pdf.js is an HTML5 technology experiment that explores building a faithful and efficient Portable Document Format (PDF) renderer without native code assistance.
pdf.js is community-driven and supported by Mozilla Labs. Our goal is to create a general-purpose, web standards-based platform for parsing and rendering PDFs, and eventually release a PDF reader extension powered by pdf.js. Integration with Firefox is a possibility if the experiment proves successful.
For an online demo, visit:
This demo provides an interactive interface for displaying and browsing PDFs using the pdf.js API.
An up-to-date Firefox extension is also available:
(The above link is updated upon every merge to our master branch).
For an experimental Chrome extension, get the code as explained below and issue
Then open Chrome with the flag
--enable-experimental-extension-apis, go to
Tools > Extension
and load the (unpackaged) extension from the directory
Getting the code
To get a local copy of the current code, clone it using git:
$ git clone git://github.com/mozilla/pdf.js.git pdfjs $ cd pdfjs
Next, you need to start a local web server as some browsers don't allow opening PDF files for a file:// url:
$ make server
If everything worked out, you can now serve
You can also view all the test pdf files on the right side serving
In order to bundle all
src/ files into a final
This will generate the file
build/pdf.js that can be included in your final project. (WARNING: That's a large file! Consider minifying it).
Here are some initial pointers to help contributors get off the ground. Additional resources are available in a separate section below.
For a "hello world" example, take a look at:
This example illustrates the bare minimum ingredients for integrating pdf.js in a custom project.
Check out the presentation by our contributor Julian Viereck on the inner workings of PDF and pdf.js:
pdf.js is a community-driven project, so contributors are always welcome. Simply fork our repo and contribute away. Good starting places for picking a bug are the top error messages and TODOs in our corpus report:
and of course our open Github issues:
For better consistency and long-term stability, please do look around the code and try to follow our conventions. More information about the contributor process can be found on the contributor wiki page.
If you don't want to hack on the project or have little spare time, you still can help! Just open PDFs in the online demo and report any breakage in rendering.
Our Github contributors so far:
You can add your name to it! :)
Running the tests
pdf.js comes with browser-level regression tests that allow one to probe whether it's able to successfully parse PDFs, as well as compare its output against reference images, pixel-by-pixel.
To run the tests, first configure the browser manifest file at:
Sample manifests for different platforms are provided in that directory.
To run all the bundled tests, type:
$ make test
and cross your fingers. Different types of tests are available, see the test manifest file at:
The test type
eq tests whether the output images are identical to reference
images. The test type
load simply tests whether the file loads without
raising any errors.
Running tests through our bot
If you are a reviewer, you can use our remote bot to issue comprehensive tests against reference images before merging pull requests.
See the bot repo for details:
Gallery of user projects and modifications:
You can read more about pdf.js here:
Talk to us on IRC:
- #pdfjs on irc.mozilla.org
Join our mailing list:
Subscribe either using lists.mozilla.org or Google Groups:
Follow us on twitter: @pdfjs
A really basic overview of PDF is described here:
A more detailed file example:
The PDF specification itself is an ISO and not freely available. However, there is a "PDF Reference" from Adobe:
Recommended chapters to read: "2. Overview", "3.4 File Structure", "4.1 Graphics Objects" that lists the PDF commands.