PyKubeGrader 0.1.2__py3-none-any.whl → 0.1.3__py3-none-any.whl

pykubegrader/build/build_folder.py

@@ -0,0 +1,1531 @@
+from dataclasses import dataclass, field
+import os
+import shutil
+import nbformat
+import subprocess
+import sys
+import argparse
+import logging
+import json
+import re
+import importlib.util
+
+
+@dataclass
+class NotebookProcessor:
+    """
+    A class for processing Jupyter notebooks in a directory and its subdirectories.
+
+    Attributes:
+        root_folder (str): The root directory containing notebooks to process.
+        solutions_folder (str): The directory where processed notebooks and solutions are stored.
+        verbose (bool): Flag for verbose output to the console.
+        log (bool): Flag to enable or disable logging.
+    """
+
+    root_folder: str
+    solutions_folder: str = field(init=False)
+    verbose: bool = False
+    log: bool = True
+
+    def __post_init__(self):
+        """
+        Post-initialization method for setting up the `NotebookProcessor` instance.
+
+        This method is automatically called after the instance is created. It performs the following tasks:
+            1. Creates a solutions folder within the root directory to store processed outputs.
+            2. Configures logging to capture detailed information about the processing.
+
+        Raises:
+            OSError: If the solutions folder cannot be created due to permissions or other filesystem issues.
+        """
+        # Define the folder to store solutions and ensure it exists
+        self.solutions_folder = os.path.join(self.root_folder, "_solutions")
+        os.makedirs(
+            self.solutions_folder, exist_ok=True
+        )  # Create the folder if it doesn't exist
+
+        # Configure logging to store log messages in the solutions folder
+        log_file_path = os.path.join(self.solutions_folder, "notebook_processor.log")
+        logging.basicConfig(
+            filename=log_file_path,  # Path to the log file
+            level=logging.INFO,  # Log messages at INFO level and above will be recorded
+            format="%(asctime)s - %(levelname)s - %(message)s",  # Log message format: timestamp, level, and message
+        )
+
+        # Initialize a global logger for the class
+        global logger
+        logger = logging.getLogger(
+            __name__
+        )  # Create a logger instance specific to this module
+        self.logger = logger  # Assign the logger instance to the class for use in instance methods
+
+    def process_notebooks(self):
+        """
+        Recursively processes Jupyter notebooks in a given folder and its subfolders.
+
+        The function performs the following steps:
+        1. Iterates through all files within the root folder and subfolders.
+        2. Identifies Jupyter notebooks by checking file extensions (.ipynb).
+        3. Checks if each notebook contains assignment configuration metadata.
+        4. Processes notebooks that meet the criteria using `otter assign` or other defined steps.
+
+        Prerequisites:
+            - The `has_assignment` method should be implemented to check if a notebook
+            contains the required configuration for assignment processing.
+            - The `_process_single_notebook` method should handle the specific processing
+            of a single notebook, including moving it to a new folder or running
+            additional tools like `otter assign`.
+
+        Raises:
+            - OSError: If an issue occurs while accessing files or directories.
+
+        Example:
+            class NotebookProcessor:
+                def __init__(self, root_folder):
+                    self.root_folder = root_folder
+
+                def has_assignment(self, notebook_path):
+                    # Implementation to check for assignment configuration
+                    return True  # Replace with actual check logic
+
+                def _process_single_notebook(self, notebook_path):
+                    # Implementation to process a single notebook
+                    self._print_and_log(f"Processing notebook: {notebook_path}")
+
+            processor = NotebookProcessor("/path/to/root/folder")
+            processor.process_notebooks()
+        """
+        ipynb_files = []
+
+        # Walk through the root folder and its subfolders
+        for dirpath, _, filenames in os.walk(self.root_folder):
+            for filename in filenames:
+                # Check if the file is a Jupyter notebook
+                if filename.endswith(".ipynb"):
+                    notebook_path = os.path.join(dirpath, filename)
+                    ipynb_files.append(notebook_path)
+
+        for notebook_path in ipynb_files:
+            # Check if the notebook has the required assignment configuration
+            if self.has_assignment(notebook_path):
+                self._print_and_log(f"notebook_path = {notebook_path}")
+
+                # Process the notebook if it meets the criteria
+                self._process_single_notebook(notebook_path)
+
+    def _print_and_log(self, message):
+        """
+        Logs a message and optionally prints it to the console.
+
+        This method is used for logging important information and optionally
+        displaying it in the console based on the `verbose` and `log` attributes.
+
+        Args:
+            message (str): The message to be logged and/or printed.
+
+        Behavior:
+            - If `self.verbose` is True, the message will be printed to the console.
+            - If `self.log` is True, the message will be logged using the class's logger.
+
+        Example:
+            self._print_and_log("Processing completed successfully.")
+
+        Raises:
+            None: This method handles exceptions internally, if any arise from logging or printing.
+        """
+
+        # Print the message to the console if verbosity is enabled
+        if self.verbose:
+            print(message)
+
+        # Log the message if logging is enabled
+        if self.log:
+            self.logger.info(message)
+
+    def _process_single_notebook(self, notebook_path):
+        """
+        Processes a single Jupyter notebook.
+
+        This method handles the preparation, validation, and processing of a given notebook. It:
+        1. Moves the notebook to a subfolder within the solutions folder.
+        2. Creates temporary and destination folders for autograder and student files.
+        3. Identifies and processes multiple-choice questions (MCQs).
+        4. Runs assignment-specific tasks like executing `otter assign` and cleaning notebooks.
+        5. Generates solution and question files and moves them to appropriate folders.
+
+        Args:
+            notebook_path (str): The file path to the Jupyter notebook to be processed.
+
+        Raises:
+            FileNotFoundError: If the notebook file or intermediate files are not found.
+            OSError: If there are issues creating or moving files/directories.
+            Exception: For unexpected errors during processing.
