import tkinter as tk
from tkinter import filedialog, simpledialog
import pandas as pd
import os

def normalize_modalities(val):
    if val.startswith("[") and val.endswith("]"):
        try:
            return ", ".join(eval(val))
        except:
            return val
    return val

def analyze_and_save():
    input_path = filedialog.askopenfilename(
        title="Select DICOM Discovery Output File",
        filetypes=[("Text Files", "*.txt")]
    )
    if not input_path:
        return

    output_path = filedialog.asksaveasfilename(
        title="Save Analysis Output As",
        defaultextension=".txt",
        filetypes=[("Text Files", "*.txt")]
    )
    if not output_path:
        return

    df = pd.read_csv(input_path, delimiter='|', dtype=str).fillna('')
    df['ModalitiesInStudy'] = df['ModalitiesInStudy'].apply(normalize_modalities)

    stats = {
        "Distinct StudyInstanceUIDs": df['StudyInstanceUID'].nunique(),
        "Distinct AccessionNumbers": df['AccessionNumber'].nunique(),
        "AccessionNumbers with 2 StudyInstanceUIDs": (df.groupby('AccessionNumber')['StudyInstanceUID'].nunique() == 2).sum(),
        "AccessionNumbers with 3 StudyInstanceUIDs": (df.groupby('AccessionNumber')['StudyInstanceUID'].nunique() == 3).sum(),
        "AccessionNumbers with >3 StudyInstanceUIDs": (df.groupby('AccessionNumber')['StudyInstanceUID'].nunique() > 3).sum(),
        "Missing AccessionNumbers": (df['AccessionNumber'] == '').sum(),
        "Missing StudyDescriptions": (df['StudyDescription'] == '').sum(),
        "*QA* in PatientID": df['PatientID'].str.contains('QA', case=False, na=False).sum(),
        "*QA* in StudyDescription": df['StudyDescription'].str.contains('QA', case=False, na=False).sum(),
        "*QA* in PatientName": df['PatientName'].str.contains('QA', case=False, na=False).sum(),
        "*TEST* in PatientID": df['PatientID'].str.contains('TEST', case=False, na=False).sum(),
        "*TEST* in StudyDescription": df['StudyDescription'].str.contains('TEST', case=False, na=False).sum(),
        "*TEST* in PatientName": df['PatientName'].str.contains('TEST', case=False, na=False).sum()
    }

    modality_counts = {
        "CR": 0, "DX": 0, "CT": 0, "MR": 0, "NM": 0, "US": 0,
        "XA": 0, "PR": 0, "SC": 0, "OT": 0, "Other": 0
    }

    for val in df['ModalitiesInStudy']:
        for modality in [m.strip().upper() for m in val.split(',') if m.strip()]:
            if modality in modality_counts:
                modality_counts[modality] += 1
            else:
                modality_counts["Other"] += 1

    with open(output_path, 'w') as f:
        f.write("DICOM Discovery Analysis Summary\n")
        f.write("="*40 + "\n\n")
        for key, value in stats.items():
            f.write(f"{key}\t{value}\n")
        f.write("\nModality Breakdown\n")
        f.write("-"*40 + "\n")
        for modality, count in modality_counts.items():
            f.write(f"{modality}\t{count}\n")

    tk.messagebox.showinfo("Analysis Complete", f"Analysis saved to:\n{output_path}")

# GUI
root = tk.Tk()
root.title("DICOM Discovery Analyzer")
root.geometry("400x150")

frame = tk.Frame(root, padx=10, pady=20)
frame.pack(expand=True)

tk.Button(frame, text="Select Input File and Analyze", command=analyze_and_save, width=30).pack(pady=10)

root.mainloop()