import tkinter as tk from tkinter import filedialog, messagebox, ttk, Toplevel import os import fitz # PyMuPDF import re class MmpiExtractorApp: def __init__(self, root): self.root = root self.root.title("MMPI-2 데이터 추출기 v2.3 (중복 해결)") self.root.geometry("700x500") self.root.configure(bg="#f0f2f5") # --- 스타일 설정 --- self.font_main = ("Segoe UI", 10) self.font_status = ("Segoe UI", 9) self.bg_color = "#f0f2f5" self.btn_color = "#ffffff" self.btn_accent_color = "#e0e8ff" self.file_paths = [] self.results = {} main_frame = tk.Frame(self.root, padx=15, pady=15, bg=self.bg_color) main_frame.pack(fill=tk.BOTH, expand=True) # --- UI 요소들 (이전과 동일) --- btn_frame = tk.Frame(main_frame, bg=self.bg_color) btn_frame.pack(fill=tk.X, pady=(0, 10)) self.add_file_button = tk.Button(btn_frame, text="PDF 파일 선택", command=self.add_files, font=self.font_main, bg=self.btn_color, relief="flat", padx=10) self.add_file_button.pack(side=tk.LEFT) self.clear_button = tk.Button(btn_frame, text="목록 지우기", command=self.clear_list, font=self.font_main, bg=self.btn_color, relief="flat", padx=10) self.clear_button.pack(side=tk.LEFT, padx=(10, 0)) list_frame = tk.Frame(main_frame, bg=self.bg_color) list_frame.pack(fill=tk.BOTH, expand=True) self.listbox = tk.Listbox(list_frame, font=self.font_main, relief="solid", bd=1, bg="#ffffff") self.listbox.pack(side=tk.LEFT, fill=tk.BOTH, expand=True) scrollbar = tk.Scrollbar(list_frame, orient="vertical", command=self.listbox.yview, relief="flat") scrollbar.pack(side=tk.RIGHT, fill=tk.Y) self.listbox.config(yscrollcommand=scrollbar.set) status_frame = tk.Frame(main_frame, bg=self.bg_color) status_frame.pack(fill=tk.X, pady=(10, 0)) self.status_label = tk.Label(status_frame, text="MMPI-2 PDF 보고서 파일을 선택하세요.", font=self.font_status, anchor="w", bg=self.bg_color, fg="#555555") self.status_label.pack(fill=tk.X) self.progress = ttk.Progressbar(status_frame, orient="horizontal", length=100, mode="determinate") self.progress.pack(fill=tk.X, pady=(5, 0)) self.extract_button = tk.Button(main_frame, text="데이터 추출 시작", command=self.start_extraction, font=(self.font_main[0], 11, 'bold'), height=2, bg=self.btn_accent_color, relief="flat") self.extract_button.pack(fill=tk.X, pady=(15, 0)) def add_files(self): new_files = filedialog.askopenfilenames(title="MMPI-2 보고서 PDF 파일을 선택하세요", filetypes=[("PDF files", "*.pdf")]) if new_files: for f in new_files: if f not in self.file_paths: self.file_paths.append(f) self.listbox.insert(tk.END, os.path.basename(f)) self.update_status(f"{len(self.file_paths)}개의 파일이 선택되었습니다.") def clear_list(self): self.file_paths.clear() self.listbox.delete(0, tk.END) self.results = {} self.update_status("파일을 선택하세요.") def update_status(self, message): self.status_label.config(text=message) self.root.update_idletasks() def extract_name(self, page): try: text = page.get_text("text") match = re.search(r"(?:이름|성명)\s*:\s*(\S+)", text) if match: return match.group(1) blocks = page.get_text("blocks") if blocks: return blocks[0][4].strip().split('\n')[0] except Exception: pass return "이름모름" def reformat_numeric_blocks(self, text): lines = text.split('\n') if not lines: return "" new_lines = [lines[0]] for i in range(1, len(lines)): line, stripped_line = lines[i], lines[i].strip() is_numeric = False if stripped_line and ' ' not in stripped_line: try: float(stripped_line) is_numeric = True except ValueError: is_numeric = False if is_numeric and new_lines and new_lines[-1].strip(): new_lines[-1] += f" {stripped_line}" else: new_lines.append(line) return "\n".join(new_lines) def start_extraction(self): if not self.file_paths: return messagebox.showwarning("경고", "추출할 PDF 파일을 먼저 선택해주세요.") total_files = len(self.file_paths) self.progress["maximum"] = total_files self.progress["value"] = 0 self.results = {} pages_to_extract = [1, 2, 3, 4] try: for i, file_path in enumerate(self.file_paths): filename = os.path.basename(file_path) self.update_status(f"({i+1}/{total_files}) 처리 중: {filename}") doc = fitz.open(file_path) if doc.page_count < 5: continue name = self.extract_name(doc.load_page(0)) if name in self.results: name = f"{name}_{i+1}" # 1. 