import json

import requests
from bs4 import BeautifulSoup
import pandas as pd

FILES_DIR = './files'


def main(filename):
    with open(input_filename, 'r') as f:
        configuration_doc_html = f.read()
    soup = BeautifulSoup(configuration_doc_html, "html.parser")

    # Step 2: Parse the webpage content
    data = []

    # Locate the configuration settings table or sections
    sections = soup.find_all("div", class_="sect1")  # Adjust this to match the webpage structure

    for section in sections:
        group = section.find("h2").get_text(strip=True)
        settings = section.find_all("div", class_="sect2")

        for setting in settings:
            setting_name = setting.select("h3 code")[0].text
            setting_info = setting.select('table tr')

            description = setting_info[0].select('td')[1].text
            valid_values = setting_info[1].select('td')[1].text
            try:
                default_value = setting_info[2].select('td')[1].text
            except IndexError:
                default_value = None

            # Check for tags indicating dynamic or deprecated
            # dynamic = "Dynamic" in row.get_text()
            # deprecated = "Deprecated" in row.get_text()

            # Append to the data list
            data.append({
                "Settings Group": group,
                "Setting": setting_name,
                "Description": "\n".join([line for line in description.splitlines() if line.strip()]),
                "Valid Values": valid_values,
                "Default Value": default_value,
                # "Dynamic": dynamic,
                # "Deprecated": deprecated
            })

    # Step 3: Create a DataFrame and export
    # df = pd.DataFrame(data)
    # df.to_excel(f'{FILES_DIR}/{output_filename}', index=False, engine='openpyxl')

    return data


def get_renamed_settings():
    doc_page = _html_to_bs(f'{FILES_DIR}/neo4j_docs/v5_changes.html')
    result = []
    rows = doc_page.find(id='_renamed').find_next_sibling().find_all('tr')
    for i, row in enumerate(rows):
        try:
            result.append(
                {
                    'v4': row.select('td')[1].text,
                    'v5': row.select('td')[2].text,
                }
            )
        except:
            pass
    return result


def get_deprecated_settings():
    doc_page = _html_to_bs(f'{FILES_DIR}/neo4j_docs/v5_changes.html')
    result = []
    rows = doc_page.find(id='_configuration_settings_deprecated_in_neo4j_5_x').find_next_sibling().find_all('tr')
    for i, row in enumerate(rows):
        try:
            version = ''
            try:
                version = row.select('td')[2].select('.label--deprecated')[0].text
            except:
                pass
            result.append(
                {
                    'deprecated_setting': row.select('td')[0].text,
                    'deprecated_setting_description': row.select('td')[1].text,
                    'deprecated_version': version
                }
            )
        except:
            pass
    return result


def _html_to_bs(filename):
    with open(filename, 'r') as f:
        configuration_doc_html = f.read()
    return BeautifulSoup(configuration_doc_html, "html.parser")


if __name__ == "__main__":
    input_filename = f'{FILES_DIR}/neo4j_docs/configuration_doc_v5.html'
    output_filename = 'neo4j_configuration_settings_v4.xlsx'
    v5_properties = main(input_filename)
    with open(f'{FILES_DIR}/neo4j_docs/v5_settings.json', 'w') as f:
        json.dump(v5_properties, f)

    # renamed_settings = get_renamed_settings()
    # with open(f'{FILES_DIR}/neo4j_docs/renamed_settings.json', 'w') as f:
    #     json.dump(renamed_settings, f)

    # deprecated_settings = get_deprecated_settings()
    # with open(f'{FILES_DIR}/neo4j_docs/deprecated_settings.json', 'w') as f:
    #     json.dump(deprecated_settings, f)