VitalSource web book page scraper

1-README.md

VitalSource web book page scraper

This pair of userscripts (to be used with any usescript manager such as ViolentMonkey) allow scraping books from the VitalSource Bookshelf web reader.

This allows creating a PDF for offline reading with free-software readers such as Calibre.

The "inner" script hooks into the book page nested iframe; it detects when a new page image is loaded and it automatically starts a browser download.

The "outer" script hooks into the main reader page and adds a "Scrape" button which automatically goes to the next page when the inner script has successfully downloaded a page.

Usage

1. Install both userscripts
2. Open the book
3. Wait for the first two downloads (the web reader will prefetch one page in the background while you read)
4. Click "Scrape this motherfucker" to start auto-advancing
5. Keep an eye on the reader as it may occasionally present a CAPTCHA; the scraping should resume once you solve it

Additional tools

• img2pdf: to merge all the pages into a PDF
• https://tinyjpg.com/: to compress the images without reducing visual quality
• OCRmyPDF: to OCR your PDF
• Krasjet/pdf.tocgen: to add a ToC
• lovasoa/pagelabels-py: to map custom page numbers

2-vitalsource-scraper-inner.userscript.js

// ==UserScript==
// @name         Scrape VitalSource (inner)
// @namespace    http://tampermonkey.net/
// @version      2024-02-16
// @description  try to take over the world!
// @author       You
// @match        https://jigsaw.vitalsource.com/books/*
// @icon         https://www.google.com/s2/favicons?sz=64&domain=vitalsource.com
// @sandbox      JavaScript
// @grant        unsafeWindow
// @grant        GM_download
// ==/UserScript==

function download(url) {
    console.log("inner frame: downloading:", url);
    GM_download({
        url: url,
        name: "page.jpg",
        saveAs: false,
        conflictAction: "uniquify",
        onerror: function (error) {
            console.error("inner frame: download failed:", url, error);
        },
        onload: function (response) {
            console.log("inner frame: downloaded:", url);

            window.top.postMessage(
                {
                    type: "pageImage",
                    frameUrl: window.location.href,
                    url: url,
                },
                "https://bookshelf.vitalsource.com/reader/books/*"
            );

            console.log("inner frame: message sent");
        },
    });
}

(function () {
    let lowResCount = 0;

    const run = function () {
        const b = document.querySelector('img#pbk-page');
        if (b == null) {
            console.log("inner frame: image not found");
            setTimeout(run, 100);
        } else {
            const url = b.src;

            if (url.endsWith("/800")) {
                lowResCount++;
                if (lowResCount < 25) {
                    console.log("inner frame: low res image, retrying:", url);
                    setTimeout(run, 500);
                    return;
                } else {
                    console.log("inner frame: low res image, download anyway:", url);
                }
            }

            console.log("inner frame: found image:", url);
            download(url);
        }
    };

    run();
})();

2-vitalsource-scraper.userscript.js

// ==UserScript==
// @name         Scrape VitalSource
// @namespace    http://tampermonkey.net/
// @version      2024-02-16
// @description  try to take over the world!
// @author       You
// @match        https://bookshelf.vitalsource.com/reader/books/*
// @icon         https://www.google.com/s2/favicons?sz=64&domain=vitalsource.com
// @grant        window.onurlchange
// @grant        unsafeWindow
// ==/UserScript==

function findElementByLabelText(labelText) {
    // Find all label elements in the document
    const labels = document.querySelectorAll('label');

    // Iterate through the found labels to find the one with the matching text
    for (let label of labels) {
        if (label.textContent.trim() === labelText) {
            // Read the "for" attribute of the label
            const forAttribute = label.getAttribute('for');
            if (forAttribute) {
                // Use the "for" attribute to fetch the referenced element
                return document.getElementById(forAttribute); // Return the found element
            }
            break; // Stop the search once the first matching label is found
        }
    }

    return null; // Return null if no matching label or referenced element is found
}


(function () {
    let lastUrl = null;
    let lastUrlChangedCallback = null;
    let currentPage = -1;
    let running = false;

    function stop() {
        running = false;
        lastUrlChangedCallback = null;
        currentPage = -1;

        const button = document.querySelector("#scrapeButton");
        button.innerHTML = "Scrape this motherfucker";
    }

    window.addEventListener("message", e => {
        // Check if e.data contains .type, and .type===pageImage
        if (e.data.type !== "pageImage") return;
        console.log("Page image event:", e);
        unsafeWindow.msg = e;

        lastUrl = e.data;
        if (lastUrlChangedCallback) {
            lastUrlChangedCallback();
        }
    });

    const worker = function () {
        const goToPageInput = findElementByLabelText('Go to Page');
        if (goToPageInput == null) {
            console.log("Go to page input not found!");
            return;
        }
        const pageNum = parseInt(goToPageInput.value);

        if (pageNum === currentPage) {
            console.log("Skipping duplicate event");
            return;
        }

        if (currentPage === -1) {
            currentPage = pageNum;
            if (currentPage !== 1) {
                // Ask permission to start from a page other than the first
                if (!confirm("Do you want to start from page " + currentPage + "?")) {
                    console.log("User cancelled");
                    stop();
                    return;
                }
            }
        }

        if (lastUrl == null) {
            console.log("No URL found");
            stop();
            return;
        }

        console.log("SCRAPER:", lastUrl.url);

        setTimeout(() => {
            const nextButton = document.querySelector('[aria-label="Next"]');
            if (nextButton == null) {
                console.log("Next button not found!");
                stop();
                return;
            }

            // check if disabled via the "disabled" attribute
            if (nextButton.hasAttribute("disabled")) {
                console.log("Next button is disabled");
                stop();
                return;
            }

            nextButton.click();
        }, 10);
    }

    const doStuff = function () {
        if (running) {
            console.log("Stopping scraper");
            stop();
            return;
        }

        console.log("Starting scraper, lastUrl:", lastUrl);
        const button = document.querySelector("#scrapeButton");
        running = true;
        lastUrlChangedCallback = worker;
        button.innerHTML = "Stop scraping";

        worker();
    };

    const inject = function () {
        const b = document.querySelector('[aria-label="Search across book"]');
        if (b == null) {
            console.log("Button not found, retrying in 1s");
            setTimeout(inject, 1000);
            return;
        }

        const a = document.createElement("a");
        a.id = "scrapeButton";
        a.href = "#";
        a.innerHTML = "Scrape this motherfucker";
        a.addEventListener("click", _ => doStuff());
        b.after(a);

        console.log("Injected!");
    };

    inject();
})();