+
+        Returns:
+            None
+        """
+
+        logging.info(f"Processing notebook: {notebook_path}")
+        notebook_name = os.path.splitext(os.path.basename(notebook_path))[0]
+        notebook_subfolder = os.path.join(self.solutions_folder, notebook_name)
+        os.makedirs(notebook_subfolder, exist_ok=True)
+
+        new_notebook_path = os.path.join(
+            notebook_subfolder, os.path.basename(notebook_path)
+        )
+
+        # makes a temp copy of the notebook
+        temp_notebook_path = os.path.join(
+            notebook_subfolder, f"{notebook_name}_temp.ipynb"
+        )
+        shutil.copy(notebook_path, temp_notebook_path)
+
+        # Determine the path to the autograder folder
+        autograder_path = os.path.join(notebook_subfolder, f"dist/autograder/")
+        os.makedirs(autograder_path, exist_ok=True)
+
+        # Determine the path to the student folder
+        student_path = os.path.join(notebook_subfolder, f"dist/student/")
+        os.makedirs(student_path, exist_ok=True)
+
+        if os.path.abspath(notebook_path) != os.path.abspath(new_notebook_path):
+            shutil.move(notebook_path, new_notebook_path)
+            self._print_and_log(f"Moved: {notebook_path} -> {new_notebook_path}")
+        else:
+            self._print_and_log(f"Notebook already in destination: {new_notebook_path}")
+
+        ### Parse the notebook for multiple choice questions
+        if self.has_assignment(temp_notebook_path, "# BEGIN MULTIPLE CHOICE"):
+            self._print_and_log(
+                f"Notebook {temp_notebook_path} has multiple choice questions"
+            )
+
+            # Extract all the multiple choice questions
+            data = extract_MCQ(temp_notebook_path)
+
+            # determine the output file path
+            solution_path = f"{os.path.splitext(new_notebook_path)[0]}_solutions.py"
+
+            # Extract the first value cells
+            value = extract_raw_cells(temp_notebook_path)
+
+            data = NotebookProcessor.merge_metadata(value, data)
+
+            for data_ in data:
+                # Generate the solution file
+                self.generate_solution_MCQ(data, output_file=solution_path)
+
+                question_path = (
+                    f"{new_notebook_path.replace(".ipynb", "")}_questions.py"
+                )
+
+            generate_mcq_file(data, output_file=question_path)
+
+            markers = ("# BEGIN MULTIPLE CHOICE", "# END MULTIPLE CHOICE")
+
+            replace_cells_between_markers(
+                data, markers, temp_notebook_path, temp_notebook_path
+            )
+
+        ### Parse the notebook for TF questions
+        if self.has_assignment(temp_notebook_path, "# BEGIN TF"):
+
+            markers = ("# BEGIN TF", "# END TF")
+
+            self._print_and_log(
+                f"Notebook {temp_notebook_path} has True False questions"
+            )
+
+            # Extract all the multiple choice questions
+            data = extract_TF(temp_notebook_path)
+
+            # determine the output file path
+            solution_path = f"{os.path.splitext(new_notebook_path)[0]}_solutions.py"
+
+            # Extract the first value cells
+            value = extract_raw_cells(temp_notebook_path, markers[0])
+
+            data = NotebookProcessor.merge_metadata(value, data)
+
+            # for data_ in data:
+            # Generate the solution file
+            self.generate_solution_MCQ(data, output_file=solution_path)
+
+            question_path = f"{new_notebook_path.replace(".ipynb", "")}_questions.py"
+
+            generate_tf_file(data, output_file=question_path)
+
+            replace_cells_between_markers(
+                data, markers, temp_notebook_path, temp_notebook_path
+            )
+
+        ### Parse the notebook for select_many questions
+        if self.has_assignment(temp_notebook_path, "# BEGIN SELECT MANY"):
+
+            markers = ("# BEGIN SELECT MANY", "# END SELECT MANY")
+
+            self._print_and_log(
+                f"Notebook {temp_notebook_path} has True False questions"
+            )
+
+            # Extract all the multiple choice questions
+            data = extract_SELECT_MANY(temp_notebook_path)
+
+            # determine the output file path
+            solution_path = f"{os.path.splitext(new_notebook_path)[0]}_solutions.py"
+
+            # Extract the first value cells
+            value = extract_raw_cells(temp_notebook_path, markers[0])
+
+            data = NotebookProcessor.merge_metadata(value, data)
+
+            # for data_ in data:
+            # Generate the solution file
+            self.generate_solution_MCQ(data, output_file=solution_path)
+
+            question_path = f"{new_notebook_path.replace(".ipynb", "")}_questions.py"
+
+            generate_select_many_file(data, output_file=question_path)
+
+            replace_cells_between_markers(
+                data, markers, temp_notebook_path, temp_notebook_path
+            )
+
+        if self.has_assignment(temp_notebook_path, "# ASSIGNMENT CONFIG"):
+            self.run_otter_assign(
+                temp_notebook_path, os.path.join(notebook_subfolder, "dist")
+            )
+            student_notebook = os.path.join(
+                notebook_subfolder, "dist", "student", f"{notebook_name}.ipynb"
+            )
+            self.clean_notebook(student_notebook)
+            NotebookProcessor.replace_temp_in_notebook(
+                student_notebook, student_notebook
+            )
+            autograder_notebook = os.path.join(
+                notebook_subfolder, "dist", "autograder", f"{notebook_name}.ipynb"
+            )
+            NotebookProcessor.replace_temp_in_notebook(
+                autograder_notebook, autograder_notebook
+            )
+            shutil.copy(student_notebook, self.root_folder)
+            self._print_and_log(
+                f"Copied and cleaned student notebook: {student_notebook} -> {self.root_folder}"
+            )
+
+        # If Otter does not run, move the student file to the main directory
+        if "student_notebook" not in locals():
+            path_ = shutil.copy(temp_notebook_path, self.root_folder)
+            self._print_and_log(
+                f"Copied and cleaned student notebook: {path_} -> {self.root_folder}"
+            )
+
+        # Move the solution file to the autograder folder
+        if "solution_path" in locals():
+            # gets importable file name
+            importable_file_name = sanitize_string(
+                os.path.splitext(os.path.basename(solution_path))[0]
+            )
+
+            # Move the solution file to the autograder folder
+            os.rename(
+                solution_path,
+                os.path.join(autograder_path, f"{importable_file_name}.py"),
+            )
+
+        if "question_path" in locals():
+            shutil.move(question_path, student_path)
+
+        # Remove the temp copy of the notebook
+        os.remove(temp_notebook_path)
+
+        # Remove all postfix from filenames in dist
+        NotebookProcessor.remove_postfix(autograder_path, "_solutions")
+        NotebookProcessor.remove_postfix(student_path, "_questions")
+        NotebookProcessor.remove_postfix(self.root_folder, "_temp")
+
+        ### CODE TO ENSURE THAT STUDENT NOTEBOOK IS IMPORTABLE
+        if "question_path" in locals():
+            
+            # question_root_path = os.path.dirname(question_path)
+            question_file_name = os.path.basename(question_path)
+            question_file_name_sanitized = sanitize_string(question_file_name.replace("_questions", ""))
+            if question_file_name_sanitized.endswith("_py"):
+                question_file_name_sanitized = question_file_name_sanitized[:-3] + ".py"
+            
+            # Rename the file
+            os.rename(os.path.join(student_path, question_file_name.replace("_questions", "")), os.path.join(student_path, question_file_name_sanitized))
+            
+            # Ensure the "questions" folder exists
+            questions_folder_jbook = os.path.join(self.root_folder, "questions")
+            os.makedirs(questions_folder_jbook, exist_ok=True)
+            
+            # Copy the renamed file to the "questions" folder
+            shutil.copy(os.path.join(student_path, question_file_name_sanitized), os.path.join(questions_folder_jbook, question_file_name_sanitized))
+
+
+    @staticmethod
+    def replace_temp_in_notebook(input_file, output_file):
+        """
+        Replaces occurrences of '_temp.ipynb' with '.ipynb' in a Jupyter Notebook.
