currently we are taking week number from the image name . is it possible to extract from…
- User
currently we are taking week number from the image name . is it possible to extract from the image. for example in one unmatched line : Unmatched line in Mahalakshmi_Week01_2025.PNG: Week 1 2025, commencing Saturday 28/12/2021: this week) you can see 'Week 1 2025'. Could you modify the program to save this week number into excel file instead of week number from image name import os import re import threading from tkinter import ( Tk, Label, Button, filedialog, messagebox, Text, Scrollbar, Entry, StringVar, END, DISABLED, NORMAL, Frame, LabelFrame, Menu ) from tkinter.ttk import Progressbar, Style from PIL import Image, ImageEnhance, ImageFilter import pytesseract import pandas as pd
---------- CONFIG ----------
pytesseract.pytesseract.tesseract_cmd = r"C:\Users\36083\timesheet_env\Tesseract-OCR\tesseract.exe" custom_config = r'--oem 3 --psm 6 -c tessedit_char_whitelist=0123456789ABCDEFGHIJKLMNOPQRSTUVWXYZ '
line_pattern = re.compile( r'(.+?)\s+([0-9O]{1,2})\s+([0-9O]{1,2})\s+([0-9O]{1,2})\s+([0-9O]{1,2})\s+([0-9O]{1,2})\s+([0-9O]{1,2})\s+([0-9O]{1,2})\s+([0-9O]{1,2})\b' )
def preprocess_image(img): img = img.convert("L") img = img.resize((img.width * 2, img.height * 2), Image.LANCZOS) img = ImageEnhance.Contrast(img).enhance(2) img = img.filter(ImageFilter.SHARPEN) return img
def classify_project(raw): compact = re.sub(r'\s+', '', raw.strip(), flags=re.I) if re.search(r'annualholiday', compact, re.I): return "Annual Holiday" if re.search(r'absence', compact, re.I): return "ODEC Absence" return "Project"
def parse_weeks(week_str): """Parse a string like '12,13,15-17' into a set of week numbers.""" weeks = set() for part in week_str.split(','): part = part.strip() if '-' in part: start, end = part.split('-') weeks.update(range(int(start), int(end)+1)) elif part.isdigit(): weeks.add(int(part)) return weeks
def process_timesheets(image_folder, output_excel, log_callback, progress_callback, week_filter=None): data = [] all_files = [] for root, dirs, files in os.walk(image_folder): for filename in files: if filename.lower().endswith(('.png', '.jpg', '.jpeg')): all_files.append((root, filename)) total_files = len(all_files) processed_files = 0
for root, filename in all_files: image_path = os.path.join(root, filename) # Parse name & week from filename name, week = "Unknown", "Unknown" file_match = re.match(r"(.+?)_Week(\d+)_\d+\.(png|jpg|jpeg)$", filename, re.IGNORECASE) if file_match: name, week = file_match.group(1), file_match.group(2) else: name = filename.rsplit('.', 1)[0] # Filter by week if needed if week_filter and week != "Unknown": if int(week) not in week_filter: processed_files += 1 progress_callback(processed_files, total_files) continue try: img = Image.open(image_path) img = preprocess_image(img) text = pytesseract.image_to_string(img, config=custom_config).strip() text = text.encode('ascii', errors='ignore').decode() log_callback(f"OCR output for {filename}:\n{text}\n") if not text: log_callback(f"No text found in image: {filename}\n") processed_files += 1 progress_callback(processed_files, total_files) continue except Exception as e: log_callback(f"Error reading image {filename}: {e}\n") processed_files += 1 progress_callback(processed_files, total_files) continue matched = False for line in text.splitlines(): match = line_pattern.search(line) if not match: if line.strip(): log_callback(f"Unmatched line in {filename}: {line.strip()}\n") continue original_project = match.group(1).strip() if original_project.lower().startswith("total"): continue # Clean digits days = [match.group(i).replace('O', '0') for i in range(2, 9)] total = match.group(9).replace('O', '0') # Validate