pdf.js
Overview
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.
Getting started
Online demo
For an online demo, visit:
This demo provides an interactive interface for displaying and browsing PDFs
using the pdf.js API.
Extension
A Firefox extension is also available:
However, note that the extension might not reflect the latest source in our master branch.
Getting the code
To get a local copy of the current code, clone it using git:
$ git clone git://github.com/andreasgal/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
Learning
Here are some initial pointers to help contributors get off the ground.
Additional resources are available in a separate section below.
Introductory video
Check out the presentation by our contributor Julian Viereck on the inner
workings of PDF and pdf.js:
Hello world
For a "hello world" example, take a look at:
This example illustrates the bare minimum ingredients for integrating pdf.js
in a custom project.
Contributing
pdf.js is a community-driven project, so contributors are always welcome.
Simply fork our repo and contribute away. A great place to start is our
open 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:
test/resources/browser_manifests/browser_manifest.json
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:
test/test_manifest.json
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:
Additional resources
Our demo site is here:
You can read more about pdf.js here:
- http://andreasgal.com/2011/06/15/pdf-js/
- http://blog.mozilla.com/cjones/2011/06/15/overview-of-pdf-js-guts/
Talk to us on IRC:
- #pdfjs on irc.mozilla.org
Join our mailing list:
Subscribe either using lists.mozilla.org or Google Groups:
- https://lists.mozilla.org/listinfo/dev-pdf-js
- https://groups.google.com/group/mozilla.dev.pdf-js/topics
Follow us on twitter: @pdfjs
PDF-related resources
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.