splitter-v1.py

import tkinter as tk
from tkinter import filedialog, messagebox, ttk
import csv
import os
from pathlib import Path

class CSVSplitterApp:
    def __init__(self, root):
        self.root = root
        self.root.title("CSV Splitter")
        self.root.geometry("600x500")
        self.root.configure(bg="#f0f0f0")
        
        self.style = ttk.Style()
        self.style.theme_use("clam")
        self.style.configure("TFrame", background="#f0f0f0")
        self.style.configure("TLabel", background="#f0f0f0", font=("Arial", 10))
        self.style.configure("TButton", font=("Arial", 10), padding=6)
        self.style.configure("Horizontal.TProgressbar", thickness=20)
        
        self.file_path = tk.StringVar()
        self.chunk_size = tk.IntVar(value=40)
        self.include_header = tk.BooleanVar(value=True)
        self.output_dir = tk.StringVar()
        
        self.create_widgets()
        
    def create_widgets(self):
        main_frame = ttk.Frame(self.root, padding="20")
        main_frame.grid(row=0, column=0, sticky=(tk.W, tk.E, tk.N, tk.S))
        
        title_label = ttk.Label(main_frame, text="CSV Splitter", font=("Arial", 16, "bold"))
        title_label.grid(row=0, column=0, columnspan=3, pady=(0, 20))
        
        ttk.Label(main_frame, text="Select CSV File:").grid(row=1, column=0, sticky=tk.W, pady=5)
        file_frame = ttk.Frame(main_frame)
        file_frame.grid(row=2, column=0, columnspan=3, sticky=(tk.W, tk.E), pady=5)
        
        ttk.Entry(file_frame, textvariable=self.file_path, width=50, state="readonly").grid(row=0, column=0, sticky=(tk.W, tk.E), padx=(0, 10))
        ttk.Button(file_frame, text="Browse", command=self.browse_file).grid(row=0, column=1)
        file_frame.columnconfigure(0, weight=1)
        
        settings_frame = ttk.LabelFrame(main_frame, text="Settings", padding="10")
        settings_frame.grid(row=3, column=0, columnspan=3, sticky=(tk.W, tk.E), pady=15)
        
        ttk.Label(settings_frame, text="Chunk Size:").grid(row=0, column=0, sticky=tk.W, pady=5)
        chunk_frame = ttk.Frame(settings_frame)
        chunk_frame.grid(row=0, column=1, sticky=tk.W, pady=5)
        ttk.Spinbox(chunk_frame, from_=1, to=1000, textvariable=self.chunk_size, width=10).grid(row=0, column=0)
        ttk.Label(chunk_frame, text="rows").grid(row=0, column=1, padx=(5, 0))
        
        ttk.Checkbutton(settings_frame, text="Include Header in Each Chunk", variable=self.include_header).grid(row=1, column=0, columnspan=2, sticky=tk.W, pady=5)
        
        ttk.Label(settings_frame, text="Output Directory:").grid(row=2, column=0, sticky=tk.W, pady=(10, 5))
        output_frame = ttk.Frame(settings_frame)
        output_frame.grid(row=3, column=0, columnspan=2, sticky=(tk.W, tk.E), pady=5)
        ttk.Entry(output_frame, textvariable=self.output_dir, width=50, state="readonly").grid(row=0, column=0, sticky=(tk.W, tk.E), padx=(0, 10))
        ttk.Button(output_frame, text="Browse", command=self.browse_output_dir).grid(row=0, column=1)
        output_frame.columnconfigure(0, weight=1)
        
        settings_frame.columnconfigure(1, weight=1)
        
        button_frame = ttk.Frame(main_frame)
        button_frame.grid(row=4, column=0, columnspan=3, pady=20)
        
        self.split_button = ttk.Button(button_frame, text="Split CSV", command=self.split_csv, state="disabled")
        self.split_button.grid(row=0, column=0, padx=5)
        ttk.Button(button_frame, text="Reset", command=self.reset_form).grid(row=0, column=1, padx=5)
        
