Agent skill

docx

Word document manipulation with python-docx - handling split placeholders, headers/footers, nested tables

Stars 897
Forks 232

Install this agent skill to your Project

npx add-skill https://github.com/benchflow-ai/skillsbench/tree/main/tasks/offer-letter-generator/environment/skills/docx

SKILL.md

Word Document Manipulation with python-docx

Critical: Split Placeholder Problem

The #1 issue with Word templates: Word often splits placeholder text across multiple XML runs. For example, {{CANDIDATE_NAME}} might be stored as:

  • Run 1: {{CANDI
  • Run 2: DATE_NAME}}

This happens due to spell-check, formatting changes, or Word's internal XML structure.

Naive Approach (FAILS on split placeholders)

python
# DON'T DO THIS - won't find split placeholders
for para in doc.paragraphs:
    for run in para.runs:
        if '{{NAME}}' in run.text:  # Won't match if split!
            run.text = run.text.replace('{{NAME}}', value)

Correct Approach: Paragraph-Level Search and Rebuild

python
import re

def replace_placeholder_robust(paragraph, placeholder, value):
    """Replace placeholder that may be split across runs."""
    full_text = paragraph.text
    if placeholder not in full_text:
        return False

    # Find all runs and their positions
    runs = paragraph.runs
    if not runs:
        return False

    # Build mapping of character positions to runs
    char_to_run = []
    for run in runs:
        for char in run.text:
            char_to_run.append(run)

    # Find placeholder position
    start_idx = full_text.find(placeholder)
    end_idx = start_idx + len(placeholder)

    # Get runs that contain the placeholder
    if start_idx >= len(char_to_run):
        return False

    start_run = char_to_run[start_idx]

    # Clear all runs and rebuild with replacement
    new_text = full_text.replace(placeholder, str(value))

    # Preserve first run's formatting, clear others
    for i, run in enumerate(runs):
        if i == 0:
            run.text = new_text
        else:
            run.text = ''

    return True

Best Practice: Regex-Based Full Replacement

python
import re
from docx import Document

def replace_all_placeholders(doc, data):
    """Replace all {{KEY}} placeholders with values from data dict."""

    def replace_in_paragraph(para):
        """Replace placeholders in a single paragraph."""
        text = para.text
        # Find all placeholders
        pattern = r'\{\{([A-Z_]+)\}\}'
        matches = re.findall(pattern, text)

        if not matches:
            return

        # Build new text with replacements
        new_text = text
        for key in matches:
            placeholder = '{{' + key + '}}'
            if key in data:
                new_text = new_text.replace(placeholder, str(data[key]))

        # If text changed, rebuild paragraph
        if new_text != text:
            # Clear all runs, put new text in first run
            runs = para.runs
            if runs:
                runs[0].text = new_text
                for run in runs[1:]:
                    run.text = ''

    # Process all paragraphs
    for para in doc.paragraphs:
        replace_in_paragraph(para)

    # Process tables (including nested)
    for table in doc.tables:
        for row in table.rows:
            for cell in row.cells:
                for para in cell.paragraphs:
                    replace_in_paragraph(para)
                # Handle nested tables
                for nested_table in cell.tables:
                    for nested_row in nested_table.rows:
                        for nested_cell in nested_row.cells:
                            for para in nested_cell.paragraphs:
                                replace_in_paragraph(para)

    # Process headers and footers
    for section in doc.sections:
        for para in section.header.paragraphs:
            replace_in_paragraph(para)
        for para in section.footer.paragraphs:
            replace_in_paragraph(para)

Headers and Footers

Headers/footers are separate from main document body:

python
from docx import Document

doc = Document('template.docx')

# Access headers/footers through sections
for section in doc.sections:
    # Header
    header = section.header
    for para in header.paragraphs:
        # Process paragraphs
        pass

    # Footer
    footer = section.footer
    for para in footer.paragraphs:
        # Process paragraphs
        pass

Nested Tables

Tables can contain other tables. Must recurse:

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

            # Recurse into nested tables
            for nested_table in cell.tables:
                process_table(nested_table, data)

