Greasy Fork is available in English.

Royal Road Download Button

Adds buttons to download Royal Road chapters

// ==UserScript==
// @name        Royal Road Download Button
// @license     MIT
// @namespace   rtonne
// @match       https://www.royalroad.com/fiction/*
// @icon        https://www.google.com/s2/favicons?sz=64&domain=royalroad.com
// @version     6.3
// @author      Rtonne
// @description Adds buttons to download Royal Road chapters
// The following @require is needed for jszip to work with @grant
// @require     data:application/javascript,window.setImmediate%20%3D%20window.setImmediate%20%7C%7C%20((f%2C%20...args)%20%3D%3E%20window.setTimeout(()%20%3D%3E%20f(args)%2C%200))%3B
// @require     https://cdn.jsdelivr.net/npm/[email protected]
// @require     https://cdn.jsdelivr.net/npm/[email protected]
// @require     https://update.greasyfork.org/scripts/498119/1399005/setupCommands.js
// @run-at      document-end
// @grant       GM.registerMenuCommand
// @grant       GM.unregisterMenuCommand
// @grant       GM.getValue
// @grant       GM.setValue
// @grant       GM.xmlHttpRequest
// ==/UserScript==

const FICTION_REGEX = new RegExp(
  /^https:\/\/www.royalroad.com\/fiction\/\d+?\/[^\/]+$/
);
const CHAPTER_REGEX = new RegExp(
  /^https:\/\/www.royalroad.com\/fiction\/\d+?\/[^\/]+\/chapter\/\d+?\/[^\/]+$/
);
// Not sure if this is the correct name for it, but it's an exception I found in
// https://www.royalroad.com/fiction/chapter/1671376?fid=52639&fslug=edge-cases-complete
const CORRUPTED_CHAPTER_REGEX = new RegExp(
  /^https:\/\/www.royalroad.com\/fiction\/chapter\/\d+?([^\d\/][^\/]*)?$/
);
const PARSER = new DOMParser();

/** @type {_Command[]} */
const command_list = [
  {
    type: "radio",
    id: "filename_prefix",
    default_value: "publish_date",
    radios: [
      {
        value: "chapter_id",
        id: "filename_prefix_chapter_id",
        text: "Chapter ID as the file prefix.",
      },
      {
        value: "publish_date",
        id: "filename_prefix_publish_date",
        text: "Chapter's publish date as the file prefix.",
      },
    ],
  },
  {
    type: "toggle",
    id: "download_images",
    text: "Include images in the download.",
  },
];
setupCommands(command_list);

if (FICTION_REGEX.test(window.location.href)) {
  // If current page is a fiction page
  setupFictionPageDownload();
} else if (
  CHAPTER_REGEX.test(window.location.href) ||
  CORRUPTED_CHAPTER_REGEX.test(window.location.href)
) {
  // If current page is a chapter page
  setupChapterPageDownload();
}

async function setupFictionPageDownload() {
  const chapter_metadata_list = await fetchChapterMetadataList();

  if (!chapter_metadata_list.length) {
    return;
  }

  const royalroad_button_computed_style = getComputedStyle(
    document.querySelector("a.button-icon-large")
  );

  // The page has multiple sets of buttons for different widths
  const royalroad_3_button_rows = document.querySelectorAll(
    "div.row.reduced-gutter"
  );

  for (const button_row of royalroad_3_button_rows) {
    const container = document.createElement("div");
    container.style.marginBottom = "10px";
    button_row.after(container);

    const button = document.createElement("a");
    button.className =
      "button-icon-large rtonne-royalroad-download-button-fiction-button";
    button.style.marginBottom = "0";
    container.append(button);

    const progress_bar = document.createElement("div");
    progress_bar.style.position = "absolute";
    progress_bar.style.top = `calc(${royalroad_button_computed_style.height} - ${royalroad_button_computed_style.borderBottomWidth})`;
    progress_bar.style.left = "0";
    progress_bar.style.height =
      royalroad_button_computed_style.borderBottomWidth;
    progress_bar.style.background = getComputedStyle(
      document.querySelector("a.btn-primary")
    ).backgroundColor;
    progress_bar.style.width = "0";
    progress_bar.className = "rtonne-royalroad-download-button-progress-bar";
    button.append(progress_bar);

    const button_icon = document.createElement("i");
    button_icon.className = "fa fa-download";
    button.append(button_icon);

