from bs4 import BeautifulSoup
from markdownify import markdownify
from requests.exceptions import HTTPError
from urllib.parse import urlparse

import requests
import textwrap

headers = {
    "User-Agent": "Mozilla/5.0 (Windows NT 10.0; Win64; x64; rv:135.0) Gecko/20100101 Firefox/135.0"
        }

class WikiError(Exception):
    pass

def fix_link(config, node, field):
    node[field] = node[field].replace("/wiki", config["site"] + "/wiki")
    node[field] = node[field].replace("/mediawiki", config["site"] + "/mediawiki")

def decompose(node, field):
    for unwanted in node.select(field):
        unwanted.decompose()

def get_search_response(config, api_root, query):
    try:
        response = requests.get(config['site'] + api_root + '/api.php?action=opensearch&format=json&formatversion=2&search=' + query + '&namespace=0&limit=6&suggest=true', headers=headers, timeout=7)
    except requests.exceptions.Timeout:
        raise WikiError("Request timed out.")
    except requests.exceptions.RequestException as e:
        print(f"Failed to complete request: {e}")
        raise WikiError("Error with wiki request.")

    try:
        response.raise_for_status()
    except HTTPError as e:
        print(f"HTTP error occured: {e}")
        raise WikiError("Bad HTTP response.")

    data = response.json()

    if not data[3]:
        raise WikiError("No such wiki entry found.")

    return data[3]

def get_page_content(config, page_name):
    try:
        response = requests.get(page_name, headers=headers, timeout=7)
    except requests.exceptions.Timeout:
        raise WikiError("Request timed out.")
    except requests.exceptions.RequestException as e:
        print(f"Failed to complete request: {e}")
        raise WikiError("Error with wiki request.")

    try:
        response.raise_for_status()
    except HTTPError as e:
        print(f"HTTP error occured: {e}")
        raise WikiError("Bad HTTP response.")

    soup = BeautifulSoup(response.content, "html.parser")
    
    parser_output = soup.select_one(".mw-parser-output")

    decompose(parser_output, "sup")
    decompose(parser_output, "div.portal")
    decompose(parser_output, "table.nottemplate")
    decompose(parser_output, "table.Anpassen")
    decompose(parser_output, "div.quotebox")
    decompose(parser_output, "div.floatleft")

    image = parser_output.find("a", {"class": "image"})
    image_link = None
    if image:
        if image["href"] == "/wiki/File:Targetdrone.gif":
            image = parser_output.find_all("a", {"class": "image"})[1]

    if image:
        if image.contents[0]:
            fix_link(config, image.contents[0], "src")
            image_link = image.contents[0]["src"]

            try:
                result = urlparse(image_link)

                if not result.netloc:
                    fix_link(config, image.contents[0], "data-src")
                    image_link = image.contents[0]["data-src"]
            except ValueError:
                fix_link(config, image.contents[0], "data-src")
                image_link = image.contents[0]["data-src"]

    decompose(parser_output, "br")
    decompose(parser_output, "table")
    decompose(parser_output, "aside")

    for p in parser_output.find_all("p"):
        if not p.get_text(strip=True):
            p.decompose()

    content = parser_output.find("p")
    for a in content.findAll("a"):
        fix_link(config, a, "href")

    header_md = markdownify(str(soup.find(id="firstHeading").contents[0]))
    if not header_md:
        header_md = markdownify(str(soup.select_one(".mw-page-title-main")))

    content_md = textwrap.shorten(markdownify(str(content)), width=1000, placeholder="...")

    return header_md, content_md, image_link