mrakopedia.net Calibre recipe

mrakopedia.recipe

#!/usr/bin/env python3
# vim: ft=python sw=4 sts=4 et si

__license__ = "GPL v3"
__author__ = "hg"

from calibre.web.feeds.recipes import BasicNewsRecipe

base_url = "https://mrakopedia.net"
logo_url = base_url + "/logos/default.gif"
exclude = ["(манга)", "(комикс)", "Каталог"]


class MrakopediaRecipe(BasicNewsRecipe):
    title = "Мракопедия"
    description = "Лучшие истории Мракопедии по голосам читателей"
    category = "copypasta, creepy, horror, fiction"

    masthead_url = logo_url
    cover_url = logo_url

    language = "ru"
    encoding = "utf-8"

    ignore_duplicate_articles = {'url'}
    no_stylesheets = True
    resolve_internal_links = True
    recipe_specific_options = {
        "save_large": {
            "short": "Сохранять большие НЕ текстовые документы",
            "long": (
                "Возможные значения: yes/no. Текстовые истории сохраняются всегда, "
                "независимо от размера. Значение 'yes' дополнительно сохраняет "
                "массивные графические документы (комиксы, мангу) и второстепенные "
                "вещи вроде архивов ссылок. 'no' отбирает только текстовые посты."
            ),
            "default": "no"
        },
        "sort_by": {
            "short": "Способ сортировки историй",
            "long": "rating — сортировать по оценке; votes — по количеству голосов.",
            "default": "rating"
        },
    }

    keep_only_tags = [
        {"name": "main", "class": "tl_article"},
    ]
    remove_tags = [
        {"name": "span", "class": "mw-editsection"},
        {"name": "div", "class": "box"},
        {"name": "div", "class": "rating_box"},
    ]

    def parse_index(self):
        soup = self.index_to_soup(base_url + "/wiki/Категория:Рейтинг")
        links = soup.find_all("a", title=lambda t: t and "Рейтинг:" in t)
        index = []

        def do_page(title, href):
            pages = self.load_rating(href)
            index.append((title, pages))
            self.log("added {} pages from '{}'".format(len(pages), title))

        do_page("Общий рейтинг", "/wiki/Рейтинг:Общий_рейтинг")

        for link in links:
            [_, title] = link["title"].split(":", maxsplit=1)
            do_page(title, link["href"])

        return index

    def load_rating(self, url):
        doc = self.index_to_soup(base_url + url)
        content = doc.find(id="mw-content-text")
        table = content.find("table")
        index = []

        skip_large = self.recipe_specific_options["save_large"] != "yes"

        for row in table.find_all("tr"):
            link = row.find("a", href=True)
            if not link:
                continue

            [title, rating, votes] = [cell.text for cell in row.find_all("td")]

            if skip_large and any([sub in title for sub in exclude]):
                self.log("skipped link", title)
                continue

            index.append({
                "title": title,
                "url": base_url + link["href"] + "?reader",
                "description": "Рейтинг: {} ({} голосов)".format(rating, votes),
                "rating": int(rating.rstrip("%")),
                "votes": int(votes),
            })

        sort = self.recipe_specific_options["sort_by"]
        return sorted(index, reverse=True, key=lambda i: i[sort])