    const button_text = document.createElement("span");
    button_text.innerText = "Download Chapters";
    button_text.className = "center";
    button.append(button_text);

    const form = document.createElement("div");
    form.className = "icon-container";
    form.style.padding = "5px";
    form.style.marginLeft = "5px";
    form.style.marginRight = "5px";
    container.append(form);

    const start_select_label = document.createElement("span");
    start_select_label.innerText = "From:";
    start_select_label.className = "tip";
    start_select_label.style.position = "unset";
    form.append(start_select_label);

    const start_select = document.createElement("select");
    start_select.className =
      "form-control rtonne-royalroad-download-button-start-select";
    start_select.style.marginBottom = "5px";
    form.append(start_select);

    const end_select_label = document.createElement("span");
    end_select_label.innerText = "To:";
    end_select_label.className = "tip";
    end_select_label.style.position = "unset";
    form.append(end_select_label);

    const end_select = document.createElement("select");
    end_select.className =
      "form-control rtonne-royalroad-download-button-end-select";
    form.append(end_select);

    for (const [index, chapter_metadata] of chapter_metadata_list.entries()) {
      const option = document.createElement("option");
      option.value = index;
      option.innerText = chapter_metadata.title;
      start_select.append(option);
      end_select.append(option.cloneNode(true));
    }

    start_select.firstChild.setAttribute("selected", "selected");
    end_select.lastChild.setAttribute("selected", "selected");

    button.addEventListener("click", () => {
      const start_index = Number(start_select.value);
      const end_index = Number(end_select.value);
      const chosen_chapters_metadata_list = chapter_metadata_list.slice(
        start_index,
        end_index + 1
      );
      let before_metadata;
      if (start_index > 0) {
        before_metadata = chapter_metadata_list[start_index - 1];
      }
      let after_metadata;
      if (end_index + 1 < chapter_metadata_list.length) {
        after_metadata = chapter_metadata_list[end_index + 1];
      }
      downloadChapters(
        chosen_chapters_metadata_list,
        before_metadata,
        after_metadata
      );
    });

    start_select.addEventListener("change", () => {
      const all_start_selects = document.querySelectorAll(
        "select.rtonne-royalroad-download-button-start-select"
      );
      for (const select of all_start_selects) {
        select.value = start_select.value;
      }
    });

    end_select.addEventListener("change", () => {
      const all_end_selects = document.querySelectorAll(
        "select.rtonne-royalroad-download-button-end-select"
      );
      for (const select of all_end_selects) {
        select.value = end_select.value;
      }
    });
  }
}

async function setupChapterPageDownload() {
  const button = document.createElement("a");
  button.className =
    "btn btn-primary rtonne-royalroad-download-button-chapter-button";

  const button_icon = document.createElement("i");
  button_icon.className = "fa fa-download";
  button.appendChild(button_icon);

  const button_text = document.createElement("span");
  button_text.innerText = " Download Chapter";
  button_text.className = "center";
  button.appendChild(button_text);

  const royalroad_fiction_page_button = document.querySelector(
    "a.btn.btn-block.btn-primary"
  );
  const royalroad_rss_button = document.querySelector("a.btn-sm.yellow-gold");

  const button_clone = button.cloneNode(true);
  button_clone.classList.add("btn-block");
  button_clone.classList.add("margin-bottom-5");
  royalroad_fiction_page_button.after(button_clone);

  button.classList.add("btn-sm");
  button.setAttribute(
    "style",
    "border-radius: 4px !important; margin-right: 5px;"
  );
  royalroad_rss_button.before(button);

  const time_element = document.querySelector("i[title='Published'] ~ time");
  const date = shortenDate(time_element.getAttribute("datetime"));

  const chapter_metadata = {
    url: window.location.href,
    date: date,
  };

  button_clone.addEventListener("click", () =>
    downloadChapters([chapter_metadata])
  );
  button.addEventListener("click", () => downloadChapters([chapter_metadata]));
}

