from pptx import Presentation def parse_slides(file_path): prs = Presentation(file_path) slides = [] for idx, slide in enumerate(prs.slides): text_list = [] for shape in slide.shapes: if hasattr(shape, "text"): txt = shape.text.strip() if txt: text_list.append(txt) slides.append({ "index": idx + 1, "text": "\n".join(text_list) }) return slides