import re
from loguru import logger
import markdownify
import msgspec
from selectolax.lexbor import LexborHTMLParser
import tqdm

decoder = msgspec.json.Decoder()
encoder = msgspec.json.Encoder()

md = markdownify.MarkdownConverter(escape_misc=False, escape_asterisks=False)

with open("ImperialRoad/royalroad-merge.jsonl", "rb") as f, open(
    "TheRoyalCarpet.jsonl", "wb"
) as f2:
    for line in tqdm.tqdm(f, desc="Processing Lines..."):
        k = decoder.decode(line)
        nodes = LexborHTMLParser(k["html"])
        anti_air = None
        for style in nodes.css("head > style"):
            style_text = style.text(strip=True).replace("\n", " ")
            if "display" in style_text and "speak" in style_text:
                anti = re.search(r"(\..*)\{", style_text)
                # logger.debug(f"Anti Detection: {}")
                anti_air = anti.group(1)
        if not anti_air:
            logger.warning("Missing Anti Air defense.")
            continue
        for node in nodes.css(anti_air):
            # logger.debug(f"Shooting bird... {node.text()}")
            node.decompose()
        novel_title = nodes.select("title")
        chapter_data = nodes.select("div.chapter-inner.chapter-content")
        if not novel_title or not chapter_data:
            continue
        # The actual novel title
        novel_title = nodes.css_first('div > a[href*="/fiction/" i] > h2')
        if not novel_title:
            logger.warning("? No novel title?")
            continue
        royal_data = {"fiction": {}, "chapter": {}, "html": "", "text": ""}
        royal_data["fiction"]["title"] = novel_title.text().rstrip()
        novel_id = nodes.css_first('head > link[rel*="canon"i]')
        # logger.debug(novel_id)
        if novel_id:
            royal_data["fiction"]["id"] = int(novel_id.attributes["href"].split("/")[4])
        if novel_title and novel_title.parent and novel_title.parent.parent:
            meta_root = novel_title.parent.parent

            author_title = meta_root.css_first("h3.font-white.inline-block")
            if not author_title:
                logger.warning("? No novel title? Parents exist")
                continue
            royal_data["fiction"]["author"] = author_title.text(strip=True)

            rating = meta_root.css_first("h4.font-red-thunderbird")

            if rating and "class" in rating.attributes:
                stars = [
                    i
                    for i in rating.attributes["class"].split(" ")
                    if i.startswith("star-")
                ][0]
                pointer_rating: float = int(stars.split("-")[-1]) / 10
            royal_data["fiction"]["rating"] = pointer_rating
            # Chapter Title
            chapter_title = meta_root.css_first("h1.font-white.break-word")
            if chapter_title:
                royal_data["chapter"]["title"] = chapter_title.text(strip=True)

            # Chapter ID
            chapter_id = nodes.css_first(
                'a.red-thunderbird[href*="/report/chapter/" i]'
            )
            if chapter_id:
                royal_data["chapter"]["id"] = int(
                    chapter_id.attributes["href"].split("/")[-1]
                )
            # Next & prev chapter selectors
            next_chapter = nodes.css_first(
                'div.row.nav-buttons > .col-lg-offset-6 > a[href*="/fiction/"]'
            )
            if next_chapter:
                if not next_chapter.attributes["href"].split("/")[-2].isdigit():
                    logger.warning(next_chapter.attributes["href"])
                    whomst = int(
                        next_chapter.attributes["href"].split("/")[3].split("?")[0]
                    )
                    royal_data["chapter"]["next"] = whomst
                    royal_data["chapter"]["next_info"] = "Expect missing"
                else:
                    royal_data["chapter"]["next"] = int(
                        next_chapter.attributes["href"].split("/")[-2]
                    )
                    royal_data["chapter"]["next_info"] = ""
            else:
                royal_data["chapter"]["next"] = None

            prev_chapter = nodes.css_first(
                'div.row.nav-buttons > div.col-xs-6.col-md-4.col-lg-3.col-xl-2 > a[href*="/fiction/"]'
            )
            if prev_chapter:
                if not prev_chapter.attributes["href"].split("/")[-2].isdigit():
                    logger.warning(prev_chapter.attributes["href"])
                    whomst = int(
                        prev_chapter.attributes["href"].split("/")[3].split("?")[0]
                    )
                    royal_data["chapter"]["prev"] = whomst
                    royal_data["chapter"]["prev_info"] = "Expect missing"
                else:
                    royal_data["chapter"]["prev"] = int(
                        prev_chapter.attributes["href"].split("/")[-2]
                    )
                    royal_data["chapter"]["prev_info"] = ""
            else:
                royal_data["chapter"]["prev"] = None
        else:
            logger.warning("? No novel author?")

        # meta["html"] = chapter_data.matches[0].html

        md_text = md.convert(chapter_data.matches[0].html)
        royal_data["html"] = chapter_data.matches[0].html
        royal_data["text"] = md_text.replace("  \n  \n", "\n\n").strip()
        f2.write(encoder.encode(royal_data) + b"\n")