import pdfplumber
from jinja2 import Environment, FileSystemLoader
import os

# === #1: Extract table from PDF ===
with pdfplumber.open("original.pdf") as pdf:
    page = pdf.pages[0]
    table = page.extract_table()

# === #2: Extract dates and initialize storage ===
headers = table[0]  # ['', 'Sun 6 Jul', 'Sun 13 Jul', ...]
dates = headers[1:]
assignments = {}  # {'Cleaner': ['Person 1, Person 2', ...]}

current_role = None

for row in table[1:]:
    role = row[0]
    people = row[1:]

    if role:  # Start of a new role
        current_role = role
        assignments[current_role] = [p if p else "" for p in people]
    elif current_role:
        # Additional person(s) for the current role
        prev = assignments[current_role]
        for i in range(len(people)):
            if people[i]:
                if prev[i]:
                    prev[i] += f", {people[i]}"
                else:
                    prev[i] = people[i]
        assignments[current_role] = prev

# === #3: Render to HTML ===
env = Environment(loader=FileSystemLoader("."))
template = env.get_template("email_template.html")

html = template.render(dates=dates, assignments=assignments)

with open("roster_output.html", "w") as f:
    f.write(html)

print("✅ Full roster HTML generated.")