모든 페이지 텍스트 합치기 combined_raw_text = "" for page_num in pages_to_extract: combined_raw_text += doc.load_page(page_num).get_text("text") + "\n" # 2. 합쳐진 텍스트에서 중복 라인 제거 lines = combined_raw_text.split('\n') unique_lines = [] seen_lines = set() for line in lines: stripped_line = line.strip() # 비어있지 않고, 처음 보는 라인일 경우에만 추가 if stripped_line and stripped_line not in seen_lines: seen_lines.add(stripped_line) unique_lines.append(line) # 비어있는 라인은 그대로 추가 (문단 구분 등 유지) elif not stripped_line: unique_lines.append(line) deduplicated_text = "\n".join(unique_lines) # 3. 중복 제거된 텍스트를 재포맷 formatted_text = self.reformat_numeric_blocks(deduplicated_text) self.results[name] = formatted_text doc.close() self.progress["value"] = i + 1 self.root.update_idletasks() if not self.results: return messagebox.showinfo("완료", "처리할 파일이 없습니다.") self.show_results_window() self.update_status("추출 완료. 결과 창을 확인하세요.") except Exception as e: messagebox.showerror("오류 발생", f"작업 중 오류가 발생했습니다:\n{e}") finally: self.progress["value"] = 0 def show_results_window(self): result_window = Toplevel(self.root) result_window.title("추출 결과") result_window.geometry("800x600") result_window.configure(bg=self.bg_color) def copy_to_clipboard(content): self.root.clipboard_clear() self.root.clipboard_append(content) messagebox.showinfo("복사 완료", "클립보드에 복사되었습니다.", parent=result_window) def copy_all(): copy_to_clipboard("\n\n".join([f"---" + name + "---" + content for name, content in self.results.items()])) def copy_individual(name): copy_to_clipboard(self.results[name]) def copy_for_ai(): copy_to_clipboard("\n\n".join(self.results.values())) top_button_frame = tk.Frame(result_window, pady=5, bg=self.bg_color) top_button_frame.pack(fill=tk.X, padx=10, pady=(10,0)) tk.Button(top_button_frame, text="[ 전체 복사 ]", command=copy_all, font=self.font_main, bg=self.btn_accent_color, relief="flat", padx=10).pack(side=tk.LEFT) tk.Button(top_button_frame, text="[ AI에게 맡기기 (이름 제외) ]", command=copy_for_ai, font=self.font_main, bg=self.btn_accent_color, relief="flat", padx=10).pack(side=tk.LEFT, padx=5) individual_button_frame = tk.Frame(result_window, pady=2, bg=self.bg_color) individual_button_frame.pack(fill=tk.X, padx=10) max_cols = 7 for i, name in enumerate(self.results.keys()): btn = tk.Button(individual_button_frame, text=f"{name}", command=lambda n=name: copy_individual(n), font=self.font_main, bg=self.btn_color, relief="flat", padx=8, pady=2) btn.grid(row=i // max_cols, column=i % max_cols, padx=2, pady=3, sticky="w") ttk.Separator(result_window, orient='horizontal').pack(fill='x', padx=10, pady=10) text_area_frame = tk.Frame(result_window, bg=self.bg_color) text_area_frame.pack(expand=True, fill="both", padx=10, pady=(0, 10)) text_area = tk.Text(text_area_frame, wrap="word", font=("맑은 고딕", 10), relief="solid", bd=1, spacing1=2, spacing3=2) text_scroll = tk.Scrollbar(text_area_frame, command=text_area.yview, relief="flat") text_area.config(yscrollcommand=text_scroll.set) text_scroll.pack(side=tk.RIGHT, fill=tk.Y) text_area.pack(expand=True, fill="both") for name, content in self.results.items(): text_area.insert(tk.END, f"============== {name} ===============\n", "header") text_area.insert(tk.END, content + "\n\n") text_area.tag_config("header", font=("Segoe UI", 12, "bold"), spacing3=10) text_area.config(state="disabled") if __name__ == "__main__": root = tk.Tk() app = MmpiExtractorApp(root) root.mainloop()