/**
 * Get chapters' contents, pack them into a ZIP file,
 * and send a download request to the user.
 * @param {[{url: string, date: string}]} chapter_metadata_list
 * @param {{url: string, date: string}} [before_metadata] The metadata of the chapter before the first chapter in `chapter_metadata_list`. Used to create the first chapter's Previous button without an extra fetch.
 * @param {{url: string, date: string}} [after_metadata] The metadata of the chapter after the last chapter in `chapter_metadata_list`. Used to create the last chapter's Next button without an extra fetch.
 */
async function downloadChapters(
  chapter_metadata_list,
  before_metadata = undefined,
  after_metadata = undefined
) {
  const fiction_buttons = document.querySelectorAll(
    "a.rtonne-royalroad-download-button-fiction-button"
  );
  const chapter_buttons = document.querySelectorAll(
    "a.rtonne-royalroad-download-button-chapter-button"
  );
  const progress_bars = document.querySelectorAll(
    "div.rtonne-royalroad-download-button-progress-bar"
  );
  // Disable all the download buttons
  for (const button of fiction_buttons) {
    button.style.pointerEvents = "none";
    button.style.background = "#060606";
    button.style.borderBottom = "2px inset rgba(256,256,256,.1)";
  }
  for (const button of chapter_buttons) {
    button.style.pointerEvents = "none";
    button.style.opacity = ".65";
  }

  const zip = new JSZip();

  let fiction_name;
  // If we are in a corrupted chapter, the fiction name may not be in the URL
  if (CORRUPTED_CHAPTER_REGEX.test(window.location.href)) {
    const fslug = new URL(window.location.href).searchParams.get("fslug");
    if (fslug) {
      fiction_name = fslug;
    }
  } else {
    fiction_name = window.location.href.split("/")[5];
  }

  let failed_image = false;
  let failed_button = false;
  let failed_chapter_fetch = false;
  for (let index = 0; index < chapter_metadata_list.length; index++) {
    const chapter_metadata = chapter_metadata_list[index];
    const html = await fetchChapterHtml(chapter_metadata.url);
    if (!html) {
      console.warn(
        `The chapter with the url ${chapter_metadata.url} couldn't be fetched.`
      );
      failed_chapter_fetch = true;
      continue;
    }

    // In case chapter is corrupted and doesn't have the fslug parameter
    // This is done in the loop so we don't have to fetch 1 extra time
    if (!fiction_name) {
      const fiction_url = html.querySelector(
        "div.fic-header > div > div.col-lg-6 > a"
      ).href;
      const fiction_url_split = fiction_url.split("/");
      fiction_name = fiction_url_split[fiction_url_split.length - 1];
    }

    // This should never fail because both date and html are set
    const chapter_filename = await getChapterFilename(
      chapter_metadata.url,
      chapter_metadata.date,
      html
    );

    let prev_date;
    let next_date;
    if (index - 1 >= 0) {
      prev_date = chapter_metadata_list[index - 1].date;
    } else if (index - 1 === -1 && before_metadata) {
      prev_date = before_metadata.date;
    }
    if (index + 1 < chapter_metadata_list.length) {
      next_date = chapter_metadata_list[index + 1].date;
    } else if (index + 1 === chapter_metadata_list.length && after_metadata) {
      next_date = after_metadata.date;
    }

    const processed_chapter = await processChapterHtml(
      chapter_metadata.url,
      html,
      prev_date,
      next_date
    );
    for (const image of processed_chapter.images) {
      zip.file(`${fiction_name}/${image.path}`, image.blob);
    }
    if (processed_chapter.failed_image) {
      failed_image = true;
    }
    if (processed_chapter.failed_button) {
      failed_button = true;
    }

    zip.file(
      `${fiction_name}/${chapter_filename}`,
      processed_chapter.html_string
    );

    // Change the progress bars
    for (const progress_bar of progress_bars) {
      progress_bar.style.width = `${
        ((index + 1) / chapter_metadata_list.length) * 100
      }%`;
    }
  }

  if (failed_image) {
    alert("1 or more images couldn't be downloaded. More info in the console.");
  }
  if (failed_button) {
    alert(
      "1 or more Previous/Next buttons's links couldn't be changed. More info in the console."
    );
  }
  if (failed_chapter_fetch) {
    alert("1 or more chapters couldn't be fetched. More info in the console.");
  }

  await zip
    .generateAsync({
      type: "blob",
      compression: "DEFLATE",
      compressionOptions: {
        level: 9,
      },
    })
    .then((blob) => {
      saveAs(blob, fiction_name + ".zip");
    });