+
+        Parameters:
+        input_file (str): Path to the input Jupyter Notebook file.
+        output_file (str): Path to the output Jupyter Notebook file.
+
+        Returns:
+        None: Writes the modified notebook to the output file.
+        """
+        # Load the notebook data
+        with open(input_file, "r", encoding="utf-8") as f:
+            notebook_data = json.load(f)
+
+        # Iterate through each cell and update its content
+        for cell in notebook_data.get("cells", []):
+            if "source" in cell:
+                # Replace occurrences of '_temp.ipynb' in the cell source
+                cell["source"] = [
+                    line.replace("_temp.ipynb", ".ipynb") for line in cell["source"]
+                ]
+
+        # Write the updated notebook to the output file
+        with open(output_file, "w", encoding="utf-8") as f:
+            json.dump(notebook_data, f, indent=2)
+
+    @staticmethod
+    def merge_metadata(raw, data):
+        """
+        Merges raw metadata with extracted question data.
+
+        This method combines metadata from two sources: raw metadata and question data.
+        It ensures that the points associated with each question are appropriately distributed
+        and added to the final merged metadata.
+
+        Args:
+            raw (list): A list of dictionaries containing raw metadata.
+                        Each dictionary must have a 'points' key with a value
+                        that can be either a list of points or a string representing a single point value.
+            data (list): A list of dictionaries containing extracted question data.
+                        Each dictionary represents a set of questions and their details.
+
+        Returns:
+            list: A list of dictionaries where each dictionary represents a question
+                with merged metadata and associated points.
+
+        Raises:
+            KeyError: If 'points' is missing from any raw metadata entry.
+            IndexError: If the number of items in `raw` and `data` do not match.
+
+        Example:
+            raw = [
+                {"points": [1.0, 2.0]},
+                {"points": "3.0"}
+            ]
+            data = [
+                {"Q1": {"question_text": "What is 2+2?"}},
+                {"Q2": {"question_text": "What is 3+3?"}}
+            ]
+            merged = merge_metadata(raw, data)
+            print(merged)
+            # Output:
+            # [
+            #     {"Q1": {"question_text": "What is 2+2?", "points": 1.0}},
+            #     {"Q2": {"question_text": "What is 3+3?", "points": 3.0}}
+            # ]
+        """
+        merged_data = []
+
+        # Loop through each question set in the data
+        for i, _data in enumerate(data):
+            # Handle 'points' from raw metadata: convert single string value to a list if necessary
+            if isinstance(raw[i]["points"], str):
+                points_ = [float(raw[i]["points"])] * len(
+                    _data
+                )  # Distribute the same point value
+            else:
+                points_ = raw[i]["points"]  # Use provided list of points
+
+            # Remove 'points' from raw metadata to avoid overwriting
+            raw[i].pop("points", None)
+
+            # Handle 'grade' from raw metadata
+            if "grade" in raw[i]:
+                grade_ = [raw[i]["grade"]]
+
+            # Merge each question's metadata with corresponding raw metadata
+            for j, (key, value) in enumerate(_data.items()):
+                # Combine raw metadata with question data
+                data[i][key] = data[i][key] | raw[i]
+                # Assign the correct point value to the question
+                data[i][key]["points"] = points_[j]
+
+                if "grade" in raw[i]:
+                    data[i][key]["grade"] = grade_
+
+        return data
+
+    @staticmethod
+    def has_assignment(notebook_path, *tags):
+        """
+        Determines if a Jupyter notebook contains any of the specified configuration tags.
+
+        This method checks for the presence of specific content in a Jupyter notebook
+        to identify whether it includes any of the required headings or tags.
+
+        Args:
+            notebook_path (str): The file path to the Jupyter notebook to be checked.
+            *tags (str): Variable-length argument list of tags to search for.
+                        Defaults to ("# ASSIGNMENT CONFIG",).
+
+        Returns:
+            bool: True if the notebook contains any of the specified tags, False otherwise.
+
+        Dependencies:
+            - The `check_for_heading` function must be implemented. It should search
+            for specific headings or content in a notebook file and return a boolean
+            value indicating if any of the tags exist.
+
+        Example:
+            def check_for_heading(notebook_path, keywords):
+                # Mock implementation of content check
+                with open(notebook_path, 'r') as file:
+                    content = file.read()
+                return any(keyword in content for keyword in keywords)
+
+            notebook_path = "path/to/notebook.ipynb"
+            # Check for default tags
+            contains_config = has_assignment(notebook_path)
+            self._print_and_log(f"Contains assignment config: {contains_config}")
+
+            # Check for custom tags
+            contains_custom = has_assignment(notebook_path, "# CUSTOM CONFIG", "# ANOTHER CONFIG")
+            self._print_and_log(f"Contains custom config: {contains_custom}")
+        """
+        # Default tags if none are provided
+        if not tags:
+            tags = ["# ASSIGNMENT CONFIG", "# BEGIN MULTIPLE CHOICE"]
+
+        # Use the helper function to check for the presence of any specified tag
+        return check_for_heading(notebook_path, tags)
+
+    @staticmethod
+    def run_otter_assign(notebook_path, dist_folder):
+        """
+        Runs `otter assign` on the given notebook and outputs to the specified distribution folder.
+        """
+        try:
+            os.makedirs(dist_folder, exist_ok=True)
+            command = ["otter", "assign", notebook_path, dist_folder]
+            subprocess.run(command, check=True)
+            logger.info(f"Otter assign completed: {notebook_path} -> {dist_folder}")
+
+            # Remove all postfix _test from filenames in dist_folder
+            NotebookProcessor.remove_postfix(dist_folder)
+
+        except subprocess.CalledProcessError as e:
+            logger.info(f"Error running `otter assign` for {notebook_path}: {e}")
+        except Exception as e:
+            logger.info(
+                f"Unexpected error during `otter assign` for {notebook_path}: {e}"
+            )
+
+    @staticmethod
+    def generate_solution_MCQ(data_list, output_file="output.py"):
+        """
+        Generates a Python file with solutions and total points based on the input data.