Conditional Sections

For {{IF_CONDITION}}...{{END_IF_CONDITION}} patterns:

python
def handle_conditional(doc, condition_key, should_include, data):
    """Remove or keep conditional sections."""
    start_marker = '{{IF_' + condition_key + '}}'
    end_marker = '{{END_IF_' + condition_key + '}}'

    for para in doc.paragraphs:
        text = para.text
        if start_marker in text and end_marker in text:
            if should_include:
                # Remove just the markers
                new_text = text.replace(start_marker, '').replace(end_marker, '')
                # Also replace any placeholders inside
                for key, val in data.items():
                    new_text = new_text.replace('{{' + key + '}}', str(val))
            else:
                # Remove entire content between markers
                new_text = ''

            # Apply to first run
            if para.runs:
                para.runs[0].text = new_text
                for run in para.runs[1:]:
                    run.text = ''

Complete Solution Pattern

python
from docx import Document
import json
import re

def fill_template(template_path, data_path, output_path):
    """Fill Word template handling all edge cases."""

    # Load data
    with open(data_path) as f:
        data = json.load(f)

    # Load template
    doc = Document(template_path)

    def replace_in_para(para):
        text = para.text
        pattern = r'\{\{([A-Z_]+)\}\}'
        if not re.search(pattern, text):
            return

        new_text = text
        for match in re.finditer(pattern, text):
            key = match.group(1)
            placeholder = match.group(0)
            if key in data:
                new_text = new_text.replace(placeholder, str(data[key]))

        if new_text != text and para.runs:
            para.runs[0].text = new_text
            for run in para.runs[1:]:
                run.text = ''

    # Main document
    for para in doc.paragraphs:
        replace_in_para(para)

    # Tables (with nesting)
    def process_table(table):
        for row in table.rows:
            for cell in row.cells:
                for para in cell.paragraphs:
                    replace_in_para(para)
                for nested in cell.tables:
                    process_table(nested)

    for table in doc.tables:
        process_table(table)

    # Headers/Footers
    for section in doc.sections:
        for para in section.header.paragraphs:
            replace_in_para(para)
        for para in section.footer.paragraphs:
            replace_in_para(para)

    doc.save(output_path)

# Usage
fill_template('template.docx', 'data.json', 'output.docx')

Common Pitfalls

  1. Forgetting headers/footers - They're not in doc.paragraphs
  2. Missing nested tables - Must recurse into cell.tables
  3. Split placeholders - Always work at paragraph level, not run level
  4. Losing formatting - Keep first run's formatting when rebuilding
  5. Conditional markers left behind - Remove {{IF_...}} markers after processing

Expand your agent's capabilities with these related and highly-rated skills.

benchflow-ai/skillsbench

csv-processing

Use this skill when reading sensor data from CSV files, writing simulation results to CSV, processing time-series data with pandas, or handling missing values in datasets.

897 232
Explore
benchflow-ai/skillsbench

pid-controller

Use this skill when implementing PID control loops for adaptive cruise control, vehicle speed regulation, throttle/brake management, or any feedback control system requiring proportional-integral-derivative control.

897 232
Explore
benchflow-ai/skillsbench

yaml-config

Use this skill when reading or writing YAML configuration files, loading vehicle parameters, or handling config file parsing with proper error handling.

897 232
Explore
benchflow-ai/skillsbench

simulation-metrics

Use this skill when calculating control system performance metrics such as rise time, overshoot percentage, steady-state error, or settling time for evaluating simulation results.

897 232
Explore
benchflow-ai/skillsbench

vehicle-dynamics

Use this skill when simulating vehicle motion, calculating safe following distances, time-to-collision, speed/position updates, or implementing vehicle state machines for cruise control modes.

897 232
Explore
benchflow-ai/skillsbench

web-interface-guidelines

Vercel's comprehensive UI guidelines for building accessible, performant web interfaces. Use this skill when reviewing or building UI components for compliance with best practices around accessibility, performance, animations, and visual stability.

897 232
Explore

Didn't find tool you were looking for?

Be as detailed as possible for better results