function _ctPdfViewport(page, scale) {
        try { return page.getViewport({ scale: scale }); }
        catch (e) { return page.getViewport(scale); }
    }

    function processFile(blob, fileName) {
        var language = (($('.extra-input input').first().val() || 'eng') + '').trim().toLowerCase();
        if (!language) language = 'eng';

        Promise.all([
            loadScriptPromise('https://cdn.jsdelivr.net/npm/pdf-lib@1.17.1/dist/pdf-lib.min.js'),
            new Promise(function(resolve) {
                if (window.PDFJS && PDFJS.getDocument) { resolve(); return; }
                var s = document.createElement('script'); s.src = '/js/pdf.js'; s.onload = resolve; s.onerror = resolve;
                document.head.appendChild(s);
            }),
            loadScriptPromise('https://cdn.jsdelivr.net/npm/tesseract.js@5.1.1/dist/tesseract.min.js')
        ]).then(async function() {
            try {
                var inputBytes = new Uint8Array(await blob.arrayBuffer());
                PDFJS.workerSrc = '/js/pdf.worker.js';
                var loadingTask;
                try { loadingTask = PDFJS.getDocument(inputBytes); }
                catch (e) { loadingTask = PDFJS.getDocument({ data: inputBytes }); }
                var srcPdf = (loadingTask && loadingTask.promise) ? await loadingTask.promise : await loadingTask;

                var outPdf = await PDFLib.PDFDocument.create();
                var font = await outPdf.embedFont(PDFLib.StandardFonts.Helvetica);
                var renderScale = 2.0;

                for (var p = 1; p <= srcPdf.numPages; p++) {
                    var srcPage = await srcPdf.getPage(p);
                    var viewport = _ctPdfViewport(srcPage, renderScale);
                    var canvas = document.createElement('canvas');
                    canvas.width = Math.ceil(viewport.width);
                    canvas.height = Math.ceil(viewport.height);
                    await srcPage.render({ canvasContext: canvas.getContext('2d'), viewport: viewport }).promise;

                    var pngBlob = await new Promise(function(resolve) { canvas.toBlob(resolve, 'image/png'); });
                    if (!pngBlob) throw new Error('Could not render PDF page image for OCR.');
                    var pngBytes = new Uint8Array(await pngBlob.arrayBuffer());
                    var embeddedImage = await outPdf.embedPng(pngBytes);

                    var outPage = outPdf.addPage([viewport.width, viewport.height]);
                    outPage.drawImage(embeddedImage, { x: 0, y: 0, width: viewport.width, height: viewport.height });

                    var result = await Tesseract.recognize(canvas.toDataURL('image/png'), language);
                    var words = (result && result.data && result.data.words) ? result.data.words : [];
                    for (var w = 0; w < words.length; w++) {
                        var word = words[w];
                        var txt = (word.text || '').trim();
                        if (!txt) continue;
                        var b = word.bbox || {};
                        var x = b.x0 || 0;
                        var yTop = b.y0 || 0;
                        var yBottom = b.y1 || yTop;
                        var size = Math.max(6, Math.min(72, (yBottom - yTop) || 10));
                        var y = viewport.height - yBottom;
                        outPage.drawText(txt, {
                            x: x,
                            y: y,
                            size: size,
                            font: font,
                            opacity: 0
                        });
                    }
                }

                var outBytes = await outPdf.save({ useObjectStreams: true });
                var base = (fileName || 'document').replace(/\.pdf$/i, '');
                add_file_output(URL.createObjectURL(new Blob([outBytes], { type: 'application/pdf' })), base + '-ocr.pdf');
            } catch (err) {
                alert('Could not OCR this PDF. Try a smaller file, fewer pages, or language code "eng".');
            }
        }).catch(function() {
            alert('Could not load OCR/PDF libraries in your browser.');
        });
    }







var _loadedScripts = {};
function loadScriptPromise(url) {
    if (_loadedScripts[url]) return _loadedScripts[url];
    _loadedScripts[url] = new Promise(function (resolve, reject) {
        var s = document.createElement('script');
        s.src = url;
        s.onload = resolve;
        s.onerror = reject;
        document.head.appendChild(s);
    });
    return _loadedScripts[url];
}

function replaceAll(find, replace, str) {
    return str.replace(new RegExp(find, 'g'), replace);
}

function beautify(str) {


    var result = '';

    var length = str.length;

    var i = 0;

    var braceCountLeft = 0;
    var braceCountRight = 0;

    var withinQuotes = false;

    while (i < length) {

        var c = str[i];

        if (c == '"' && (i == 0 || c[i - 1] != '\\')) {
            // non-escaped quotes
            withinQuotes = !withinQuotes;
        }

        if (!withinQuotes && (c == '}' || c == '{' || c == ',')) {

            console.log('Start####' + result);

            // look back and remove carriage returns and whitespace that are already there

            var resultIndex = result.length - 1;
            while (resultIndex >= 0 && (result[resultIndex] == ' ' || result[resultIndex] == '\r' || result[resultIndex] == '\n' || result[resultIndex] == '\t')) {

                resultIndex = resultIndex - 1;
                result = result.substr(0, resultIndex + 1);

                console.log('char ' + result[resultIndex] + '-----' + result + 'zzz ' + result.length + ' ' + resultIndex);

            }


            if (c == '{') {
                braceCountLeft++;
                result += c + '\r' + GetTabs(braceCountLeft - braceCountRight);
            } else if (c == '}') {
                braceCountRight++;
                // precede with carriage return
                result += '\r' + GetTabs(braceCountLeft - braceCountRight) + c;
            } else if (c == ',') {
                result += c + '\r' + GetTabs(braceCountLeft - braceCountRight);
            }

            var nextChar = '';

            // advance through whitespace and remove carriage returns that are already there
            while (i < length && (str[i + 1] == ' ' || str[i + 1] == '\r' || str[i + 1] == '\n' || str[i + 1] == '\t')) {
                i++;
            }

        } else {

            result += str[i];

        }


        i++;

    }

    return result;

}

function GetTabs(count) {

    var result = '';

    for (var i = 0; i < count; i++) {
        result += '    ';
    }

    return result;

}