try: days = [int(d) for d in days] total = int(total) except ValueError: log_callback(f"Skipping malformed line in {filename}: {line}\n") continue project = classify_project(original_project) data.append({ "Name": name, "Week": week, "Project": project, "Sat": days[0], "Sun": days[1], "Mon": days[2], "Tue": days[3], "Wed": days[4], "Thu": days[5], "Fri": days[6], "Total": total }) matched = True if not matched: log_callback(f"No valid timesheet lines found in {filename}\n") processed_files += 1 progress_callback(processed_files, total_files) if data: df = pd.DataFrame(data) # Separate "Project" and non-"Project" rows project_rows = df[df["Project"] == "Project"] other_rows = df[df["Project"] != "Project"] # Group and sum "Project" rows by Name and Week if not project_rows.empty: summed = ( project_rows .groupby(["Name", "Week"], as_index=False)[["Sat", "Sun", "Mon", "Tue", "Wed", "Thu", "Fri"]] .sum() ) # Recompute Total as sum of Sat-Fri summed["Total"] = summed[["Sat", "Sun", "Mon", "Tue", "Wed", "Thu", "Fri"]].sum(axis=1) summed["Project"] = "Project" # Reorder columns to match original summed = summed[["Name", "Week", "Project", "Sat", "Sun", "Mon", "Tue", "Wed", "Thu", "Fri", "Total"]] else: summed = pd.DataFrame(columns=["Name", "Week", "Project", "Sat", "Sun", "Mon", "Tue", "Wed", "Thu", "Fri", "Total"]) # Combine summed "Project" rows with other rows final_df = pd.concat([summed, other_rows], ignore_index=True) # Optional: sort for neatness final_df = final_df.sort_values(["Name", "Week", "Project"]) final_df.to_excel(output_excel, index=False) return True, output_excel else: return False, None---------- GUI ----------
class TimesheetGUI: def init(self, master): self.master = master master.title("Timesheet Extractor") master.geometry("800x600") master.minsize(700, 500)
# Set a modern theme style = Style() style.theme_use('clam') # Menu bar menubar = Menu(master) helpmenu = Menu(menubar, tearoff=0) helpmenu.add_command(label="About", command=self.show_about) helpmenu.add_command(label="Help", command=self.show_help) menubar.add_cascade(label="Help", menu=helpmenu) master.config(menu=menubar) # Top Frame for title top_frame = Frame(master, bg="#e6f2ff") top_frame.pack(fill="x", padx=10, pady=5) Label(top_frame, text="Timesheet Extractor", font=("Arial", 18, "bold"), bg="#e6f2ff").pack(side="left", padx=10) # Input Frame input_frame = LabelFrame(master, text="Input", padx=10, pady=10) input_frame.pack(fill="x", padx=10, pady=5) Label(input_frame, text="Select the input folder containing timesheet images:").pack(anchor="w") self.select_button = Button(input_frame, text="Browse Folder", command=self.browse_folder) self.select_button.pack(anchor="w", pady=2) self.folder_label = Label(input_frame, text="", fg="blue") self.folder_label.pack(anchor="w", pady=2) Label(input_frame, text="Week Filter:").pack(anchor="w") self.week_var = StringVar() self.week_entry = Entry(input_frame, textvariable=self.week_var, width=40) self.week_entry.pack(anchor="w", pady=2) self.week_entry.insert(0, "Optional: Enter week(s) e.g. 12,13,15-17") self.submit_button = Button(input_frame, text="Submit", command=self.run_extraction, state="disabled") self.submit_button.pack(anchor="w", pady=5) # Progress Frame progress_frame = Frame(master) progress_frame.pack(fill="x", padx=10, pady=5) self.progress = Progressbar(progress_frame, orient="horizontal", length=400, mode="determinate") self.progress.pack(side="left", padx=5, pady=5) self.progress_label = Label(progress_frame, text="0%") self.progress_label.pack(side="left", padx=5) # Log Frame log_frame = LabelFrame(master, text="Log", padx=10, pady=10) log_frame.pack(fill="both", expand=True, padx=10, pady=5) self.log_text = Text(log_frame, height=12, width=90, state=DISABLED, wrap="word") self.log_text.pack(side="left", fill="both", expand=True) self.scrollbar = Scrollbar(log_frame, command=self.log_text.yview) self.log_text['yscrollcommand'] = self.scrollbar.set self.scrollbar.pack(side="right", fill="y") self.clear_log_button = Button(log_frame, text="Clear Log", command=self.clear_log) self.clear_log_button.pack(anchor="ne", padx=5, pady=2) # Status Bar self.status = Label(master, text="Ready", bd=1, relief="sunken", anchor="w", bg="#f0f0f0") self.status.pack(side="bottom", fill="x") self.input_folder = "" self.output_excel = "" def browse_folder(self): folder_selected = filedialog.askdirectory() if folder_selected: self.input_folder = folder_selected self.folder_label.config(text=self.input_folder) self.output_excel = os.path.join( self.input_folder, "Extracted_Timesheet.xlsx" ) self.submit_button.config(state="normal") self.status.config(text="Folder selected. Ready to process.") def log(self, message): self.log_text.config(state=NORMAL) self.log_text.insert(END, message) self.log_text.see(END) self.log_text.config(state=DISABLED) def clear_log(self): self.log_text.config(state=NORMAL) self.log_text.delete(1.0, END) self.log_text.config(state=DISABLED) def update_progress(self, value, maximum): self.progress["maximum"] = maximum self.progress["value"] = value percent = int((value / maximum) * 100) if maximum else 0 self.progress_label.config(text=f"{percent}%") self.status.config(text=f"Processing... {value} of {maximum} files") self.master.update_idletasks() def run_extraction(self): if not self.input_folder or not os.path.isdir(self.input_folder): messagebox.showerror("Error", "Please select a valid input folder.") return self.submit_button.config(state="disabled") self.status.config(text="Processing, please wait...") self.progress["value"] = 0 self.progress_label.config(text="0%") self.clear_log() threading.Thread(target=self._run_extraction_thread).start() def _run_extraction_thread(self): week_filter = None week_str = self.week_var.get().strip() if week_str and not week_str.lower().startswith("optional"): try: week_filter = parse_weeks(week_str) except Exception as e: self.log(f"Invalid week filter: {e}\n") week_filter = None success, output_path = process_timesheets( self.input_folder, self.output_excel, self.log, self.update_progress, week_filter ) if success: self.status.config(text="Done! Data written to Excel.") messagebox.showinfo("Done", f"Timesheet data successfully written to:\n{output_path}") else: self.status.config(text="No data extracted.") messagebox.showwarning("No Data", "No timesheet data extracted from any image.") self.submit_button.config(state="normal") self.progress["value"] = 0 self.progress_label.config(text="0%") def show_about(self): messagebox.showinfo("About", "Timesheet Extractor\nVersion 1.0\n\nDeveloped by Your Team.") def show_help(self): messagebox.showinfo( "Help", "1. Click 'Browse Folder' to select the folder with timesheet images.\n" "2. (Optional) Enter week numbers to filter (e.g., 12,13,15-17).\n" "3. Click 'Submit' to start extraction.\n" "4. Progress and logs will be shown below.\n" "5. The result Excel file will be saved in the selected folder." )if name == "main": root = Tk() gui = TimesheetGUI(root) root.mainloop()
- Assistant
Certainly! Below is the modified version of your program. The key changes are in the