  // Re-enable all the download buttons, and reset the progress bars
  for (const button of fiction_buttons) {
    button.style.pointerEvents = null;
    button.style.background = null;
    button.style.borderBottom = null;
  }
  for (const button of chapter_buttons) {
    button.style.pointerEvents = null;
    button.style.opacity = null;
  }
  for (const progress_bar of progress_bars) {
    progress_bar.style.width = "0";
  }
}

/**
 * Turn RoyalRoad's chapter html into one better for offline reading.
 * @param {string} chapter_url
 * @param {HTMLHtmlElement} html
 * @param {string} [prev_date] The publish date of the previous chapter.
 * @param {string} [next_date] The publish date of the next chapter.
 * @returns {{html_string: string, images: [{path: string, blob: Blob}], failed_image: boolean, failed_button: boolean}}
 */
async function processChapterHtml(chapter_url, html, prev_date, next_date) {
  // Edit spoilers so they function the same offline
  const spoilers = html.querySelectorAll(".spoiler-new, .spoiler");
  for (const spoiler of spoilers) {
    if (spoiler.classList.contains("spoiler")) {
      spoiler.innerHTML = spoiler.querySelector(".spoiler-inner").innerHTML;
    }
    spoiler.className = "spoiler";
    const spoiler_content = document.createElement("div");
    for (const child of spoiler.children) {
      spoiler_content.append(child);
    }
    const spoiler_label = document.createElement("label");
    spoiler_label.innerText = "Spoiler";
    const spoiler_checkbox = document.createElement("input");
    spoiler_checkbox.type = "checkbox";
    spoiler_label.append(spoiler_checkbox);
    spoiler.append(spoiler_label);
    spoiler.append(spoiler_content);
  }

  // Edit the header links so they work offline
  const chapter_header = html.querySelector(
    "div.fic-header > div > div.col-lg-6"
  );
  const chapter_header_links = chapter_header.querySelectorAll("a");
  for (const link of chapter_header_links) {
    // This may seem redundant, but .href returns the whole URL
    link.setAttribute("href", link.href);
  }
  chapter_header.querySelector(
    "h1"
  ).innerHTML = `<a href="${chapter_url}" class="font-white">${
    chapter_header.querySelector("h1").innerHTML
  }</a>`;

  // Add publishing date and last edit date to the header
  // Not using html.querySelector(".profile-info > ul") so that we only get the date items
  const dates_container = document.createElement("ul");
  dates_container.className = "list-inline";
  const date_elements = html.querySelectorAll(
    ".profile-info > ul > li:has(i[title='Published']), .profile-info > ul > li:has(i[title='Last Edit'])"
  );
  for (const element of date_elements) {
    const date_time_element = element.querySelector("time");
    const date_icon_element = element.querySelector("i");
    // I could use Date.prototype.toLocaleString, but I personally prefer this
    date_time_element.innerText =
      date_time_element
        .getAttribute("datetime")
        .split(".")[0]
        .replaceAll("T", " ") + " UTC";
    date_icon_element.innerText = date_icon_element.getAttribute("title") + ":";
    dates_container.append(element);
  }
  chapter_header.append(dates_container);

  let chapter =
    getCustomChapterHeader() +
    chapter_header.outerHTML +
    '<div class="portlet">';

  let images = [];
  let failed_image = false;
  let failed_button = false;
  const chapterElements = html.querySelector("div.chapter-content").parentNode
    .children;
  for (const element of chapterElements) {
    if (
      element.classList.contains("chapter-content") ||
      element.classList.contains("author-note-portlet")
    ) {
      // Add chapter content and author notes
      if (element.classList.contains("chapter-content")) {
        // Remove unnecessary classes from <p> in the chapter content
        const paragraphs = element.querySelectorAll(":scope > p");
        for (let paragraph of paragraphs) {
          paragraph.removeAttribute("class");
        }
      }
      if (await GM.getValue("download_images", false)) {
        const img_elements = element.querySelectorAll("img");
        for (const img_element of img_elements) {
          try {
            // Not using fetch() because images may require being redirected
            // which I couldn't make work
            const request = await GM.xmlHttpRequest({
              url: img_element.src,
              responseType: "blob",
            });
            if (request.status >= 400 && request.status < 600) {
              throw request;
            }
            const image_path = turnImageUrlIntoFilepath(img_element.src);
            img_element.setAttribute(
              "onerror",
              `this.onerror=null; this.src='${img_element.src}'`
            );
            img_element.src = image_path;
            images.push({ path: image_path, blob: request.response });
          } catch (err) {
            console.warn(
              `Failed to download image with the url ${img_element.src}, on the chapter with the url ${chapter_url}`,
              err
            );
            failed_image = true;
          }
        }
      }
      chapter += element.outerHTML;
    } else if (
      element.classList.contains("nav-buttons") ||
      element.classList.contains("margin-bottom-10")
    ) {
      // Add prev/next/index buttons and make them work offline
      const buttons = element.querySelectorAll("a");
      for (const button of buttons) {
        if (button.innerText.includes("Index")) {
          button.setAttribute("href", ".");
          continue;
        }

