flyinglines.py

# -*- coding: utf-8 -*-
import logging
from datetime import datetime
from hashlib import md5
from urllib.parse import urlparse

from lncrawl.core.crawler import Crawler

logger = logging.getLogger(__name__)


class FlyingLinesCrawler(Crawler):
    base_url = "https://www.flying-lines.com/"

    def read_novel_info(self):
        logger.debug("Visiting %s", self.novel_url)
        soup = self.get_soup(self.novel_url)

        possible_title = soup.select_one(".novel-info .title h2")
        assert possible_title, "No novel title"
        self.novel_title = possible_title.text
        logger.info("Novel title: %s", self.novel_title)

        self.novel_cover = self.absolute_url(
            soup.select_one(".novel .novel-thumb img")["data-src"]
        )
        logger.info("Novel cover: %s", self.novel_cover)

        authors = [x.text.strip() for x in soup.select(".novel-info ul.profile li")]
        self.novel_author = ", ".join(authors)
        logger.info("%s", self.novel_author)

        self.novel_id = urlparse(self.novel_url).path.split("/")[2]
        logger.info("Novel id: %s", self.novel_id)

        for a in soup.select("ul.volume-chapters li a"):
            chap_id = int(a["data-chapter-number"])
            vol_id = 1 + (chap_id - 1) // 100
            if len(self.chapters) % 100 == 0:
                self.volumes.append({"id": vol_id})
            self.chapters.append(
                {
                    "id": chap_id,
                    "volume": vol_id,
                    "title": a.text.strip(),
                    "url": self.absolute_url(a["href"]),
                }
            )

    def download_chapter_body(self, chapter):
        # How to get the key?
        # - check `app.js` (Sources/cdn.flying-lines.com/front/../js/app.js)
        # - locate the mapped value of `./what` (e.g.: "./what": 57,)
        # - locate the mapped valued function. (e.g: search `57:`)
        # - Copy key from the function. (return "4ec781a70c3f11ea9b46f23c918347ac")
        key = "4ec781a70c3f11ea9b46f23c918347ac"

        # How to find the sign generation method?
        # - check `app.js`
        # - search the mapped value of `./nobody`
        # - locate the mapped valued function. (e.g: search `48:`)
        time = str(int(datetime.now().timestamp() * 1000))
        sign = "chapterNum=%s&novelPath=%s&time=%s&webdriver=0&key=%s" % (
            chapter["id"],
            self.novel_id,
            time,
            key,
        )
        sign = md5(sign.encode("utf8")).hexdigest()

        # How to generate the URL?
        # - check `app.js`
        # - search for `e.data.meta`
        url = (
            "%s/h5/novel/%s/%s?accessToken=&isFirstEnter=1&webdriver=0&time=%s&sign=%s"
        )
        url = url % (self.home_url.strip("/"), self.novel_id, chapter["id"], time, sign)

        logger.info("Downloading %s", url)
        data = self.post_json(
            url,
            headers={
                "Origin": self.home_url.strip("/"),
                "Referer": self.home_url.strip("/") + "/novel/" + self.novel_id,
            },
        )

        # How to find the decryption method?
        # - check `app.js`
        # - search for `TurnCharIntoString:` function
        encrypted = data["data"]["content"]
        n = data["data"]["nowTime"]
        even_shift = int(n[-1])
        odd_shift = int(n[-2])
        content = ""
        for i, ch in enumerate(encrypted):
            if i % 2 == 0:
                content += chr(ord(ch) - even_shift)
            else:
                content += chr(ord(ch) - odd_shift)
        return content