import re

from selenium.webdriver.common.by import By

# Labels shown in the MassHealth Member Detail block, e.g.
#   Date of Birth: 4/21/1958
#   Gender:        Female
#   Address:       169 CYPRESS ST, NEWTON, MA 02459
#   Home Phone:    6036789376
#   Registered     Not Registered in Member Community
_LABELS = ("Date of Birth", "Gender", "Address", "Home Phone", "Phone", "Registered")


def _label_value(text, label):
    """Value next to "Label:" — same line or the next line (innerText of a
    two-column layout can put the value on its own line). Empty when the
    portal leaves the field blank."""
    m = re.search(rf"(?:^|\n)[ \t]*{re.escape(label)}:[ \t]*\n?[ \t]*([^\n]*)", text)
    if not m:
        return ""
    value = m.group(1).strip()
    # Blank field — we grabbed the next row's label instead of a value
    if value.endswith(":") or any(value.startswith(f"{l}:") or value == l for l in _LABELS):
        return ""
    return value


def extract_member_demographics(driver):
    """Read gender, address and phone from the MassHealth Member Detail page.

    Returns only the fields that were found:
      gender  — "Female" / "Male" as shown by the portal
      address, city, state, zipCode — split from "STREET, CITY, ST 12345"
      phone   — 10 digits
      memberName — the heading above "Date of Birth:" (e.g. "OLHA SVYSHCH")
    """
    result = {}
    try:
        text = driver.find_element(By.TAG_NAME, "body").get_attribute("innerText") or ""

        # Name is the last non-empty line before the "Date of Birth:" row
        dob = re.search(r"(?:^|\n)[ \t]*Date of Birth:", text)
        if dob:
            lines = [l.strip() for l in text[:dob.start()].split("\n") if l.strip()]
            if lines and ":" not in lines[-1]:
                result["memberName"] = lines[-1]

        gender = _label_value(text, "Gender")
        if gender:
            result["gender"] = gender

        address = _label_value(text, "Address")
        if address:
            # "169 CYPRESS ST, NEWTON, MA 02459" — street may itself contain commas (APT 2)
            m = re.match(r"^(.*),\s*([^,]+?),?\s+([A-Z]{2})\s+(\d{5})(?:-?\d{4})?$", address)
            if m:
                result["address"] = m.group(1).strip()
                result["city"] = m.group(2).strip()
                result["state"] = m.group(3)
                result["zipCode"] = m.group(4)
            else:
                result["address"] = address

        phone = _label_value(text, "Home Phone") or _label_value(text, "Phone")
        digits = re.sub(r"\D", "", phone)
        if len(digits) == 11 and digits.startswith("1"):
            digits = digits[1:]
        if len(digits) == 10:
            result["phone"] = digits

        print(f"[extract_member_demographics] {result}")
    except Exception as e:
        print(f"[extract_member_demographics] failed: {e}")
    return result