        let chapter_filename = "unknown";
        if (button.innerText.includes("Previous")) {
          chapter_filename = await getChapterFilename(button.href, prev_date);
          if (!chapter_filename) {
            console.warn(
              `The url of the Previous button of the chapter with the url ${chapter_url} couldn't be changed.`
            );
            chapter_filename = ".";
            failed_button = true;
          }
        } else if (button.innerText.includes("Next")) {
          chapter_filename = await getChapterFilename(button.href, next_date);
          if (!chapter_filename) {
            console.warn(
              `The url of the Next button of the chapter with the url ${chapter_url} couldn't be changed.`
            );
            chapter_filename = ".";
            failed_button = true;
          }
        }

        button.setAttribute("href", chapter_filename);
      }
      chapter += element.outerHTML;
    }
  }

  chapter += getCustomChapterFooter();

  /* Regex explanation:
    Deletes whitespace at the start of each line.
    Deletes whitespace at the end of each line, including the newline.
  */
  chapter = chapter.replace(/^\s+|(\s*\n)/gm, "");

  return {
    html_string: chapter,
    images: images,
    failed_image: failed_image,
    failed_button: failed_button,
  };
}

/**
 * @param {string} chapter_url
 * @returns {Promise<HTMLHtmlElement | void>}
 */
async function fetchChapterHtml(chapter_url) {
  const html = await fetch(chapter_url, {
    credentials: "omit",
  })
    .then((response) => response.text())
    .then((text) => PARSER.parseFromString(text, "text/html"))
    .catch((error) => {
      // If this fails, other chapters can still be downloaded, so we check outside this function.
      console.error(error);
      return;
    });

  return html;
}

/**
 * Gets all the chapters from a fiction page.
 * If url is null, the current page is used.
 * @param {string} [url]
 * @returns {Promise<[{title: string, url: string, date: string}]>}
 */
async function fetchChapterMetadataList(url = null) {
  // If its not a fiction page, return an empty chapter list
  if (
    (url === null && !FICTION_REGEX.test(window.location.href)) ||
    (url !== null && !FICTION_REGEX.test(url))
  ) {
    return [];
  }

  let html;

  if (url === null) {
    // Use current page
    html = document.querySelector("html").cloneNode(true);
    url = window.location.href;
  } else {
    // Fetch new page
    html = await fetch(url, {
      credentials: "omit",
    })
      .then((response) => response.text())
      .then((text) => PARSER.parseFromString(text, "text/html"))
      .catch((error) => {
        // If we can't get the list, nothing else would work so we can just throw.
        alert(
          "An error has ocurred while fetching the chapter list. Please refresh and try again. More details in the console."
        );
        throw error;
      });
  }

  const chapter_metadata_list = [
    ...html.querySelectorAll("tr.chapter-row"),
  ].map((element) => {
    const left_link = element.querySelector("td:not(.text-right) a");
    const time_element = element.querySelector("td.text-right a time");
    return {
      title: left_link.innerText.trim(),
      url: left_link.href,
      date: shortenDate(time_element.getAttribute("datetime")),
    };
  });

  // Because javascript hides chapters from the list
  // we check and retry if chapters are hidden
  if (
    chapter_metadata_list.length === 20 &&
    html.querySelectorAll(".pagination-small").length > 0
  ) {
    return fetchChapterMetadataList(url);
  }

  return chapter_metadata_list;
}

/**
 * Gets the filename a chapter will have when downloaded.
 * @param {string} chapter_url
 * @param {string} [date] Not required, but if set there may be one less fetch request.
 * @param {HTMLHtmlElement} [html] Not required, but if set there may be one less fetch request.
 * @returns {Promise<string | void>}
 */
async function getChapterFilename(
  chapter_url,
  date = undefined,
  html = undefined
) {
  const use_chapter_id_as_prefix =
    (await GM.getValue("filename_prefix", "publish_date")) === "chapter_id";
  const is_corrupted_chapter = CORRUPTED_CHAPTER_REGEX.test(chapter_url);

