import re
import time
from datetime import datetime, timedelta
from pathlib import Path

from dateutil.relativedelta import relativedelta


GOOGLE_REVIEW_SCROLL_XPATH = (
    '//*[@id="QA0Szd"]/div/div/div[1]/div[2]/div/div[1]/div/div/div[5]/div[2]'
)
GOOGLE_REVIEW_SCROLL_SCRIPT = (
    'document.getElementsByClassName("dS8AEf")[0].scrollTop = '
    'document.getElementsByClassName("dS8AEf")[0].scrollHeight'
)


def build_google_chrome_options(
    headless=False,
    log_level=None,
    ignore_certificate_errors=False,
    ignore_certificate_errors_spki_list=False,
    ignore_ssl_errors=False,
    remote_debugging_pipe=False,
):
    from selenium import webdriver

    options = webdriver.ChromeOptions()
    if log_level is not None:
        options.add_argument(f"--log-level={log_level}")
    if ignore_certificate_errors:
        options.add_argument("--ignore-certificate-error")
    if ignore_certificate_errors_spki_list:
        options.add_argument("--ignore-certificate-errors-spki-list")
    if ignore_ssl_errors:
        options.add_argument("--ignore-ssl-errors")
    if headless:
        options.add_argument("--headless")
    options.add_argument("--lang=en-US")
    options.add_argument("--disable-blink-features=AutomationControlled")
    if remote_debugging_pipe:
        options.add_argument("--remote-debugging-pipe")
    options.add_experimental_option("excludeSwitches", ["enable-automation"])
    options.add_experimental_option("useAutomationExtension", False)
    return options


def set_stealth_driver(driver):
    driver.execute_script(
        "Object.defineProperty(navigator, 'webdriver', {get: () => undefined})"
    )


def count_google_review_pages(driver):
    result = driver.find_element("class name", 'Qha3nb').text
    result = result.replace(',', '')
    result = result.split(' ')
    result = result[0].split('\n')
    return int(result[0]) // 10 + 1


def scroll_google_reviews(driver, review_count, label='google_scroll...', delay_seconds=3):
    print(label, end="")
    time.sleep(delay_seconds)
    scrollable_div = driver.find_element("xpath", GOOGLE_REVIEW_SCROLL_XPATH)
    google_scroller = None
    for _ in range(review_count):
        try:
            google_scroller = driver.execute_script(
                GOOGLE_REVIEW_SCROLL_SCRIPT,
                scrollable_div,
            )
            time.sleep(delay_seconds)
            print('.', end="")
        except Exception as error:
            print(f"Error while {label.strip('.')}: {error}")
            break
    print('')
    return google_scroller


def sanitize_review_name(name):
    return re.sub(r'[^a-zA-Z0-9]', '', name)


def normalize_google_media_url(style_text):
    return re.sub(r'=\S*-p-k-no', '=-no', (re.findall(r"['\"](.*?)['\"]", style_text))[0])


def build_review_media_path(review_name, visit_date, filename, base_dir="./Output/Pics"):
    review_dir = Path(base_dir) / sanitize_review_name(review_name) / visit_date
    review_dir.mkdir(parents=True, exist_ok=True)
    return str(review_dir / filename)


def normalize_wordpress_date(date_text, now=None):
    reference_time = now or datetime.now()
    normalized = str(date_text).strip()
    lowered = normalized.lower()

    if lowered == "a day ago":
        publish_at = reference_time - timedelta(days=1)
    elif "day" in lowered:
        publish_at = reference_time - timedelta(days=int(re.sub(r'[^0-9]', '', lowered) or 0))
    elif lowered == "a week ago":
        publish_at = reference_time - relativedelta(weeks=1)
    elif "week" in lowered:
        publish_at = reference_time - relativedelta(weeks=int(re.sub(r'[^0-9]', '', lowered) or 0))
    elif lowered == "a month ago":
        publish_at = reference_time - relativedelta(months=1)
    elif "month" in lowered:
        publish_at = reference_time - relativedelta(months=int(re.sub(r'[^0-9]', '', lowered) or 0))
    elif lowered == "a year ago":
        publish_at = reference_time - relativedelta(years=1)
    elif "year" in lowered:
        publish_at = reference_time - relativedelta(years=int(re.sub(r'[^0-9]', '', lowered) or 0))
    else:
        compact_date = normalized.replace(" ", "")
        publish_at = datetime.strptime(compact_date[:3] + "/" + compact_date[3:] + "/01", '%b/%Y/%d')

    return publish_at.strftime('%Y-%m-%dT22:00:00')