# type: ignore

import json
from pathlib import Path

import requests
from bs4 import BeautifulSoup
from sentence_transformers import SentenceTransformer

KNOWLEDGE_BASE_URL = "https://help.suno.com/en"
KNOWLEDGE_BASE_URL_BASE = "https://help.suno.com"


def scrape_category_cards_for_urls(url, a_tag_class="kb-category-card"):
    try:
        response = requests.get(url)
        soup = BeautifulSoup(response.content, "html.parser")

        links = soup.find_all("a", class_=a_tag_class)
        urls = [
            None
            if link.get("href")[:8] == "https://"
            else f"{KNOWLEDGE_BASE_URL_BASE}{link.get('href')}"
            for link in links
            if link.get("href")
        ]

        return urls
    except Exception as e:
        print(f"Error scraping URL: {e}")
        return []


def scrape_knowledge_base():
    try:
        articles = []
        article_index = 0
        category_urls = scrape_category_cards_for_urls(KNOWLEDGE_BASE_URL)
        print(category_urls)
        if len(category_urls):
            for url in category_urls:
                if url is None:
                    continue
                print(f"Scraping {url} for categories...")
                subcategory_urls = scrape_category_cards_for_urls(url)
                for subcategory_url in subcategory_urls:
                    if subcategory_url is None:
                        continue
                    article_urls = scrape_category_cards_for_urls(
                        subcategory_url, a_tag_class="article-card"
                    )
                    for article_url in article_urls:
                        if article_url is None:
                            continue
                        article_title, article_text_chunks = scrape_webpage(article_url)
                        for chunk in article_text_chunks:
                            chunk_data = {
                                "text": chunk,
                                "url": article_url,
                                "title": article_title or f"article-{article_index}",
                            }
                            articles.append(chunk_data)
                        article_index += 1
        else:
            print(f"No articles found for {KNOWLEDGE_BASE_URL}")
            return []
        return articles

    except Exception as e:
        print(f"Error scraping {url}: {e}")
        return []


def scrape_webpage(url):
    """Scrape text content from a webpage"""
    print(f"Scraping {url}...")
    try:
        response = requests.get(url, timeout=10)
        soup = BeautifulSoup(response.content, "html.parser")

        # Remove script and style elements
        for script in soup(["script", "style"]):
            script.decompose()

        # Get text and title
        title = None
        title_el = soup.find("h1", class_="article-title")
        if title_el:
            title = title_el.get_text()
        content_el = soup.find("div", class_="article-content")
        text = None
        if content_el:
            text = content_el.get_text()

        # Clean up whitespace
        if text:
            chunks = (line.strip().replace("\n", " ") for line in text.splitlines())
            # chunks = (phrase.strip() for line in lines for phrase in line.split("  "))
        else:
            chunks = []

        return (title, [chunk for chunk in chunks if chunk])
    except Exception as e:
        print(f"Error scraping {url}: {e}")
        return ""


if __name__ == "__main__":
    articles = scrape_knowledge_base()  # returns { title -> text_chunks } dict
    model = SentenceTransformer("all-MiniLM-L6-v2")
    if articles:
        embeddings = model.encode([article["text"] for article in articles], show_progress_bar=True)
        for i, embedding in enumerate(embeddings):
            articles[i]["embedding"] = embedding.tolist()

    script_dir = Path(__file__).parent
    output_file = script_dir / ".." / "suno_orpheus" / "services" / "assets" / "knowledge_base.json"
    with open(output_file, "w", encoding="utf-8") as f:
        json.dump({"knowledge_chunks": articles}, f, ensure_ascii=False, indent=2)