        self.progress = ttk.Progressbar(main_frame, orient="horizontal", mode="determinate", style="Horizontal.TProgressbar")
        self.progress.grid(row=5, column=0, columnspan=3, sticky=(tk.W, tk.E), pady=10)
        
        self.status_label = ttk.Label(main_frame, text="", font=("Arial", 9))
        self.status_label.grid(row=6, column=0, columnspan=3, pady=5)
        
        self.root.columnconfigure(0, weight=1)
        self.root.rowconfigure(0, weight=1)
        main_frame.columnconfigure(0, weight=1)
        main_frame.rowconfigure(6, weight=1)
        
    def browse_file(self):
        filename = filedialog.askopenfilename(
            title="Select CSV file",
            filetypes=[("CSV files", "*.csv"), ("All files", "*.*")]
        )
        if filename:
            self.file_path.set(filename)
            if not self.output_dir.get():
                self.output_dir.set(os.path.dirname(filename))
            self.split_button.config(state="normal")
    
    def browse_output_dir(self):
        directory = filedialog.askdirectory(title="Select Output Directory")
        if directory:
            self.output_dir.set(directory)
            
    def split_csv(self):
        if not self.file_path.get():
            messagebox.showerror("Error", "Please select a CSV file")
            return
            
        if not self.output_dir.get():
            messagebox.showerror("Error", "Please select an output directory")
            return
            
        try:
            self.split_button.config(state="disabled")
            self.status_label.config(text="Processing...")
            self.root.update()
            
            self._split_csv_process()
            
            self.status_label.config(text="Splitting completed successfully!")
            messagebox.showinfo("Success", "CSV file has been split successfully!")
            
        except Exception as e:
            messagebox.showerror("Error", f"An error occurred: {str(e)}")
            self.status_label.config(text="Error occurred")
        finally:
            self.split_button.config(state="normal")
    
    def _split_csv_process(self):
        input_file = self.file_path.get()
        chunk_size = self.chunk_size.get()
        include_header = self.include_header.get()
        output_dir = self.output_dir.get()
        
        Path(output_dir).mkdir(parents=True, exist_ok=True)
        
        base_filename = os.path.splitext(os.path.basename(input_file))[0]
        
        with open(input_file, 'r', newline='', encoding='utf-8') as csvfile:
            reader = csv.reader(csvfile)
            
            header = None
            if include_header:
                try:
                    header = next(reader)
                except StopIteration:
                    messagebox.showwarning("Warning", "CSV file is empty")
                    return
            
            csvfile.seek(0)
            total_rows = sum(1 for row in csvfile) - (1 if include_header else 0)
            csvfile.seek(0)
            if include_header:
                next(csvfile)
                
            self.progress['maximum'] = total_rows
            processed_rows = 0
            chunk_num = 1
            writer = None
            output_file = None
            
            for i, row in enumerate(reader):
                if i % chunk_size == 0:
                    if output_file:
                        output_file.close()
                    
                    chunk_filename = f"{base_filename}_part_{chunk_num}.csv"
                    chunk_filepath = os.path.join(output_dir, chunk_filename)
                    output_file = open(chunk_filepath, 'w', newline='', encoding='utf-8')
                    writer = csv.writer(output_file)
                    
                    if include_header and header:
                        writer.writerow(header)
                    
                    chunk_num += 1
                
                writer.writerow(row)
                processed_rows += 1
                
                if processed_rows % 10 == 0 or processed_rows == total_rows:
                    self.progress['value'] = processed_rows
                    self.status_label.config(text=f"Processing... ({processed_rows}/{total_rows} rows)")
                    self.root.update_idletasks()
            
            if output_file:
                output_file.close()
                
            self.progress['value'] = total_rows
    
    def reset_form(self):
        self.file_path.set("")
        self.chunk_size.set(40)
        self.include_header.set(True)
        self.output_dir.set("")
        self.progress['value'] = 0
        self.status_label.config(text="")
        self.split_button.config(state="disabled")

if __name__ == "__main__":
    root = tk.Tk()
    app = CSVSplitterApp(root)
    root.mainloop()