  // Get the dates here if the button exists but the date is null
  // (can occur if a chapter comes out after we get the chapter metadata list)
  // Will also be used for the chapter page button
  let filename_prefix;
  if (use_chapter_id_as_prefix) {
    if (is_corrupted_chapter) {
      const chapter_url_split = chapter_url.split("?")[0].split("/");
      const chapter_id = chapter_url_split[chapter_url_split.length - 1];

      filename_prefix = chapter_id;
    } else {
      const chapter_url_split = chapter_url.split("/");
      const chapter_id = chapter_url_split[chapter_url_split.length - 2];

      filename_prefix = chapter_id;
    }
  } else {
    if (!date) {
      if (!html) {
        html = await fetchChapterHtml(chapter_url);
        // If fetching the html failed, this function also fails
        // but throw outside for extra context
        if (!html) {
          return;
        }
      }
      const time_element = html.querySelector("i[title='Published'] ~ time");
      date = shortenDate(time_element.getAttribute("datetime"));
    }
    filename_prefix = shortenDate(date);
  }

  if (is_corrupted_chapter) {
    return `${filename_prefix}.html`;
  }

  const chapter_url_split = chapter_url.split("/");
  const chapter_name = chapter_url_split[chapter_url_split.length - 1];
  return `${filename_prefix}_${chapter_name}.html`;
}

/**
 * Turns any image URL and turns it into a file path.
 * @param {string} url
 * @returns {string}
 */
function turnImageUrlIntoFilepath(url) {
  // Getting the filename from a URL https://stackoverflow.com/a/36756650
  const filename = (url.match(/^\w+:(\/+([^\/#?\s]+)){2,}/) || [])[2] || "";

  // Put filename at the end of the URL so that all other parts (including params) create folders but the filename is correct
  // Make path compatible with Windows/Android
  // Make folder names have a maximum length of 100
  // Remove repeated "/"
  const filepath = (
    "images/" +
    url
      .replace(filename, "")
      .replaceAll(/[\\:*?"<>|]/g, "/")
      .replaceAll(/([^\/]{100})()/g, "$1/") +
    "/" +
    filename
  ).replaceAll(/[\/]+/g, "/");

  return filepath;
}

/**
 * Turns a date with a format like "2023-06-21T22:02:45.0000000Z"
 * into something like "20230621220245".
 * @param {string} date
 * @returns {string}
 */
function shortenDate(date) {
  if (!date) {
    return "";
  }
  return date
    .split(".")[0]
    .replaceAll("-", "")
    .replaceAll(":", "")
    .replaceAll("T", "");
}

