

    doc.save(output_path)


def verify_no_remaining_placeholders(doc_path):
    """Check that no {{...}} tokens remain in the filled document."""
    doc = Document(doc_path)
    remaining = []

    for para in doc.paragraphs:
        if '{{' in para.text:
            remaining.append(f'body: {para.text}')

    def check_table(table):
        for row in table.rows:
            for cell in row.cells:
                for para in cell.paragraphs:
                    if '{{' in para.text:
                        remaining.append(f'table: {para.text}')
                for nested in cell.tables:
                    check_table(nested)

    for table in doc.tables:
        check_table(table)

    return remaining  # empty list means success

Recommended Workflow

# 1. Load data (filter only string-valued keys needed for replacement)
with open('employee_data.json') as f:
    data = json.load(f)

# 2. Process conditionals FIRST (before general replacement)
#    See run2_docx-conditional-sections skill

# 3. Fill remaining placeholders
fill_template('template.docx', data, 'output.docx')

# 4. Verify
leftovers = verify_no_remaining_placeholders('output.docx')
if leftovers:
    print("Unfilled placeholders:", leftovers)
else:
    print("Success!")

Common Pitfalls

Pitfall	Fix
Searching para.runs for {{...}}	Always use para.text (handles split runs)
Only iterating doc.paragraphs	Also iterate doc.tables with recursion
Forgetting headers/footers	Iterate doc.sections, access .header and .footer
Processing placeholders before conditionals	Handle {{IF_X}}...{{END_IF_X}} first
No verification step	Always check for remaining {{ tokens after filling