process_timesheetsfunction, where we now extract the week number from the OCR text instead of the image filename.import os import re import threading from tkinter import ( Tk, Label, Button, filedialog, messagebox, Text, Scrollbar, Entry, StringVar, END, DISABLED, NORMAL, Frame, LabelFrame, Menu ) from tkinter.ttk import Progressbar, Style from PIL import Image, ImageEnhance, ImageFilter import pytesseract import pandas as pd # ---------- CONFIG ---------- pytesseract.pytesseract.tesseract_cmd = r"C:\Users\36083\timesheet_env\Tesseract-OCR\tesseract.exe" custom_config = r'--oem 3 --psm 6 -c tessedit_char_whitelist=0123456789ABCDEFGHIJKLMNOPQRSTUVWXYZ ' line_pattern = re.compile( r'(.+?)\s+([0-9O]{1,2})\s+([0-9O]{1,2})\s+([0-9O]{1,2})\s+([0-9O]{1,2})\s+([0-9O]{1,2})\s+([0-9O]{1,2})\s+([0-9O]{1,2})\s+([0-9O]{1,2})\b' ) week_pattern = re.compile(r'Week (\d+) \d{4}') def preprocess_image(img): img = img.convert("L") img = img.resize((img.width * 2, img.height * 2), Image.LANCZOS) img = ImageEnhance.Contrast(img).enhance(2) img = img.filter(ImageFilter.SHARPEN) return img def classify_project(raw): compact = re.sub(r'\s+', '', raw.strip(), flags=re.I) if re.search(r'annualholiday', compact, re.I): return "Annual Holiday" if re.search(r'absence', compact, re.I): return "ODEC Absence" return "Project" def parse_weeks(week_str): """Parse a string like '12,13,15-17' into a set of week numbers.""" weeks = set() for part in week_str.split(','): part = part.strip() if '-' in part: start, end = part.split('-') weeks.update(range(int(start), int(end)+1)) elif part.isdigit(): weeks.add(int(part)) return weeks def process_timesheets(image_folder, output_excel, log_callback, progress_callback, week_filter=None): data = [] all_files = [] for root, dirs, files in os.walk(image_folder): for filename in files: if filename.lower().endswith(('.png', '.jpg', '.jpeg')): all_files.append((root, filename)) total_files = len(all_files) processed_files = 0 for root, filename in all_files: image_path = os.path.join(root, filename) # Parse name from filename name = "Unknown" file_match = re.match(r"(.+?)_Week\d+_\d+\.(png|jpg|jpeg)$", filename, re.IGNORECASE) if file_match: name = file_match.group(1) else: name = filename.rsplit('.', 1)[0] try: img = Image.open(image_path) img = preprocess_image(img) text = pytesseract.image_to_string(img, config=custom_config).strip() text = text.encode('ascii', errors='ignore').decode() log_callback(f"OCR output for {filename}:\n{text}\n") if not text: log_callback(f"No text found in image: {filename}\n") processed_files += 1 progress_callback(processed_files, total_files) continue except Exception as e: log_callback(f"Error reading image {filename}: {e}\n") processed_files += 1 progress_callback(processed_files, total_files) continue # Extract week number from OCR text week_match = week_pattern.search(text) week = "Unknown" if week_match: week = week_match.group(1) else: log_callback(f"Week number not found in OCR text for {filename}\n") processed_files += 1 progress_callback(processed_files, total_files) continue # Filter by week if needed if week_filter and week != "Unknown": if int(week) not in week_filter: processed_files += 1 progress_callback(processed_files, total_files) continue matched = False for line in text.splitlines(): match = line_pattern.search(line) if not match: if line.strip(): log_callback(f"Unmatched line in {filename}: {line.strip()}\n") continue original_project = match.group(1).strip() if original_project.lower().startswith("total"): continue # Clean digits days = [match.group(i).replace('O', '0') for i in range(2, 9)] total = match.group(9).replace('O', '0') # Validate try: days = [int(d) for d in days] total = int(total) except ValueError: log_callback(f"Skipping malformed line in {filename}: {line}\n") continue project = classify_project(original_project) data.append({ "Name": name, "Week": week, "Project": project, "Sat": days[0], "Sun": days[1], "Mon": days[2], "Tue": days[3], "Wed": days[4], "Thu": days[5], "Fri": days[6], "Total": total }) matched = True if