+        If the file already exists, it appends new solutions to the existing solution dictionary.
+
+        Args:
+            data_list (list): A list of dictionaries containing question metadata.
+            output_file (str): Path to the output Python file.
+        """
+
+        solutions = {}
+        total_points = 0.0
+
+        # If the output file exists, load the existing solutions and total_points
+        if os.path.exists(output_file):
+            spec = importlib.util.spec_from_file_location(
+                "existing_module", output_file
+            )
+            existing_module = importlib.util.module_from_spec(spec)
+            spec.loader.exec_module(existing_module)  # Load the module dynamically
+
+            # Attempt to read existing solutions and total_points
+            if hasattr(existing_module, "solutions"):
+                solutions.update(existing_module.solutions)
+            if hasattr(existing_module, "total_points"):
+                total_points += existing_module.total_points
+
+        # Process new question data and update solutions and total_points
+        for question_set in data_list:
+            for key, question_data in question_set.items():
+                solution_key = f"q{question_data['question number']}-{question_data['subquestion_number']}-{key}"
+                solutions[solution_key] = question_data["solution"]
+                total_points += question_data["points"]
+
+        # Write updated total_points and solutions back to the file
+        with open(output_file, "w", encoding="utf-8") as f:
+            f.write("from typing import Any\n\n")
+            f.write(f"total_points: float = {total_points}\n\n")
+
+            f.write("solutions: dict[str, Any] = {\n")
+            for key, solution in solutions.items():
+                # For safety, we assume solutions are strings, but if not, repr would be safer
+                f.write(f'    "{key}": {repr(solution)},\n')
+            f.write("}\n")
+
+    @staticmethod
+    def generate_solution_MCQ(data_list, output_file="output.py"):
+        """
+        Generates a Python file with solutions and total points based on the input data.
+        If the file already exists, it appends new solutions to the existing solution dictionary.
+
+        Args:
+            data_list (list): A list of dictionaries containing question metadata.
+            output_file (str): Path to the output Python file.
+        """
+
+        solutions = {}
+        total_points = 0.0
+
+        # If the output file exists, load the existing solutions and total_points
+        if os.path.exists(output_file):
+            spec = importlib.util.spec_from_file_location(
+                "existing_module", output_file
+            )
+            existing_module = importlib.util.module_from_spec(spec)
+            spec.loader.exec_module(existing_module)  # Load the module dynamically
+
+            # Attempt to read existing solutions and total_points
+            if hasattr(existing_module, "solutions"):
+                solutions.update(existing_module.solutions)
+            if hasattr(existing_module, "total_points"):
+                total_points += existing_module.total_points
+
+        # Process new question data and update solutions and total_points
+        for question_set in data_list:
+            for key, question_data in question_set.items():
+                solution_key = f"q{question_data['question number']}-{question_data['subquestion_number']}-{key}"
+                solutions[solution_key] = question_data["solution"]
+                total_points += question_data["points"]
+
+        # Write updated total_points and solutions back to the file
+        with open(output_file, "w", encoding="utf-8") as f:
+            f.write("from typing import Any\n\n")
+            f.write(f"total_points: float = {total_points}\n\n")
+
+            f.write("solutions: dict[str, Any] = {\n")
+            for key, solution in solutions.items():
+                # For safety, we assume solutions are strings, but if not, repr would be safer
+                f.write(f'    "{key}": {repr(solution)},\n')
+            f.write("}\n")
+
+    def extract_MCQ(ipynb_file):
+        """
+        Extracts questions from markdown cells and organizes them as a nested dictionary,
+        including subquestion numbers.
+
+        Args:
+            ipynb_file (str): Path to the .ipynb file.
+
+        Returns:
+            dict: A nested dictionary where the first-level key is the question name (text after ##),
+                and the value is a dictionary with keys: 'name', 'subquestion_number',
+                'question_text', 'OPTIONS', and 'solution'.
+        """
+        try:
+            # Load the notebook file
+            with open(ipynb_file, "r", encoding="utf-8") as f:
+                notebook_data = json.load(f)
+
+            cells = notebook_data.get("cells", [])
+            results = {}
+            within_section = False
+            subquestion_number = 0  # Counter for subquestions
+
+            for cell in cells:
+                if cell.get("cell_type") == "raw":
+                    # Check for the start and end labels in raw cells
+                    raw_content = "".join(cell.get("source", []))
+                    if "# BEGIN MULTIPLE CHOICE" in raw_content:
+                        within_section = True
+                        subquestion_number = (
+                            0  # Reset counter at the start of a new section
+                        )
+                        continue
+                    elif "# END MULTIPLE CHOICE" in raw_content:
+                        within_section = False
+                        continue
+
+                if within_section and cell.get("cell_type") == "markdown":
+                    # Parse markdown cell content
+                    markdown_content = "".join(cell.get("source", []))
+
+                    # Extract title (## heading)
+                    title_match = re.search(
+                        r"^##\s*(.+)", markdown_content, re.MULTILINE
+                    )
+                    title = title_match.group(1).strip() if title_match else None
+
+                    if title:
+                        subquestion_number += (
+                            1  # Increment the subquestion number for each question
+                        )
+
+                        # Extract question text (### heading)
+                        question_text_match = re.search(
+                            r"^###\s*\*\*(.+)\*\*", markdown_content, re.MULTILINE
+                        )
+                        question_text = (
+                            question_text_match.group(1).strip()
+                            if question_text_match
+                            else None
+                        )
+
+                        # Extract OPTIONS (lines after #### options)
+                        options_match = re.search(
+                            r"####\s*options\s*(.+?)(?=####|$)",
+                            markdown_content,
+                            re.DOTALL | re.IGNORECASE,
+                        )
+                        options = (
+                            [
+                                line.strip()
+                                for line in options_match.group(1).strip().splitlines()
+                                if line.strip()
+                            ]
+                            if options_match
+                            else []
+                        )
+
+                        # Extract solution (line after #### SOLUTION)
+                        solution_match = re.search(
+                            r"####\s*SOLUTION\s*(.+)", markdown_content, re.IGNORECASE
+                        )
+                        solution = (
+                            solution_match.group(1).strip() if solution_match else None
+                        )
+
+                        # Create nested dictionary for the question
+                        results[title] = {
+                            "name": title,
+                            "subquestion_number": subquestion_number,
+                            "question_text": question_text,
+                            "OPTIONS": options,
+                            "solution": solution,
+                        }
+
+            return results
+
+        except FileNotFoundError:
+            print(f"File {ipynb_file} not found.")
