# Kyler Olsen
# July 2025
import requests
import bs4
import datetime
import json

# This script downloads the final exam schedule from Snow College's registrar page
# and parses it to extract the semester and exam times for each course.

# Last tested for Fall 2025 on 2025-07-10

def parse_days(days_str):
    # Mon, Wed, or Fri
    # Tue or Thu
    days = days_str.split(" ")
    days = [day.strip(',') for day in days if day.strip(',')]
    day_map = {
        "Mon": "M",
        "Tue": "T",
        "Wed": "W",
        "Thu": "Th",
        "Fri": "F",
    }
    return [day_map[day] for day in days if day in day_map]

def parse_class_time(time_str) -> tuple[datetime.time, datetime.time]:
    # 7:00 or 7:30am
    # *5:00pm or later
    am_or_pm = "am" if "am" in time_str else "pm"
    time_parts = time_str.replace("*", "").replace(am_or_pm, "").split(" ")
    start_time = datetime.datetime.strptime(time_parts[0] + am_or_pm, "%I:%M%p").time()
    if "later" in time_str:
        end_time = datetime.time(23, 59)
    else:
        end_time = datetime.datetime.strptime(time_parts[-1] + am_or_pm, "%I:%M%p").time()
    return start_time, end_time

def parse_schedule(content):
    soup = bs4.BeautifulSoup(content, 'html.parser')
    semester = soup.select_one("h2 strong").get_text(strip=True) # type: ignore
    year = semester.split()[-1]
    # schedule = {}
    schedule = []
    date = datetime.date(int(year), 1, 1)
    earliest_date = datetime.date(int(year)+1, 1, 1)
    for row in soup.select("tbody tr"):
        date_label = row.find("th")
        new_date = f"{date_label.get_text(strip=True)}, {year}" if date_label and date_label.get_text(strip=True) else None
        if new_date:
            new_date = new_date.replace(' ,', ',')
            date = datetime.datetime.strptime(new_date, "%A, %B %d, %Y").date()
            earliest_date = min(earliest_date, date)
        cols = row.find_all("td")
        if len(cols) == 3:
            days = cols[0].get_text(strip=True)
            class_time = cols[1].get_text(strip=True)
            exam_time = cols[2].get_text(strip=True)
            exam_datetime = datetime.datetime.strptime(f"{date} {exam_time.split(' - ')[0]}", "%Y-%m-%d %I:%M%p")
            for day in parse_days(days):
                start_time, end_time = parse_class_time(class_time)
                # schedule[(day, start_time, end_time)] = exam_datetime
                schedule.append({
                    "day": day,
                    "start_time": start_time.strftime("%H:%M"),
                    "end_time": end_time.strftime("%H:%M"),
                    "exam_datetime": exam_datetime.strftime("%Y-%m-%d %H:%M")
                })
        else: print("Unexpected number of columns in row:", len(cols))
    return semester, schedule, earliest_date.strftime("%Y-%m-%d")

def download_schedule(url):
    response = requests.get(url)
    if response.status_code == 200:
        return response.content
    else:
        raise Exception(f"Failed to retrieve schedule. Status code: {response.status_code}")

if __name__ == "__main__":
    url = "https://www.snow.edu/offices/registrar/final_exams.html"
    content = download_schedule(url)
    # with open("final_exams.html", "r") as file:
    #     content = file.read()
    semester, schedule, earliest_date = parse_schedule(content)
    with open("www/final_exams.json", "w") as file:
        json.dump({
            "url": url,
            "semester": semester,
            "updated_date": datetime.date.today().strftime("%Y-%m-%d"),
            "earliest_date": earliest_date,
            "schedule": schedule,
        }, file, indent=4)
    # print("Semester:", semester)
    # print("Schedule:")
    # for (day, start_time, end_time), date in schedule.items():
    #     print(f"{day} {start_time} - {end_time}: {date}")