qidianunderground.py

# -*- coding: utf-8 -*-
"""
Decryptor: https://github.com/Pioverpie/privatebin-api/blob/master/privatebinapi/download.py
"""
import logging
import re
from datetime import datetime
from urllib.parse import urlsplit

import regex

from lncrawl.core.crawler import Crawler

logger = logging.getLogger(__name__)

novel_list_url = "https://toc.qidianunderground.org/api/v1/pages/public"
chapter_list_url = "https://toc.qidianunderground.org/api/v1/pages/public/%s/chapters"
chapter_body_url = "/?pasteid=%s"


class QidianComCrawler(Crawler):
    base_url = [
        "https://toc.qidianunderground.org/",
        "https://tocqidianunderground.blogspot.com/",
    ]

    def initialize(self):
        self.init_executor(1)
        self.chapter_cache = {}
        self.set_header("Accept", "application/json")

    @property
    def novel_list(self):
        if not hasattr(self, "_novel_list"):
            data = self.get_json(novel_list_url)
            self._novel_list = {x["ID"]: x for x in data}
        return self._novel_list

    def search_novel(self, query):
        query = query.strip().lower()
        spaces = len(query.split(" "))
        query = regex.compile("(%s){e<=%d}" % (query, spaces))

        results = []
        for novel in self.novel_list.values():
            m = query.search(novel["Name"].lower())
            if m:
                last_update = datetime.fromtimestamp(novel["LastUpdated"])
                last_update = last_update.strftime("%Y-%m-%d %I:%M:%S %p")
                results.append(
                    {
                        "title": novel["Name"],
                        "url": chapter_list_url % novel["ID"],
                        "info": "Last Updated: %s" % last_update,
                        "score": sum(len(x) for x in m.groups()),
                    }
                )
        return list(sorted(results, key=lambda x: -x["score"]))[:10]

    def read_novel_info(self):
        if self.novel_url.startswith("https://tocqidianunderground"):
            soup = self.get_soup(self.novel_url)
            meta = soup.select_one('meta[property="og:title"]')
            assert meta, "No title found"
            data = self.search_novel(meta["content"])
            assert len(data) > 0, "No such novel found"
            self.novel_url = data[0]["url"]

        novel_id = self.novel_url.split("/")[-2]
        self.novel_title = self.novel_list[novel_id]["Name"]

        data = self.get_json(self.novel_url)
        for vol_id, item in enumerate(data, 1):
            start_ch, final_ch = re.findall(r"(\d+) - (\d+)", item["Text"])[0]
            self.volumes.append(
                {
                    "id": vol_id,
                    "title": "Chapters %s - %s" % (start_ch, final_ch),
                }
            )
            for j in range(int(start_ch), int(final_ch) + 1):
                self.chapters.append(
                    {
                        "id": j,
                        "volume": vol_id,
                        "url": item["Href"],
                    }
                )

    def download_chapter_body(self, chapter):
        from lncrawl.utils.pbincli import PasteV2

        url_data = urlsplit(chapter["url"])
        pasteHost = url_data.scheme + "://" + url_data.netloc
        pasteId = url_data.query
        passphrase = url_data.fragment

        if pasteId in self.chapter_cache:
            soup = self.chapter_cache[pasteId]
        else:
            data = self.get_json(pasteHost + (chapter_body_url % pasteId))
            paste = PasteV2()
            paste.setHash(passphrase)
            paste.loadJSON(data)
            paste.decrypt()
            soup = self.make_soup(paste.getText())
            self.chapter_cache[pasteId] = soup

        a = soup.select_one('#toc a[href*="chapter-%d"]' % chapter["id"])
        chapter["title"] = a.text.strip()
        logger.debug(chapter["title"])

        logger.debug("Chapter Id: %s", a["href"])
        contents = soup.find("div", attrs={"id": a["href"][1:]})
        contents = contents.find("div", attrs={"class": "well"})
        for bad in contents.select("h2, br"):
            bad.extract()

        body = contents.text.split("\n\n")
        return "<p>" + "</p><p>".join(body) + "</p>"