Brightspace MCP Server

""" Brightspace MCP Server - Playwright-based web scraping for Purdue University """ from playwright.sync_api import sync_playwright, Browser, Page import time import json import os from typing import Dict, List, Optional from dataclasses import dataclass from dotenv import load_dotenv # Load environment variables load_dotenv() @dataclass class Course: """Data class for course information""" name: str code: str instructor: str url: str @dataclass class Assignment: """Data class for assignment information""" title: str due_date: str course: str status: str url: str class BrightspaceScraper: """Main class for scraping Purdue Brightspace data""" def __init__(self, headless: bool = False): self.headless = headless self.browser: Optional[Browser] = None self.page: Optional[Page] = None def __enter__(self): """Context manager entry""" self.playwright = sync_playwright().start() self.browser = self.playwright.chromium.launch( headless=self.headless, args=['--no-sandbox', '--disable-dev-shm-usage'] ) self.page = self.browser.new_page() return self def __exit__(self, exc_type, exc_val, exc_tb): """Context manager exit""" if self.browser: self.browser.close() if hasattr(self, 'playwright'): self.playwright.stop() def login(self, username: str, password: str) -> bool: """ Login to Purdue Brightspace with Duo Mobile 2FA Args: username: Purdue Career Account username password: Purdue Career Account password Returns: bool: True if login successful, False otherwise """ try: # Navigate to Purdue Brightspace login print("Navigating to Purdue Brightspace...") self.page.goto("https://purdue.brightspace.com") # Wait for page to load self.page.wait_for_load_state("networkidle") # Click on Purdue West Lafayette / Indianapolis button print("Looking for Purdue institution selector...") try: purdue_button = self.page.wait_for_selector("text=Purdue West Lafayette / Indianapolis", timeout=5000) if purdue_button: print("Clicking Purdue West Lafayette button...") purdue_button.click() self.page.wait_for_load_state("networkidle") except Exception as e: print(f"Note: Institution selector not found or not needed: {e}") # Wait for login form to load print("Waiting for login form...") self.page.wait_for_selector("#username", timeout=10000) # Fill in credentials print("Entering credentials...") self.page.fill("#username", username) self.page.fill("#password", password) # Click login button print("Clicking login button...") self.page.click("button[type='submit']") # Wait for Duo Mobile prompt print("Waiting for Duo Mobile authentication...") print("Please approve the login request on your Duo Mobile app...") # Wait for successful authentication by checking for Brightspace homepage try: # Wait for navigation away from Duo page to Brightspace self.page.wait_for_function( "() => window.location.href.includes('/d2l/home')", timeout=60000 ) print("Duo Mobile authentication successful!") return True except Exception as e: print(f"Authentication failed or timed out: {e}") return False except Exception as e: print(f"Login error: {e}") return False def get_courses(self) -> List[Course]: """ Scrape list of enrolled courses Returns: List[Course]: List of course objects """ try: # Navigate to home page print("Navigating to homepage...") self.page.goto("https://purdue.brightspace.com/d2l/home") # Wait for the My Courses widget to load (it's a web component) print("Waiting for My Courses widget to load...") self.page.wait_for_selector("d2l-my-courses", timeout=10000) # Wait a bit more for the courses to render inside the component time.sleep(3) # The courses are loaded dynamically, so we need to wait for them # Try multiple possible selectors based on the actual HTML structure course_selectors = [ "d2l-enrollment-card .d2l-card-container", # Main course containers ".d2l-card-container", # Course card containers "d2l-enrollment-card a[href*='/d2l/home/']", # Course links "a[href*='/d2l/home/']", # Any course links ] courses = [] for selector in course_selectors: print(f"Trying selector: {selector}") course_elements = self.page.query_selector_all(selector) if course_elements: print(f"Found {len(course_elements)} course elements with {selector}") for element in course_elements: try: # If this is a link element, use it directly if element.evaluate("el => el.tagName.toLowerCase()") == "a": link_element = element else: # Otherwise, look for a link inside this element link_element = element.query_selector("a[href*='/d2l/home/']") if not link_element: continue # Get course URL url = link_element.get_attribute("href") if not url: continue # Get course name from the span inside the link name_span = link_element.query_selector(".d2l-card-link-text") if name_span: name = name_span.text_content().strip() else: # Fallback: get from link text or title name = ( link_element.get_attribute("title") or link_element.get_attribute("aria-label") or