Created
February 11, 2015 02:54
-
-
Save jdeng/cbfad9cb21e452127c81 to your computer and use it in GitHub Desktop.
pdf to image using pdf.js
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
<html> | |
<body> | |
<script type="text/javascript" src="//mozilla.github.io/pdf.js/build/pdf.js"></script> | |
<script type="text/javascript"> | |
var url = "https://docs.google.com/document/export?format=pdf&id=1ML11ZyyMpnAr6clIAwWrXD53pQgNR-DppMYwt9XvE6s&token=AC4w5Vg7fSWH1Hq0SgNckx4YCvnGPaScyw%3A1423618416864"; | |
var pages = [], heights = [], width = 0, height = 0, currentPage = 1; | |
var scale = 1.5; | |
function draw() { | |
var canvas = document.createElement('canvas'), ctx = canvas.getContext('2d'); | |
canvas.width = width; | |
canvas.height = height; | |
for(var i = 0; i < pages.length; i++) | |
ctx.putImageData(pages[i], 0, heights[i]); | |
document.body.appendChild(canvas); | |
} | |
PDFJS.disableWorker = true; // due to CORS | |
PDFJS.getDocument(url).then(function (pdf) { | |
getPage(); | |
function getPage() { | |
pdf.getPage(currentPage).then(function(page) { | |
console.log("Printing " + currentPage); | |
var viewport = page.getViewport(scale); | |
var canvas = document.createElement('canvas') , ctx = canvas.getContext('2d'); | |
var renderContext = { canvasContext: ctx, viewport: viewport }; | |
canvas.height = viewport.height; | |
canvas.width = viewport.width; | |
page.render(renderContext).then(function() { | |
pages.push(ctx.getImageData(0, 0, canvas.width, canvas.height)); | |
heights.push(height); | |
height += canvas.height; | |
if (width < canvas.width) width = canvas.width; | |
if (currentPage < pdf.numPages) { | |
currentPage++; | |
getPage(); | |
} | |
else { | |
draw(); | |
} | |
}); | |
}); | |
} | |
}); | |
</script> | |
</body> | |
</html> |
const convertDataUrlToFile = (dataUrl, filename) => {
return fetch(dataUrl)
.then(res => res.arrayBuffer())
.then(buf => new File([buf], filename, { type: 'image/png' }))
};
const convertFiletoDataUrl = (file) => new Promise((resolve, reject) => {
const reader = new FileReader();
reader.readAsDataURL(file);
reader.onload = () => resolve(reader.result.normalize());
reader.onerror = error => reject(error);
});
const convertPdfPageToImage = (pdf, pageNumber, fileName) => new Promise(resolve => {
pdf.getPage(pageNumber).then(page => {
const viewport = page.getViewport({ scale: 1 });
const canvas = document.createElement('canvas');
const ctx = canvas.getContext('2d');
const renderContext = { canvasContext: ctx, viewport: viewport };
canvas.height = viewport.height;
canvas.width = viewport.width;
page.render(renderContext).promise.then(() => {
const dataUrl = canvas.toDataURL();
convertDataUrlToFile(dataUrl, `${fileName}-${pageNumber}.png`).then(resolve);
});
});
})
const convertPdfFilesToImages = (files) => new Promise(resolve => {
pdfjsLib.GlobalWorkerOptions.workerSrc = `//cdnjs.cloudflare.com/ajax/libs/pdf.js/${pdfjsLib.version}/pdf.worker.js`;
const worker = new pdfjsLib.PDFWorker();
const promises = Array.from(Array(files.length), (_, index) => new Promise(resolve => {
const file = files[index];
if (file.type === 'application/pdf') {
convertFiletoDataUrl(file).then(dataUrl =>
pdfjsLib.getDocument({ url: dataUrl, worker }).promise.then(pdf => {
const promises = Array.from(Array(pdf.numPages), (_, index) => new Promise(resolve => {
const pageNumber = index + 1;
convertPdfPageToImage(pdf, pageNumber, file.name).then(resolve);
}));
Promise.all(promises).then(resolve);
})
);
} else {
resolve(file);
}
}));
Promise.all(promises).then(files => resolve(files.flat()));
});
@itamardavidyan this is good script and thanks a lot for that
but i couldn't use it because image quality is very bad🤔
@agoalofalife
In this line you can change scale to enchance quality :D
const viewport = page.getViewport({ scale: 1 });
e.g.
const viewport = page.getViewport({ scale: 4 });
Using angular how to use the above code?
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Hello sir,
I'm beginner in coding world..i'm applying this script for convert pdf to image, but unfortunately, it's wont work.
I see error in console like "PDFJS is not defined" even pdf.js is integrated. what's the solution for this..Please help me to move ahead.
Thanks in advance,
Ankur