+            return {}
+        except json.JSONDecodeError:
+            print("Invalid JSON in notebook file.")
+            return {}
+
+    @staticmethod
+    def remove_postfix(dist_folder, suffix="_temp"):
+        logging.info(f"Removing postfix '{suffix}' from filenames in {dist_folder}")
+        for root, _, files in os.walk(dist_folder):
+            for file in files:
+                if suffix in file:
+                    old_file_path = os.path.join(root, file)
+                    new_file_path = os.path.join(root, file.replace(suffix, ""))
+                    os.rename(old_file_path, new_file_path)
+                    logging.info(f"Renamed: {old_file_path} -> {new_file_path}")
+
+    @staticmethod
+    def clean_notebook(notebook_path):
+        """
+        Cleans a Jupyter notebook to remove unwanted cells and set cell metadata.
+        """
+        clean_notebook(notebook_path)
+
+
+def extract_raw_cells(ipynb_file, heading="# BEGIN MULTIPLE CHOICE"):
+    """
+    Extracts all metadata from value cells in a Jupyter Notebook file for a specified heading.
+
+    Args:
+        ipynb_file (str): Path to the .ipynb file.
+        heading (str): The heading to search for in value cells.
+
+    Returns:
+        list of dict: A list of dictionaries containing extracted metadata for each heading occurrence.
+    """
+    try:
+        with open(ipynb_file, "r", encoding="utf-8") as f:
+            notebook_data = json.load(f)
+
+        # Extract value cell content
+        raw_cells = [
+            "".join(
+                cell.get("source", [])
+            )  # Join multiline sources into a single string
+            for cell in notebook_data.get("cells", [])
+            if cell.get("cell_type") == "raw"
+        ]
+
+        # Process each value cell to extract metadata
+        metadata_list = []
+        for raw_cell in raw_cells:
+            metadata_list.extend(_extract_metadata_from_heading(raw_cell, heading))
+
+        return metadata_list
+
+    except FileNotFoundError:
+        print(f"File {ipynb_file} not found.")
+        return []
+    except json.JSONDecodeError:
+        print("Invalid JSON in notebook file.")
+        return []
+
+
+def _extract_metadata_from_heading(raw_cell, heading="# BEGIN MULTIPLE CHOICE"):
+    """
+    Extracts metadata for a single value cell string each time the heading is found.
+
+    Args:
+        raw_cell (str): String containing value cell content.
+        heading (str): The heading to identify sections.
+
+    Returns:
+        list of dict: A list of dictionaries containing extracted key-value pairs.
+    """
+    metadata_list = []
+    lines = raw_cell.split("\n")
+    current_metadata = None
+
+    for line in lines:
+        if line.startswith(heading):
+            if current_metadata:
+                metadata_list.append(current_metadata)  # Save previous metadata
+            current_metadata = {}  # Start new metadata block
+        elif line.startswith("##") and current_metadata is not None:
+            # Extract key and value from lines
+            key, value = line[3:].split(":", 1)
+            current_metadata[key.strip()] = value.strip()
+
+    if current_metadata:  # Append the last metadata block
+        metadata_list.append(current_metadata)
+
+    return metadata_list
+
+
+def extract_SELECT_MANY(ipynb_file):
+    """
+    Extracts questions marked by `# BEGIN SELECT MANY` and `# END SELECT MANY` in markdown cells,
+    including all lines under the SOLUTION header until the first blank line or whitespace-only line.
+
+    Args:
+        ipynb_file (str): Path to the .ipynb file.
+
+    Returns:
+        list: A list of dictionaries, where each dictionary corresponds to questions within
+              a section. Each dictionary contains parsed questions with details like
+              'name', 'subquestion_number', 'question_text', and 'solution'.
+    """
+    try:
+        # Load the notebook file
+        with open(ipynb_file, "r", encoding="utf-8") as f:
+            notebook_data = json.load(f)
+
+        cells = notebook_data.get("cells", [])
+        sections = []  # List to store results for each section
+        current_section = {}  # Current section being processed
+        within_section = False
+        subquestion_number = 0  # Counter for subquestions
+
+        for cell in cells:
+            if cell.get("cell_type") == "raw":
+                # Check for the start and end labels in raw cells
+                raw_content = "".join(cell.get("source", []))
+                if "# BEGIN SELECT MANY" in raw_content:
+                    within_section = True
+                    subquestion_number = (
+                        0  # Reset counter at the start of a new section
+                    )
+                    current_section = {}  # Prepare a new section dictionary
+                    continue
+                elif "# END SELECT MANY" in raw_content:
+                    within_section = False
+                    if current_section:
+                        sections.append(current_section)  # Save the current section
+                    continue
+
+            if within_section and cell.get("cell_type") == "markdown":
+                # Parse markdown cell content
+                markdown_content = "".join(cell.get("source", []))
+
+                # Extract title (## heading)
+                title_match = re.search(r"^##\s*(.+)", markdown_content, re.MULTILINE)
+                title = title_match.group(1).strip() if title_match else None
+
+                if title:
+                    subquestion_number += (
+                        1  # Increment subquestion number for each question
+                    )
+
+                    # Extract question text (### heading)
+                    question_text_match = re.search(
+                        r"^###\s*\*\*(.+)\*\*", markdown_content, re.MULTILINE
+                    )
+                    question_text = (
+                        question_text_match.group(1).strip()
+                        if question_text_match
+                        else None
+                    )
+
+                    # Extract OPTIONS (lines after #### options)
+                    options_match = re.search(
+                        r"####\s*options\s*(.+?)(?=####|$)",
+                        markdown_content,
+                        re.DOTALL | re.IGNORECASE,
+                    )
+                    options = (
+                        [
+                            line.strip()
+                            for line in options_match.group(1).strip().splitlines()
+                            if line.strip()
+                        ]
+                        if options_match
+                        else []
+                    )
+
+                    # Extract all lines under the SOLUTION header
+                    solution_start = markdown_content.find("#### SOLUTION")
+                    if solution_start != -1:
+                        solution = []
+                        lines = markdown_content[solution_start:].splitlines()
+                        for line in lines[1:]:  # Skip the "#### SOLUTION" line
+                            if line.strip():  # Non-blank line after trimming spaces
+                                solution.append(line.strip())
+                            else:
+                                break
+
+                    # Add question details to the current section
+                    current_section[title] = {
+                        "name": title,
+                        "subquestion_number": subquestion_number,
+                        "question_text": question_text,
+                        "solution": solution,
+                        "OPTIONS": options,
+                    }
+
+        return sections
+
+    except FileNotFoundError:
+        print(f"File {ipynb_file} not found.")
