Open HTML to PDF is a pure-Java library for rendering arbitrary well-formed XML/XHTML (and even HTML5) using CSS 2.1 for layout and formatting, outputting to PDF or images.
Use this library to generated nice looking PDF documents. But be aware that you can not throw modern HTML5+ at this engine and expect a great result. You must special craft the HTML document for this library and use it's extended CSS feature like #31 or #32 to get good results. Avoid floats near page breaks and use table layouts.
- Showcase Document - PDF
- Integration guide - get maven artifacts and code to get started.
- Documentation wiki
- Template Author Guide - PDF - DEPRECATED - Prefer wiki - Moving info to wiki
- RC18 Online Sandbox - With more samples!
- Sample Project - Pretty Resume Generator
- Uses the well-maintained and open-source (LGPL compatible) PDFBOX as PDF library, rather than iText.
- Proper support for generating accessible PDFs (Section 508, PDF/UA, WCAG 2.0).
- Proper support for generating PDF/A standards compliant PDFs.
- New, faster renderer means this project can be several times faster for very large documents.
- Better support for CSS3 transforms.
- Automatic visual regression testing of PDFs, with many end-to-end tests.
- Ability to insert pages for cut-off content.
- Built-in plugins for SVG and MathML.
- Font fallback support.
- Limited support for RTL and bi-directional documents.
- On the negative side, no support for OpenType fonts.
- Much more. See changelog below.
Open HTML to PDF is distributed under the LGPL. Open HTML to PDF itself is licensed under the GNU Lesser General Public License, version 2.1 or later, available at http://www.gnu.org/copyleft/lesser.html. You can use Open HTML to PDF in any way and for any purpose you want as long as you respect the terms of the license. A copy of the LGPL license is included as license-lgpl-2.1.txt or license-lgpl-3.txt in our distributions and in our source tree.
An exception to this is the pdf-a testing module, which is licensed under the GPL. This module is not distributed to Maven Central and is for testing only.
Open HTML to PDF uses a couple of FOSS packages to get the job done. A list of these can be found in the dependency graph.
Open HTML to PDF is based on Flying-saucer. Credit goes to the contributors of that project. Code will also be used from neoFlyingSaucer
- OPEN HTML TO PDF is tested with OpenJDK 8 and 11. We are not currently testing with Oracle JDKs. As of RC18, it requires at least Java 8 to run.
- No, you can not use it on Android.
- You should be able to use it on Google App Engine (Java 8 or greater environment). Let us know your experience.
Flowing columns are not implemented.Implemented in RC12.- No, it's not a web browser.
Test cases, failing or working are welcome, please place them
in /openhtmltopdf-examples/src/main/resources/testcases/
and run them
from /openhtmltopdf-examples/src/main/java/com/openhtmltopdf/testcases/TestcaseRunner.java.
- #372 Much improved sizing support for
img,svgandmathelements. - #344 Use PDFs in
imgtag:<img src="some.pdf" page="1" alt="Some alt text" />.
- #361 The SVG renderer now uses Batik in a more secure mode (no scripts, no external resource requests) by default. If you need the old behavior that allowed external resource requests and possibly scripts, please see the new BatikSVGDrawer constructor (only for trusted SVGs). Thanks @krabbenpuler.
- #363 Upgrade PDFBOX to 2.0.16. Thanks @rototor.
- #353 Better error handling around SVGs linked from
imgtag. Thanks @ieugen. - #342 Fixed text-justification/letter-spacing when fallback fonts are in use. Thanks @daliuss.
- #351 Improved text-justification by removing spaces at ends of lines. Thanks @halcsi.
IMPORTANT: This release was brought forward due to a CVE in PDFBOX. While not directly affecting this project (it affects parsing of untrusted PDFs), it is better not to have a vulnerable library on your classpath.
- #349 Upgrade PDF-BOX to 2.0.15 to avoid CVE in 2.0.14. Thanks @BryceMehring.
- #347 Add document language and title preference for PDF/A documents to satisfy Acrobat Pro validator. Thanks @mattstjean.
- #339 Mark Jsoup DOM converter module as deprecated (for removal). Please see integration guide for replacement. This module may also pull in an insecure version of Guava so please migrate now.
- #336 Fix for broken image links causing an NPE. Thanks @svenfrauen.
- #334 Allow the user to supply
PDPageobjects via page supplier. Thanks @DSW-PS.
- Please start using the fast renderer (
builder.useFastMode()) as the old renderer will be removed in a future version. - #180 Fast renderer is finally ready for production. The fast renderer comes with:
- Nearly 150 automated end-to-end regression tests. This is about 150 more than the old renderer.
- Improved performance. This renderer scales linearly with the number of pages, compared to the old renderer which scaled with the page count squared.
- Far better support for transforms, including nested transforms, multiple transforms and transforms interacting with hidden overflow, etc.
- Better support for hidden overflow, with boxes now not escaping except with accordance to the standard.
- Support for inserted cut off overflow pages. See Cut-off page support on the wiki.
- Link areas and their hash link targets now repect transforms.
- Bookmark targets now respect transforms.
- Improved page placement for boxes. Now respects overflow and tranform properties.
- Greater understanding which should make fixes and feature improvements easier.
- Visual testing API is now available to use in the PDFBOX module. Please see testing your PDF on the wiki. Thanks @red6.
- #333 Upgraded PDFBox to 2.0.14 and PDFBox-Graphics2D to 0.21.
- #315, #79 Accessible and tagged PDF support. See PDF Accessibility (PDF UA, WCAG, Section 508) Support on the wiki.
- #326 Proper support for PDF/A standards with automatic regression testing. See PDF/A Standards Compliance on the wiki.
- #328 SVG with
pagerule was crashing in certain circumstances. - #324 Better logging with invalid or missing fonts.
- #320 NPE prevention in case of incorrect font configuration.
- a145329 Were using incorrect font-metrics in certain situations.
- #303 Fixed: Table borders are partly transparent.
- #297 Fixed: Border not printed with "overflow: hidden".
- #304 Fix warnings for icon font without space inside PDF/A, add tests.
- #301 Make loading resources from classpath work when openhtmltopdf is a named module.
- #232 Were using JRE internal APIs.
- #289 System.out.println("Getting image") in NaiveUserAgent.
Thanks to these people for pull-requests:
- @rototor
- @brundipub
- @zimmi
- @dnguyenminh
Finally, a big thanks to all issue reporters and extra thanks to those who help out in issues.