not matched: log_callback(f"No valid timesheet lines found in {filename}\n") processed_files += 1 progress_callback(processed_files, total_files) if data: df = pd.DataFrame(data) # Separate "Project" and non-"Project" rows project_rows = df[df["Project"] == "Project"] other_rows = df[df["Project"] != "Project"] # Group and sum "Project" rows by Name and Week if not project_rows.empty: summed = ( project_rows .groupby(["Name", "Week"], as_index=False)[["Sat", "Sun", "Mon", "Tue", "Wed", "Thu", "Fri"]] .sum() ) # Recompute Total as sum of Sat-Fri summed["Total"] = summed[["Sat", "Sun", "Mon", "Tue", "Wed", "Thu", "Fri"]].sum(axis=1) summed["Project"] = "Project" # Reorder columns to match original summed = summed[["Name", "Week", "Project", "Sat", "Sun", "Mon", "Tue", "Wed", "Thu", "Fri", "Total"]] else: summed = pd.DataFrame(columns=["Name", "Week", "Project", "Sat", "Sun", "Mon", "Tue", "Wed", "Thu", "Fri", "Total"]) # Combine summed "Project" rows with other rows final_df = pd.concat([summed, other_rows], ignore_index=True) # Optional: sort for neatness final_df = final_df.sort_values(["Name", "Week", "Project"]) final_df.to_excel(output_excel, index=False) return True, output_excel else: return False, None # ---------- GUI ---------- class TimesheetGUI: def __init__(self, master): self.master = master master.title("Timesheet Extractor") master.geometry("800x600") master.minsize(700, 500) # Set a modern theme style = Style() style.theme_use('clam') # Menu bar menubar = Menu(master) helpmenu = Menu(menubar, tearoff=0) helpmenu.add_command(label="About", command=self.show_about) helpmenu.add_command(label="Help", command=self.show_help) menubar.add_cascade(label="Help", menu=helpmenu) master.config(menu=menubar) # Top Frame for title top_frame = Frame(master, bg="#e6f2ff") top_frame.pack(fill="x", padx=10, pady=5) Label(top_frame, text="Timesheet Extractor", font=("Arial", 18, "bold"), bg="#e6f2ff").pack(side="left", padx=10) # Input Frame input_frame = LabelFrame(master, text="Input", padx=10, pady=10) input_frame.pack(fill="x", padx=10, pady=5) Label(input_frame, text="Select the input folder containing timesheet images:").pack(anchor="w") self.select_button = Button(input_frame, text="Browse Folder", command=self.browse_folder) self.select_button.pack(anchor="w", pady=2) self.folder_label = Label(input_frame, text="", fg="blue") self.folder_label.pack(anchor="w", pady=2) Label(input_frame, text="Week Filter:").pack(anchor="w") self.week_var = StringVar() self.week_entry = Entry(input_frame, textvariable=self.week_var, width=40) self.week_entry.pack(anchor="w", pady=2) self.week_entry.insert(0, "Optional: Enter week(s) e.g. 12,13,15-17") self.submit_button = Button(input_frame, text="Submit", command=self.run_extraction, state="disabled") self.submit_button.pack(anchor="w", pady=5) # Progress Frame progress_frame = Frame(master) progress_frame.pack(fill="x", padx=10, pady=5) self.progress = Progressbar(progress_frame, orient="horizontal", length=400, mode="determinate") self.progress.pack(side="left", padx=5, pady=5) self.progress_label = Label(progress_frame, text="0%") self.progress_label.pack(side="left", padx=5) # Log Frame log_frame = LabelFrame(master, text="Log", padx=10, pady=10) log_frame.pack(fill="both", expand=True, padx=10, pady=5) self.log_text = Text(log_frame, height=12, width=90, state=DISABLED, wrap="word") self.log_text.pack(side="left", fill="both", expand=True) self.scrollbar = Scrollbar(log_frame, command=self.log_text.yview) self.log_text['yscrollcommand'] = self.scrollbar.set self.scrollbar.pack(side="right", fill="y") self.clear_log_button = Button(log_frame, text="Clear Log", command=self.clear_log) self.clear_log_button.pack(anchor="ne", padx=5, pady=2) # Status