+        return []
+    except json.JSONDecodeError:
+        print("Invalid JSON in notebook file.")
+        return []
+
+
+def extract_TF(ipynb_file):
+    """
+    Extracts True False questions from markdown cells within sections marked by
+    `# BEGIN TF` and `# END TF`.
+
+    Args:
+        ipynb_file (str): Path to the .ipynb file.
+
+    Returns:
+        list: A list of dictionaries, where each dictionary corresponds to questions within
+              a section. Each dictionary contains parsed questions with details like
+              'name', 'subquestion_number', 'question_text', and 'solution'.
+    """
+    try:
+        # Load the notebook file
+        with open(ipynb_file, "r", encoding="utf-8") as f:
+            notebook_data = json.load(f)
+
+        cells = notebook_data.get("cells", [])
+        sections = []  # List to store results for each section
+        current_section = {}  # Current section being processed
+        within_section = False
+        subquestion_number = 0  # Counter for subquestions
+
+        for cell in cells:
+            if cell.get("cell_type") == "raw":
+                # Check for the start and end labels in raw cells
+                raw_content = "".join(cell.get("source", []))
+                if "# BEGIN TF" in raw_content:
+                    within_section = True
+                    subquestion_number = (
+                        0  # Reset counter at the start of a new section
+                    )
+                    current_section = {}  # Prepare a new section dictionary
+                    continue
+                elif "# END TF" in raw_content:
+                    within_section = False
+                    if current_section:
+                        sections.append(current_section)  # Save the current section
+                    continue
+
+            if within_section and cell.get("cell_type") == "markdown":
+                # Parse markdown cell content
+                markdown_content = "".join(cell.get("source", []))
+
+                # Extract title (## heading)
+                title_match = re.search(r"^##\s*(.+)", markdown_content, re.MULTILINE)
+                title = title_match.group(1).strip() if title_match else None
+
+                if title:
+                    subquestion_number += (
+                        1  # Increment subquestion number for each question
+                    )
+
+                    # Extract question text (### heading)
+                    question_text_match = re.search(
+                        r"^###\s*\*\*(.+)\*\*", markdown_content, re.MULTILINE
+                    )
+                    question_text = (
+                        question_text_match.group(1).strip()
+                        if question_text_match
+                        else None
+                    )
+
+                    # Extract solution (line after #### SOLUTION)
+                    solution_match = re.search(
+                        r"####\s*SOLUTION\s*(.+)", markdown_content, re.IGNORECASE
+                    )
+                    solution = (
+                        solution_match.group(1).strip() if solution_match else None
+                    )
+
+                    # Add question details to the current section
+                    current_section[title] = {
+                        "name": title,
+                        "subquestion_number": subquestion_number,
+                        "question_text": question_text,
+                        "solution": solution,
+                    }
+
+        return sections
+
+    except FileNotFoundError:
+        print(f"File {ipynb_file} not found.")
+        return []
+    except json.JSONDecodeError:
+        print("Invalid JSON in notebook file.")
+        return []
+
+
+def extract_MCQ(ipynb_file):
+    """
+    Extracts multiple-choice questions from markdown cells within sections marked by
+    `# BEGIN MULTIPLE CHOICE` and `# END MULTIPLE CHOICE`.
+
+    Args:
+        ipynb_file (str): Path to the .ipynb file.
+
+    Returns:
+        list: A list of dictionaries, where each dictionary corresponds to questions within
+              a section. Each dictionary contains parsed questions with details like
+              'name', 'subquestion_number', 'question_text', 'OPTIONS', and 'solution'.
+    """
+    try:
+        # Load the notebook file
+        with open(ipynb_file, "r", encoding="utf-8") as f:
+            notebook_data = json.load(f)
+
+        cells = notebook_data.get("cells", [])
+        sections = []  # List to store results for each section
+        current_section = {}  # Current section being processed
+        within_section = False
+        subquestion_number = 0  # Counter for subquestions
+
+        for cell in cells:
+            if cell.get("cell_type") == "raw":
+                # Check for the start and end labels in raw cells
+                raw_content = "".join(cell.get("source", []))
+                if "# BEGIN MULTIPLE CHOICE" in raw_content:
+                    within_section = True
+                    subquestion_number = (
+                        0  # Reset counter at the start of a new section
+                    )
+                    current_section = {}  # Prepare a new section dictionary
+                    continue
+                elif "# END MULTIPLE CHOICE" in raw_content:
+                    within_section = False
+                    if current_section:
+                        sections.append(current_section)  # Save the current section
+                    continue
+
+            if within_section and cell.get("cell_type") == "markdown":
+                # Parse markdown cell content
+                markdown_content = "".join(cell.get("source", []))
+
+                # Extract title (## heading)
+                title_match = re.search(r"^##\s*(.+)", markdown_content, re.MULTILINE)
+                title = title_match.group(1).strip() if title_match else None
+
+                if title:
+                    subquestion_number += (
+                        1  # Increment subquestion number for each question
+                    )
+
+                    # Extract question text (### heading)
+                    question_text_match = re.search(
+                        r"^###\s*\*\*(.+)\*\*", markdown_content, re.MULTILINE
+                    )
+                    question_text = (
+                        question_text_match.group(1).strip()
+                        if question_text_match
+                        else None
+                    )
+
+                    # Extract OPTIONS (lines after #### options)
+                    options_match = re.search(
+                        r"####\s*options\s*(.+?)(?=####|$)",
+                        markdown_content,
+                        re.DOTALL | re.IGNORECASE,
+                    )
+                    options = (
+                        [
+                            line.strip()
+                            for line in options_match.group(1).strip().splitlines()
+                            if line.strip()
+                        ]
+                        if options_match
+                        else []
+                    )
+
+                    # Extract solution (line after #### SOLUTION)
+                    solution_match = re.search(
+                        r"####\s*SOLUTION\s*(.+)", markdown_content, re.IGNORECASE
+                    )
+                    solution = (
+                        solution_match.group(1).strip() if solution_match else None
+                    )
+
+                    # Add question details to the current section
+                    current_section[title] = {
+                        "name": title,
+                        "subquestion_number": subquestion_number,
+                        "question_text": question_text,
+                        "OPTIONS": options,
+                        "solution": solution,
+                    }
+
+        return sections
+
+    except FileNotFoundError:
+        print(f"File {ipynb_file} not found.")
+        return []
+    except json.JSONDecodeError:
+        print("Invalid JSON in notebook file.")
+        return []
+
+
+def check_for_heading(notebook_path, search_strings):
+    """
+    Checks if a Jupyter notebook contains a heading cell whose source matches any of the given strings.