link_element.text_content() ) if name: name = name.strip() # Extract course code from name course_code = self._extract_course_code(name) if name else "Unknown" if name and name != "": courses.append(Course( name=name, code=course_code, instructor="", url=url )) print(f"Found course: {name} ({course_code})") except Exception as e: print(f"Error extracting course from element: {e}") continue # If we found courses, don't try other selectors if courses: break # If no courses found with any selector, take a screenshot for debugging if not courses: print("No courses found! Taking screenshot...") self.page.screenshot(path="homepage_debug.png") print("Screenshot saved as homepage_debug.png") print("Printing page HTML to debug...") # Get the inner HTML of the my-courses widget my_courses = self.page.query_selector("d2l-my-courses") if my_courses: print(f"My Courses widget found, but no course elements inside") else: print("My Courses widget not found!") return courses except Exception as e: print(f"Error getting courses: {e}") self.page.screenshot(path="error_getting_courses.png") return [] def get_assignments(self, course_url: str) -> List[Assignment]: """ Scrape assignments for a specific course Args: course_url: URL of the course Returns: List[Assignment]: List of assignment objects """ try: # Ensure course_url is a full URL if not course_url.startswith("http"): course_url = f"https://purdue.brightspace.com{course_url}" self.page.goto(course_url) # Navigate to assignments (adjust selector as needed) assignments_link = self.page.query_selector("a[href*='assignments']") if assignments_link: assignments_link.click() self.page.wait_for_load_state("networkidle") assignments = [] # Extract assignment information (selectors will need adjustment) assignment_elements = self.page.query_selector_all(".assignment-row") for element in assignment_elements: try: title_element = element.query_selector(".assignment-title") due_date_element = element.query_selector(".due-date") status_element = element.query_selector(".status") title = title_element.text_content().strip() if title_element else "Unknown" due_date = due_date_element.text_content().strip() if due_date_element else "No due date" status = status_element.text_content().strip() if status_element else "Unknown" assignments.append(Assignment( title=title, due_date=due_date, course=self._extract_course_from_url(course_url), status=status, url=title_element.get_attribute("href") if title_element else "" )) except Exception as e: print(f"Error extracting assignment: {e}") continue return assignments except Exception as e: print(f"Error getting assignments: {e}") return [] def _extract_course_code(self, course_name: str) -> str: """Extract course code from course name""" # Simple regex to extract course codes like "CS 18000" or "MATH 16500" import re match = re.search(r'([A-Z]{2,4})\s+(\d{5})', course_name) if match: return f"{match.group(1)} {match.group(2)}" return course_name.split(' - ')[0] if ' - ' in course_name else course_name def _extract_course_from_url(self, url: str) -> str: """Extract course name from URL""" # Extract course identifier from URL parts = url.split('/') for part in parts: if 'course' in part.lower(): return part return "Unknown Course" def save_data(self, data: Dict, filename: str): """Save scraped data to JSON file""" try: with open(filename, 'w') as f: json.dump(data, f, indent=2) print(f"Data saved to {filename}") except Exception as e: print(f"Error saving data: {e}") def main(): """Example usage of the Brightspace scraper""" # Load credentials from environment variables USERNAME = os.getenv("PURDUE_USERNAME") PASSWORD = os.getenv("PURDUE_PASSWORD") if not USERNAME or not PASSWORD: print("❌ Error: PURDUE_USERNAME and PURDUE_PASSWORD must be set in .env file") print("Please create a .env file with:") print("PURDUE_USERNAME=your_username") print("PURDUE_PASSWORD=your_password") return print(f"Using username: {USERNAME}") # Use the scraper with context manager with BrightspaceScraper(headless=False) as scraper: # Login if scraper.login(USERNAME, PASSWORD): print("Login successful! Scraping data...") # Get courses courses = scraper.get_courses() print(f"Found {len(courses)} courses:") for course in courses: print(f" - {course.name} ({course.code})") # Get assignments for first course (if any) if courses: first_course_url = courses[0].url if first_course_url: assignments = scraper.get_assignments(first_course_url) print(f"\nFound {len(assignments)} assignments in {courses[0].name}:") for assignment in assignments: print(f" - {assignment.title} (Due: {assignment.due_date})") # Save data data = { "courses": [ { "name": course.name, "code": course.code, "url": course.url } for course in courses ], "scraped_at": time.strftime("%Y-%m-%d %H:%M:%S") } scraper.save_data(data, "brightspace_data.json") else: print("Login failed!") if __name__ == "__main__": main()