function getCustomChapterHeader() {
  const header = /*html*/ `<!DOCTYPE html>
<html lang="en">
<head>
<meta charset="UTF-8" />
<meta name="viewport" content="width=device-width, initial-scale=1.0" />
<style>
.portlet-body p,
p {
  margin-top: 0;
}
body {
  background: #181818;
  font-family: Open Sans, open-sans, Helvetica Neue, Helvetica, Roboto,
    Arial, sans-serif;
  line-height: 1.42857143;
  font-size: 16px;
  margin: 0;
}
.font-white {
  color: #fff !important;
}
.portlet {
  background: #131313;
  border: 1px solid hsla(0, 0%, 100%, 0.1);
  color: hsla(0, 0%, 100%, 0.8);
  padding: 1em 20px 0;
  margin: 10px 0;
  display: flex;
  flex-direction: column;
}
.author-note-portlet {
  background: #393939;
  color: hsla(0, 0%, 100%, 0.8);
  border: 0;
  padding: 0 10px 10px;
  margin: 0 0 1em;
}
.portlet-title {
  border-bottom: 0;
  margin-bottom: 10px;
  min-height: 41px;
  padding: 0;
  margin-left: 15px;
}
.caption {
  padding: 16px 0 2px;
  display: inline-block;
  float: left;
  font-size: 18px;
  line-height: 18px;
}
.uppercase {
  text-transform: uppercase !important;
}
.bold {
  font-weight: 700 !important;
}
a {
  color: #337ab7;
  text-shadow: none;
  text-decoration: none;
}
.portlet-body {
  padding: 10px 15px;
}
p {
  margin-bottom: 1em;
}
.col-md-5 {
  min-height: 1px;
  background: #2a3642;
  margin-left: -15px;
  margin-right: -15px;
  padding: 10px;
}
.text-center {
  text-align: center;
}
.container {
  margin-left: auto;
  margin-right: auto;
  padding-left: 15px;
  padding-right: 15px;
  width: "100%";
}
.col-md-5 > *,
.col-md-5 > * > * {
  font-weight: 300;
  margin: 10px 0;
}
table {
  background: #004b7a;
  border: none;
  border-collapse: separate;
  border-spacing: 2px;
  box-shadow: 1px 1px 1px rgba(0, 0, 0, 0.75);
  margin: 10px auto;
  width: 90%;
}
table td {
  background: rgba(0, 0, 0, 0.1);
  border: 1px solid hsla(0, 0%, 100%, 0.25) !important;
  color: #ccc;
  margin: 3px;
  padding: 5px;
}
.btn-primary {
  box-shadow: none;
  outline: none;
  line-height: 1.44;
  background-color: #337ab7;
  color: #fff;
  padding: 6px 0;
  text-align: center;
  display: inline-block;
  font-size: 14px;
  font-weight: 400;
  border: 1px solid #2e6da4;
}
.btn-primary[disabled] {
  cursor: not-allowed;
  opacity: 0.65;
}
.col-xs-12 {
  width: 100%;
}
.col-xs-6 {
  width: 50%;
  position: relative;
  float: left;
}
.visible-xs,
.visible-xs-block {
  display: none;
}
.col-xs-4 {
  width: 33.33333333%;
  margin: 0;
}
.row {
  display: flex;
  margin-bottom: 1em;
}
@media (min-width: 992px) {
  .container {
    width: 970px;
  }
  .col-md-4 {
    width: 33.33333333%;
  }
  .col-md-offset-4 {
    margin-left: 33.33333333%;
  }
}
@media (min-width: 1200px) {
  .container {
    width: 1170px;
  }
  .col-lg-3 {
    width: 25%;
  }
  .col-lg-offset-6 {
    margin-left: 50%;
  }
}
.spoiler > label > input {
  position: absolute;
  opacity: 0;
  z-index: -1;
}
.spoiler > label {
  font-weight: bold;
  cursor: pointer;
}
.spoiler > label::after {
  content: "Show";
  background: #2c2c2c;
  border: 1px solid rgba(61, 61, 61, 0.31);
  color: hsla(0, 0%, 100%, 0.8) !important;
  font-size: 12px;
  padding: 1px 5px;
  font-weight: 400;
  margin-left: 5px;
}
.spoiler > label:has(> input:checked)::after {
  content: "Hide";
}
.spoiler > label:hover::after {
  background: #3e3e3e;
}
.spoiler > div {
  display: none;
  margin-top: 20px;
}
.spoiler > label:has(> input:checked) ~ div {
  display: block;
}
img {
  height: auto !important;
  max-width: 100%;
}
.list-inline {
  list-style: none;
  padding-left: 0;
  color: hsla(0, 0%, 100%, 0.8);
}
.list-inline > li {
  display: inline-block;
  padding-left: 15px;
  padding-right: 15px;
  margin: 0;
}
hr {
  margin: 20px 0;
  border: 0;
  border-top: 1px solid hsla(0, 0%, 100%, .3);
}
</style>
</head>
<body>
<div class="container">`;

  /* Regex explanation:
  Deletes whitespace at the start of each line.
  Deletes whitespace at the end of each line, including the newline.
  Deletes whitespace before "{", "(", "}", ")", "/", ":", ",", "<", ">".
  ("?=" means that the character ahead is found but not selected for deletion)
  Deletes whitespace after "{", "(", "}", ")", "/", ":", ",", "<", ">".
  ("?<=" means that the character behind is found but not selected for deletion)
  */
  const no_newlines_header = header.replace(
    /^\s+|(\s*\n)|(\s+(?=[\{\(\}\)\/:,<>]))|((?<=[\{\(\}\)\/:,<>])\s+)/gm,
    ""
  );
  return no_newlines_header;
}

function getCustomChapterFooter() {
  return "</div></div></body></html>";
}