+    """
+    try:
+        with open(notebook_path, "r", encoding="utf-8") as f:
+            notebook = nbformat.read(f, as_version=4)
+            for cell in notebook.cells:
+                if cell.cell_type == "raw" and cell.source.startswith("#"):
+                    if any(
+                        search_string in cell.source for search_string in search_strings
+                    ):
+                        return True
+    except Exception as e:
+        logger.info(f"Error reading notebook {notebook_path}: {e}")
+    return False
+
+
+def clean_notebook(notebook_path):
+    """
+    Removes specific cells and makes Markdown cells non-editable and non-deletable by updating their metadata.
+    """
+    try:
+        with open(notebook_path, "r", encoding="utf-8") as f:
+            notebook = nbformat.read(f, as_version=4)
+
+        cleaned_cells = []
+        for cell in notebook.cells:
+            if not hasattr(cell, "cell_type") or not hasattr(cell, "source"):
+                continue
+
+            if (
+                "## Submission" not in cell.source
+                and "# Save your notebook first," not in cell.source
+            ):
+                if cell.cell_type == "markdown":
+                    cell.metadata["editable"] = cell.metadata.get("editable", False)
+                    cell.metadata["deletable"] = cell.metadata.get("deletable", False)
+                if cell.cell_type == "code":
+                    cell.metadata["tags"] = cell.metadata.get("tags", [])
+                    if "skip-execution" not in cell.metadata["tags"]:
+                        cell.metadata["tags"].append("skip-execution")
+
+                cleaned_cells.append(cell)
+            else:
+                (f"Removed cell: {cell.source.strip()[:50]}...")
+
+        notebook.cells = cleaned_cells
+
+        with open(notebook_path, "w", encoding="utf-8") as f:
+            nbformat.write(notebook, f)
+        logger.info(f"Cleaned notebook: {notebook_path}")
+
+    except Exception as e:
+        logger.info(f"Error cleaning notebook {notebook_path}: {e}")
+
+
+def ensure_imports(output_file, header_lines):
+    """
+    Ensures specified header lines are present at the top of the file.
+
+    Args:
+        output_file (str): The path of the file to check and modify.
+        header_lines (list of str): Lines to ensure are present at the top.
+
+    Returns:
+        str: The existing content of the file (without the header).
+    """
+    existing_content = ""
+    if os.path.exists(output_file):
+        with open(output_file, "r", encoding="utf-8") as f:
+            existing_content = f.read()
+
+    # Determine missing lines
+    missing_lines = [line for line in header_lines if line not in existing_content]
+
+    # Write the updated content back to the file
+    with open(output_file, "w", encoding="utf-8") as f:
+        # Add missing lines at the top
+        f.writelines(missing_lines)
+        # Retain the existing content
+        f.write(existing_content)
+
+    return existing_content
+
+
+def replace_cells_between_markers(data, markers, ipynb_file, output_file):
+    """
+    Replaces the cells between specified markers in a Jupyter Notebook (.ipynb file)
+    with provided replacement cells and writes the result to the output file.
+
+    Parameters:
+    data (list): A list of dictionaries with data for creating replacement cells.
+    markers (tuple): A tuple containing two strings: the BEGIN and END markers.
+    ipynb_file (str): Path to the input Jupyter Notebook file.
+    output_file (str): Path to the output Jupyter Notebook file.
+
+    Returns:
+    None: Writes the modified notebook to the output file.
+    """
+    begin_marker, end_marker = markers
+    file_name_ipynb = ipynb_file.split("/")[-1].replace("_temp.ipynb", "")
+
+    file_name_ipynb = sanitize_string(file_name_ipynb)
+
+    # Iterate over each set of replacement data
+    for data_ in data:
+        dict_ = data_[next(iter(data_.keys()))]
+
+        # Create the replacement cells
+        replacement_cells = {
+            "cell_type": "code",
+            "metadata": {},
+            "source": [
+                "# Run this block of code by pressing Shift + Enter to display the question\n",
+                f"from questions.{file_name_ipynb} import Question{dict_['question number']}\n",
+                f"Question{dict_['question number']}().show()\n",
+            ],
+            "outputs": [],
+            "execution_count": None,
+        }
+
+        # Process the notebook cells
+        new_cells = []
+        inside_markers = False
+        done = False
+
+        # Load the notebook data
+        with open(ipynb_file, "r", encoding="utf-8") as f:
+            notebook_data = json.load(f)
+
+        for cell in notebook_data["cells"]:
+            if cell.get("cell_type") == "raw" and not done:
+                if any(begin_marker in line for line in cell.get("source", [])):
+                    # Enter the marked block
+                    inside_markers = True
+                    new_cells.append(replacement_cells)
+                    continue
+                elif inside_markers:
+                    if any(end_marker in line for line in cell.get("source", [])):
+                        # Exit the marked block
+                        inside_markers = False
+                        done = True
+                        continue
+                    else:
+                        continue
+                else:
+                    new_cells.append(cell)
+            elif inside_markers:
+                # Skip cells inside the marked block
+                continue
+            else:
+                new_cells.append(cell)
+                continue
+
+            if done:
+                # Add cells outside the marked block
+                new_cells.append(cell)
+                continue
+
+        # Update the notebook with modified cells, preserving metadata
+        notebook_data["cells"] = new_cells
+
+        # Write the modified notebook to the output file
+        with open(output_file, "w", encoding="utf-8") as f:
+            json.dump(notebook_data, f, indent=2)
+
+        # Update ipynb_file to the output file for subsequent iterations
+        ipynb_file = output_file
+
+
+def generate_mcq_file(data_dict, output_file="mc_questions.py"):
+    """
+    Generates a Python file defining an MCQuestion class from a dictionary.
+
+    Args:
+        data_dict (dict): A nested dictionary containing question metadata.
+        output_file (str): The path for the output Python file.
