import os
import pandas as pd

# Path settings
input_file = r'G:\Users\Chipistil\Desktop\High-throughput-T2NP-Memristor-based-Dual-clock-edge-sampling-TRNG\Data Conversion\tdelay+tr.txt'
output_file = r'G:\Users\Chipistil\Desktop\High-throughput-T2NP-Memristor-based-Dual-clock-edge-sampling-TRNG\Data Conversion\td.xlsx'

data = []
with open(input_file, 'r') as f:
    for line in f:
        line = line.strip()
        if line:
            cols = line.split()
            if len(cols) >= 3:
                col1 = round(float(cols[0]), 4)
                col2 = round(float(cols[1]), 4)
                col3 = round(float(cols[2]), 4)
                data.append((col1, col2, col3))

td_list = []
n = len(data)
i = 0

def find_dg_cluster(start_idx):
    cluster_start = None
    consecutive_230 = 0
    in_cluster = False
    sep_count = 0
    
    for j in range(start_idx, n):
        current = data[j][2]
        
        if current == 2.30:
            if not in_cluster:
                if cluster_start is None:
                    cluster_start = j
                consecutive_230 += 1
                if consecutive_230 >= 6:
                    in_cluster = True
                    sep_count = 0
            else:
                consecutive_230 += 1
                sep_count = 0
            
        elif current == 0.00 and in_cluster:
            sep_count += 1
            if sep_count >= 6:
                return (cluster_start, j - sep_count), j + 1
        
        else:
            if in_cluster:
                return (cluster_start, j-1), j
            else:
                cluster_start = None
                consecutive_230 = 0
                sep_count = 0
    
    if in_cluster:
        return (cluster_start, n-1), n
    return None, n

def locate_key_points(start, end):
    dg1_start = next((j for j in range(start, end+1) if data[j][2] > 0.1), None)
    dg1_end = next((j for j in range(start, end+1) if data[j][2] == 2.30), None)
    
    if not dg1_start or not dg1_end:
        return None, None
    
    dg1_idx = (dg1_start + dg1_end) / 2
    
    col2_values = [data[j][1] for j in range(start, end+1)]
    col2_indices = list(range(start, end+1))
    
    rise_start = None
    rise_end = None
    
    for idx in range(len(col2_values)):
        if col2_values[idx] > 0.4 and rise_start is None:
            if idx + 10 <= len(col2_values):
                next_10_values = col2_values[idx:idx+10]
                count_above_2 = sum(1 for val in next_10_values if val > 2)
                if count_above_2 >= 5:
                    rise_start = idx
        if rise_start is not None and col2_values[idx] > 2.9:
            rise_end = idx
            break
    
    if rise_start is not None and rise_end is None:
        max_val = max(col2_values[rise_start:])
        for idx in range(rise_start, len(col2_values)):
            if col2_values[idx] == max_val:
                rise_end = idx
                break
    
    if rise_start is not None and rise_end is not None:
        n_rise = rise_end - rise_start + 1
        if n_rise % 2 == 1:
            t2_idx = col2_indices[rise_start + n_rise // 2]
        else:
            mid1 = col2_indices[rise_start + n_rise // 2 - 1]
            mid2 = col2_indices[rise_start + n_rise // 2]
            t2_idx = (mid1 + mid2) / 2
    else:
        t2_idx = None
    
    return dg1_idx, t2_idx

while i < n:
    dg_range, new_i = find_dg_cluster(i)
    
    if dg_range:
        start, end = dg_range
        
        dg1_idx, t2_idx = locate_key_points(start, end)
        
        if all([dg1_idx, t2_idx]):
            dg1_time = (data[int(dg1_idx)][0] + data[int(dg1_idx) + 1][0]) / 2
            
            if isinstance(t2_idx, float):
                t2_time = (data[int(t2_idx)][0] + data[int(t2_idx) + 1][0]) / 2
            else:
                t2_time = data[t2_idx][0]
            
            td = round(t2_time - dg1_time, 4)
            td_list.append((td, round(dg1_time, 4), round(t2_time, 4)))
        
        i = new_i
    else:
        i += 1

if td_list:
    df = pd.DataFrame(td_list, columns=['td', 't1', 't2'])
    df.to_excel(output_file, index=False)
    print(f"Processing completed, found {len(td_list)} valid td values, written to Excel file")
else:
    print("No valid td values found")