from docx import Document from docx.shared import Pt, RGBColor, Inches from docx.enum.text import WD_ALIGN_PARAGRAPH from docx.oxml import OxmlElement from docx.oxml.ns import qn import redef parse_markdown_table(table_text): “””Parse a markdown table into rows and columns””” lines = [line.strip() for line in table_text.strip().split(‘\n’) if line.strip()] if len(lines) < 2: return [] # Remove separator line rows = [lines[0]] + lines[2:] parsed_rows = [] for row in rows: cells = [cell.strip() for cell in row.split('|')] cells = [c for c in cells if c] # Remove empty cells if cells: parsed_rows.append(cells) return parsed_rowsdef convert_latex_to_word_equation(latex_text): """Convert LaTeX math to Word OMML format (simplified)""" # This is a simplified conversion - for complex formulas, you may need additional libraries latex_text = latex_text.strip('$').strip() # Basic conversions conversions = { r'\frac': '/', r'\times': '×', r'\div': '÷', r'\pm': '±', r'\leq': '≤', r'\geq': '≥', r'\neq': '≠', r'\sqrt': '√', r'\sum': '∑', r'\int': '∫', r'\infty': '∞', r'\alpha': 'α', r'\beta': 'β', r'\gamma': 'γ', r'\delta': 'δ', r'\pi': 'π', r'\theta': 'θ', r'\lambda': 'λ', r'\mu': 'μ', r'\sigma': 'σ', } result = latex_text for latex, symbol in conversions.items(): result = result.replace(latex, symbol) # Remove common LaTeX commands result = re.sub(r'\\[a-zA-Z]+', '', result) result = result.replace('{', '').replace('}', '') return resultdef process_text_with_math(text, paragraph): """Process text that may contain inline math formulas""" # Match LaTeX formulas: $...$ or $$...$$ pattern = r'\$\$(.+?)\$\$|\$(.+?)\$' last_end = 0 for match in re.finditer(pattern, text): # Add text before formula if match.start() > last_end: paragraph.add_run(text[last_end:match.start()]) # Add formula (converted) formula = match.group(1) or match.group(2) formula_run = paragraph.add_run(convert_latex_to_word_equation(formula)) formula_run.italic = True formula_run.font.color.rgb = RGBColor(0, 0, 128) last_end = match.end() # Add remaining text if last_end < len(text): paragraph.add_run(text[last_end:])def convert_ai_text_to_word(input_text, output_filename="document_converti.docx"): """Convert AI-generated text (with tables and formulas) to Word document""" doc = Document() # Add title title = doc.add_heading('Document Converti depuis IA', 0) title.alignment = WD_ALIGN_PARAGRAPH.CENTER # Split text into blocks blocks = input_text.split('\n\n') i = 0 while i < len(blocks): block = blocks[i].strip() if not block: i += 1 continue # Detect and process tables if '|' in block and block.count('|') > 2: # Check if it’s a markdown table lines = block.split(‘\n’) if any(‘—‘ in line or ‘===’ in line for line in lines): # This is a table table_data = parse_markdown_table(block) if table_data: # Create Word table table = doc.add_table(rows=len(table_data), cols=len(table_data[0])) table.style = ‘Light Grid Accent 1’ for row_idx, row_data in enumerate(table_data): for col_idx, cell_data in enumerate(row_data): cell = table.rows[row_idx].cells[col_idx] # Process cell content for formulas if ‘$’ in cell_data: process_text_with_math(cell_data, cell.paragraphs[0]) else: cell.text = cell_data # Bold header row if row_idx == 0: for paragraph in cell.paragraphs: for run in paragraph.runs: run.bold = True doc.add_paragraph() # Add space after table i += 1 continue # Detect headings if block.startswith(‘#’): level = len(block) – len(block.lstrip(‘#’)) heading_text = block.lstrip(‘#’).strip() doc.add_heading(heading_text, level=min(level, 3)) # Detect bullet points elif block.startswith(‘-‘) or block.startswith(‘*’) or block.startswith(‘•’): lines = block.split(‘\n’) for line in lines: clean_line = line.lstrip(‘-*•’).strip() if clean_line: p = doc.add_paragraph(style=’List Bullet’) process_text_with_math(clean_line, p) # Detect numbered lists elif re.match(r’^\d+\.’, block): lines = block.split(‘\n’) for line in lines: clean_line = re.sub(r’^\d+\.\s*’, ”, line).strip() if clean_line: p = doc.add_paragraph(style=’List Number’) process_text_with_math(clean_line, p) # Regular paragraph with potential formulas else: p = doc.add_paragraph() process_text_with_math(block, p) i += 1 # Save document doc.save(output_filename) return output_filename# Example with tables and mathematical formulas example_text = “””# Rapport Scientifique## IntroductionVoici un exemple de conversion avec des formules mathématiques et des tableaux.## Formules MathématiquesL’équation quadratique est donnée par $ax^2 + bx + c = 0$ et sa solution est:$$x = \\frac{-b \\pm \\sqrt{b^2 – 4ac}}{2a}$$La formule d’Einstein: $E = mc^2$Intégrale: $\\int_0^\\infty e^{-x} dx = 1$## Tableau de Données| Nom | Âge | Score | Formule | |—–|—–|——-|———| | Alice | 25 | 95 | $x^2 + 2x$ | | Bob | 30 | 87 | $\\sqrt{x}$ | | Charlie | 28 | 92 | $\\pi r^2$ |## Tableau de Résultats| Expérience | Valeur | Écart-type | P-value | |————|——–|————|———| | Test A | 12.5 | $\\pm 0.3$ | < 0.05 | | Test B | 15.8 | $\\pm 0.5$ | < 0.01 | | Test C | 10.2 | $\\pm 0.2$ | < 0.001 |## ConclusionLes résultats montrent que $\\alpha = 0.05$ est significatif.- Point important avec formule: $\\theta = 45°$ - Autre observation: $\\lambda = 550 nm$ - Résultat final: $\\sigma = 2.3$ """# Create the document output_file = convert_ai_text_to_word(example_text) print(f"Document créé: {output_file}")