+
+    Returns:
+        None
+    """
+
+    # Define header lines
+    header_lines = [
+        "from pykubegrader.widgets.multiple_choice import MCQuestion, MCQ\n",
+        "import pykubegrader.initialize\n",
+        "import panel as pn\n\n",
+        "pn.extension()\n\n",
+    ]
+
+    # Ensure header lines are present
+    existing_content = ensure_imports(output_file, header_lines)
+
+    for question_dict in data_dict:
+        with open(output_file, "a", encoding="utf-8") as f:
+            for i, (q_key, q_value) in enumerate(question_dict.items()):
+                if i == 0:
+                    # Write the MCQuestion class
+                    f.write(
+                        f"class Question{q_value['question number']}(MCQuestion):\n"
+                    )
+                    f.write("    def __init__(self):\n")
+                    f.write("        super().__init__(\n")
+                    f.write(f"            title=f'{q_value['question_text']}',\n")
+                    f.write("            style=MCQ,\n")
+                    f.write(
+                        f"            question_number={q_value['question number']},\n"
+                    )
+                break
+
+            keys = []
+            for i, (q_key, q_value) in enumerate(question_dict.items()):
+                # Write keys
+                keys.append(f"q{q_value['subquestion_number']}-{q_value['name']}")
+
+            f.write(f"            keys={keys},\n")
+
+            options = []
+            for i, (q_key, q_value) in enumerate(question_dict.items()):
+                # Write options
+                options.append(q_value["OPTIONS"])
+
+            f.write(f"            options={options},\n")
+
+            descriptions = []
+            for i, (q_key, q_value) in enumerate(question_dict.items()):
+                # Write descriptions
+                descriptions.append(q_value["question_text"])
+            f.write(f"            descriptions={descriptions},\n")
+
+            points = []
+            for i, (q_key, q_value) in enumerate(question_dict.items()):
+                # Write points
+                points.append(q_value["points"])
+
+            f.write(f"            points={points},\n")
+            f.write("        )\n")
+
+
+def generate_select_many_file(data_dict, output_file="select_many_questions.py"):
+    """
+    Generates a Python file defining an MCQuestion class from a dictionary.
+
+    Args:
+        data_dict (dict): A nested dictionary containing question metadata.
+        output_file (str): The path for the output Python file.
+
+    Returns:
+        None
+    """
+
+    # Define header lines
+    header_lines = [
+        "from pykubegrader.widgets.select_many import MultiSelect, SelectMany\n",
+        "import pykubegrader.initialize\n",
+        "import panel as pn\n\n",
+        "pn.extension()\n\n",
+    ]
+
+    # Ensure header lines are present
+    existing_content = ensure_imports(output_file, header_lines)
+
+    for question_dict in data_dict:
+        with open(output_file, "a", encoding="utf-8") as f:
+            for i, (q_key, q_value) in enumerate(question_dict.items()):
+                if i == 0:
+                    # Write the MCQuestion class
+                    f.write(
+                        f"class Question{q_value['question number']}(SelectMany):\n"
+                    )
+                    f.write("    def __init__(self):\n")
+                    f.write("        super().__init__(\n")
+                    f.write(f"            title=f'{q_value['question_text']}',\n")
+                    f.write("            style=MultiSelect,\n")
+                    f.write(
+                        f"            question_number={q_value['question number']},\n"
+                    )
+                break
+
+            keys = []
+            for i, (q_key, q_value) in enumerate(question_dict.items()):
+                # Write keys
+                keys.append(f"q{q_value['subquestion_number']}-{q_value['name']}")
+
+            f.write(f"            keys={keys},\n")
+
+            descriptions = []
+            for i, (q_key, q_value) in enumerate(question_dict.items()):
+                # Write descriptions
+                descriptions.append(q_value["question_text"])
+            f.write(f"            descriptions={descriptions},\n")
+
+            options = []
+            for i, (q_key, q_value) in enumerate(question_dict.items()):
+                # Write options
+                options.append(q_value["OPTIONS"])
+
+            f.write(f"            options={options},\n")
+
+            points = []
+            for i, (q_key, q_value) in enumerate(question_dict.items()):
+                # Write points
+                points.append(q_value["points"])
+
+            f.write(f"            points={points},\n")
+
+            first_key = next(iter(question_dict))
+            if "grade" in question_dict[first_key]:
+                grade = question_dict[first_key]["grade"]
+                f.write(f"            grade='{grade}',\n")
+
+            f.write("        )\n")
+
+
+def generate_tf_file(data_dict, output_file="tf_questions.py"):
+    """
+    Generates a Python file defining an MCQuestion class from a dictionary.
+
+    Args:
+        data_dict (dict): A nested dictionary containing question metadata.
+        output_file (str): The path for the output Python file.
+
+    Returns:
+        None
+    """
+
+    # Define header lines
+    header_lines = [
+        "from pykubegrader.widgets.true_false import TFQuestion, TrueFalse_style\n",
+        "import pykubegrader.initialize\n",
+        "import panel as pn\n\n",
+        "pn.extension()\n\n",
+    ]
+
+    # Ensure header lines are present
+    existing_content = ensure_imports(output_file, header_lines)
+
+    for question_dict in data_dict:
+        with open(output_file, "a", encoding="utf-8") as f:
+            for i, (q_key, q_value) in enumerate(question_dict.items()):
+                if i == 0:
+                    # Write the MCQuestion class
+                    f.write(
+                        f"class Question{q_value['question number']}(TFQuestion):\n"
+                    )
+                    f.write("    def __init__(self):\n")
+                    f.write("        super().__init__(\n")
+                    f.write(f"            title=f'{q_value['question_text']}',\n")
+                    f.write("            style=TrueFalse_style,\n")
+                    f.write(
+                        f"            question_number={q_value['question number']},\n"
+                    )
+                break
+
+            keys = []
+            for i, (q_key, q_value) in enumerate(question_dict.items()):
+                # Write keys
+                keys.append(f"q{q_value['subquestion_number']}-{q_value['name']}")
+
+            f.write(f"            keys={keys},\n")
+
+            descriptions = []
+            for i, (q_key, q_value) in enumerate(question_dict.items()):
+                # Write descriptions
+                descriptions.append(q_value["question_text"])
+            f.write(f"            descriptions={descriptions},\n")
+
+            points = []
+            for i, (q_key, q_value) in enumerate(question_dict.items()):
+                # Write points
+                points.append(q_value["points"])
+
+            f.write(f"            points={points},\n")
+            f.write("        )\n")
+
+
+def sanitize_string(input_string):
+    """
+    Converts a string into a valid Python variable name.
+
+    Args:
+        input_string (str): The string to convert.
+
+    Returns:
+        str: A valid Python variable name.
+    """
+    # Replace invalid characters with underscores
+    sanitized = re.sub(r"\W|^(?=\d)", "_", input_string)
+    return sanitized
+
+
+def main():
+    parser = argparse.ArgumentParser(
+        description="Recursively process Jupyter notebooks with '# ASSIGNMENT CONFIG', move them to a solutions folder, and run otter assign."
+    )
+    parser.add_argument(
+        "root_folder", type=str, help="Path to the root folder to process"
+    )
+    args = parser.parse_args()
+
+    processor = NotebookProcessor(args.root_folder)
+    processor.process_notebooks()
+
+
+if __name__ == "__main__":
+    sys.exit(main())