Bar self.status = Label(master, text="Ready", bd=1, relief="sunken", anchor="w", bg="#f0f0f0") self.status.pack(side="bottom", fill="x") self.input_folder = "" self.output_excel = "" def browse_folder(self): folder_selected = filedialog.askdirectory() if folder_selected: self.input_folder = folder_selected self.folder_label.config(text=self.input_folder) self.output_excel = os.path.join( self.input_folder, "Extracted_Timesheet.xlsx" ) self.submit_button.config(state="normal") self.status.config(text="Folder selected. Ready to process.") def log(self, message): self.log_text.config(state=NORMAL) self.log_text.insert(END, message) self.log_text.see(END) self.log_text.config(state=DISABLED) def clear_log(self): self.log_text.config(state=NORMAL) self.log_text.delete(1.0, END) self.log_text.config(state=DISABLED) def update_progress(self, value, maximum): self.progress["maximum"] = maximum self.progress["value"] = value percent = int((value / maximum) * 100) if maximum else 0 self.progress_label.config(text=f"{percent}%") self.status.config(text=f"Processing... {value} of {maximum} files") self.master.update_idletasks() def run_extraction(self): if not self.input_folder or not os.path.isdir(self.input_folder): messagebox.showerror("Error", "Please select a valid input folder.") return self.submit_button.config(state="disabled") self.status.config(text="Processing, please wait...") self.progress["value"] = 0 self.progress_label.config(text="0%") self.clear_log() threading.Thread(target=self._run_extraction_thread).start() def _run_extraction_thread(self): week_filter = None week_str = self.week_var.get().strip() if week_str and not week_str.lower().startswith("optional"): try: week_filter = parse_weeks(week_str) except Exception as e: self.log(f"Invalid week filter: {e}\n") week_filter = None success, output_path = process_timesheets( self.input_folder, self.output_excel, self.log, self.update_progress, week_filter ) if success: self.status.config(text="Done! Data written to Excel.") messagebox.showinfo("Done", f"Timesheet data successfully written to:\n{output_path}") else: self.status.config(text="No data extracted.") messagebox.showwarning("No Data", "No timesheet data extracted from any image.") self.submit_button.config(state="normal") self.progress["value"] = 0 self.progress_label.config(text="0%") def show_about(self): messagebox.showinfo("About", "Timesheet Extractor\nVersion 1.0\n\nDeveloped by Your Team.") def show_help(self): messagebox.showinfo( "Help", "1. Click 'Browse Folder' to select the folder with timesheet images.\n" "2. (Optional) Enter week numbers to filter (e.g., 12,13,15-17).\n" "3. Click 'Submit' to start extraction.\n" "4. Progress and logs will be shown below.\n" "5. The result Excel file will be saved in the selected folder." ) if __name__ == "__main__": root = Tk() gui = TimesheetGUI(root) root.mainloop()Changes Made:
-
Week Extraction from OCR Text:
- Added a
week_patternto match the week number in the OCR text. - Modified the
process_timesheetsfunction to extract the week number usingweek_patternfrom the OCR text instead of the filename.
- Added a
-
Filename Parsing:
- Removed the week number extraction from the filename since it is now extracted from the OCR text.
- Kept the name extraction from the filename.
-
Week Filtering:
- The week filtering logic remains the same, but it now uses the week number extracted from the OCR text.
With these changes, the program will now correctly extract the week number from the OCR text and save it to the Excel file.
-
Other models, same or similar prompt
Responses were produced in different, uncontrolled contexts (system prompts, settings, prior turns). This is not a benchmark.