research.py

import json
import csv
import time
import psutil
import os
from memory_profiler import memory_usage
import pandas as pd

# Constants
ITERATIONS = 10
DATASET_SIZES = ['small', 'medium', 'large']
DATA_DIR = 'Datasets'

def load_csv(file_path):
    df = pd.read_csv(file_path)
    return df

def load_json(file_path):
    with open(file_path, 'r') as f:
        raw = f.read()
        data = json.loads(raw)  # parse stringified JSON
        if isinstance(data, str):
            data = json.loads(data)  # if double-encoded

    if not isinstance(data, list):
        raise ValueError(f"Expected list of dicts in JSON, got {type(data)}")
    
    df = pd.DataFrame(data)
    return df

def process_data(df):
    # Perform data processing operation
    # Sort the 'Name' column in ascending order
    df.sort_values('Name', ascending=True, inplace=True)
    return df

def measure_execution_time(func, *args):
    start_time = time.time()
    result = func(*args)
    end_time = time.time()
    return end_time - start_time, result

def measure_memory_usage(func, *args):
    mem_usage = memory_usage((func, args), max_iterations=1)
    return max(mem_usage)

def measure_cpu_usage(func, *args):
    process = psutil.Process(os.getpid())
    cpu_percent_sum = 0
    cpu_percent_count = 0
    
    result = func(*args)
    
    while True:
        cpu_percent = process.cpu_percent(interval=0.1)
        cpu_percent_sum += cpu_percent
        cpu_percent_count += 1
        
        if cpu_percent == 0.0:
            break
    
    avg_cpu_percent = cpu_percent_sum / cpu_percent_count
    return avg_cpu_percent, result

def run_experiment(dataset_size):
    csv_file = f'{DATA_DIR}/{dataset_size}.csv'
    json_file = f'{DATA_DIR}/{dataset_size}.json'

    if not os.path.exists(csv_file):
        print(f"❌ CSV file not found: {csv_file}")
        return
    
    if not os.path.exists(json_file):
        print(f"❌ JSON file not found: {json_file}")
        return
    
    csv_times = []
    csv_memory = []
    csv_cpu = []
    
    json_times = []
    json_memory = []  
    json_cpu = []
    
    # print("-" * 50)
    for i in range(ITERATIONS):
        # print(f"▶️ Iteration {i+1}/{ITERATIONS} - {dataset_size.upper()}")

        # CSV
        csv_time, csv_df = measure_execution_time(load_csv, csv_file)
        csv_times.append(csv_time)
        
        csv_mem = measure_memory_usage(load_csv, csv_file)
        csv_memory.append(csv_mem)
        
        csv_cpu_usage, csv_processed_df = measure_cpu_usage(process_data, csv_df)
        csv_cpu.append(csv_cpu_usage)
        
        # JSON  
        json_time, json_df = measure_execution_time(load_json, json_file)
        json_times.append(json_time)
        
        json_mem = measure_memory_usage(load_json, json_file)
        json_memory.append(json_mem)

        json_cpu_usage, json_processed_df = measure_cpu_usage(process_data, json_df)  
        json_cpu.append(json_cpu_usage)

    # Calculate average metrics
    avg_csv_time = sum(csv_times) / ITERATIONS
    avg_csv_memory = sum(csv_memory) / ITERATIONS 
    avg_csv_cpu = sum(csv_cpu) / ITERATIONS

    avg_json_time = sum(json_times) / ITERATIONS
    avg_json_memory = sum(json_memory) / ITERATIONS
    avg_json_cpu = sum(json_cpu) / ITERATIONS

    # Print results  
    # print(f"Dataset Size: {dataset_size}")
    print(f"📊 Results for Dataset Size: {dataset_size.upper()}")
    print(f"CSV - Avg Execution Time: {avg_csv_time:.3f} seconds")
    print(f"CSV - Avg Memory Usage: {avg_csv_memory:.2f} MB")  
    print(f"CSV - Avg CPU Usage: {avg_csv_cpu:.2f}%")
    print()
    print(f"JSON - Avg Execution Time: {avg_json_time:.3f} seconds")
    print(f"JSON - Avg Memory Usage: {avg_json_memory:.2f} MB")
    print(f"JSON - Avg CPU Usage: {avg_json_cpu:.2f}%") 
    print()
    print(f"✅ Finished {dataset_size.upper()} experiment in {ITERATIONS} iteration(s)")
    print("-" * 50)
    # print()

if __name__ == '__main__':
    for size in DATASET_SIZES:
        run_experiment(size)