OPEN HTML TO PDF
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.
- Templates for Openhtmltopdf - MIT licensed templates that work with this project.
- 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
- 1.0.4 Online Sandbox - Now with logs!
- Sample Project - Pretty Resume Generator
DIFFERENCES WITH FLYING SAUCER
- 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 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.
Test cases, failing or working are welcome, please place them
and run them
head - 1.0.9-SNAPSHOT
- See commit log.
- #675 Update PDFBOX to 2.0.23 to avoid CVEs. Thanks for reporting @Samuel3.
NOTE: These CVEs relate to the loading of untrusted PDFs in PDFBOX and thus this project is not directly affected. However, it is not a good idea to have CVEs on your classpath.
- #650 Support for multiple background images on the one element. Thanks for requesting @baedorf.
- #669 Support fallback fonts. Thanks for requesting @asu2 and assisting @draco1023.
- #640 Implement file embeds via the download attribute on links. Thanks for original PR @syjer and for requesting @lindamarieb and @vader.
- #666 API to get the bottom-most y position of rendered content to be able to position follow on content with other tools. Thanks for extensive reviewing of PR @stechio and for request by @DSW-AK.
- #664 Improved support for PDF/A and PDF/UA standards. Thanks for PR @qligier.
- #653 Fix for inline-block elements with a z-index or transform were being output twice. Thanks for reporting @hannes123bsi.
- #655 Correct layout of ordered lists in RTL direction. Thanks for PR @johnnyaug.
- #658 Implement
contentproperty. Thanks for PR @BenjaminVega.
- #647 Fix race condition in setting up logger in multi-threaded environments. Thanks for PR @syjer.
- #638 Ability to plug-in external resource control based on resource type and url. Thanks for original PR @syjer.
- #628 Use enhanced image embedding methods from PDF-BOX. Thanks for PR @rototor and your work in PDF-BOX implementing this.
- #627 Fix regression where a null font style was causing NPE. Thaks for PR @rototor.
- #338 Implement read-only radio button group. Thanks for investigating, reporting and patience @ThoSchCon, @aleks-shbln, @dmitry-weirdo, @syjer and @paulito-bandito.
IMPORTANT: #615 This is a bug fix release for an endless loop issue when using break-word with floating elements with a top/bottom margin.
- #624 Update PDFBOX to 2.0.22 and pdfbox-graphics2d to 0.30. Thanks @rototor.
- #467 Prevent possibility of CSS import loop.
- #621 Allow spaces in data uris. Thanks @syjer.
SECURITY: #609 Updates Apache Batik SVG renderer to latest version to avoid security issue. If you are using this project to render untrusted SVGs (advised against), you should update immediately. Thanks a lot @halvorbmundal.
IMPORTANT: The fast renderer is now the default in preparation of removing the old slow renderer. To temporarily use the slow renderer, you can call the deprecated method
builder.useSlowMode() (PDF output only).
IMPORTANT: #543 This version stays on PDFBOX version 2.0.20 due to a bug with non-breaking spaces in version 2.0.21. Please make sure version 2.0.21 is not on your classpath. This bug has been fixed in the upcoming 2.0.22.
- #544 Code to create a website for pre-canned PDF templates in thymeleaf and raw XHTML format. Check out the template website to preview templates.
- #533 Barcode plugin. Very useful PR supplied by @syjer. Barcode plugin docs.
- #521 Move Java2D image output to fast renderer and general improvements. Java2D image output docs.
- 9ffd0e #568 Filter out problematic characters that are visible in some fonts but should not be such as soft-hyphen. Thanks @StephanSchrader.
- #587 Fix for white-space: nowrap cutting off instead of wrapping. Thanks @vipcxj for finally fixing via PR.
- #577 Add foreground PDF drawer plugin (useful especially for watermarks). Thanks @rototor for PR and @sillen102 for persisting.
- #566 Rename
baseDocumentUriand improve javadoc to avoid confusion. Thanks for reporting @NehalDamania.
- 801780 Update junit test dependency to 4.13.1 to avoid security scanner warnings (the specific security problem did not impact this library).
- #553 Fix for ContentLimitContainer causing NPEs when negative margins are used. Thanks for reporting @adilxoxo.
- #552 Optimize the log formatter for j.u.l logging. Thanks for the impressive PR @syjer.
- #542 Improve list-decoration placement. Thanks for PR @syjer and reporting @mndzielski.
- #458 Fix for list-decorations being output (clipped) in page margin area.
- #525 Remove unused schema/DTDs. Significantly reduces size of jar. Thanks for PR @syjer.
- #592 Allow unit (px, cm, em, etc) values in the width/height attributes of linked SVG images. Thanks @DanielWulfert.
- #594 #458 Fix for more repeating content and PDF/UA crash. Thanks @ThomHurks, @fungc.
- #599 Fix RuntimeException ocurring on InlineText.setSubstring. Thanks @LAlves91.
- #605 Fix to make justification work with surrogate pairs. Thanks @EmanuelCozariz.
- #601 Move CI to Github actions. Thanks @syjer.
- #597 Generalize data uri support. Thanks @syjer, @Leostat86.
- #613 Allow adding fonts for SVG, MathML as files instead of input streams to avoid JDK bug. Thanks @syjer, @sureshkumar-ramalingam, @olayinkasf.
- b88538 Fix for endless loop when using
word-wrap: break-word. Thanks for reporting, testing and investigating @swarl. Thanks for tests and debugging @rototor and @syjer.
- #492 Lots of testing of the line-breaking algorithm to avoid future endless loops. By @danfickle.
- #515 Pass document CSS styles applying to SVG element to SVG implementation. Thanks for requesting and contributing @amckain92.
- #514 FIX: Correctly position boxes when justifying rtl lines. Thanks for reporting and testing @lzhy1101.
- #512 #507 #502 Cleanup code including deleting unused code, generics, etc. Thanks for PRs @syjer.
- #489 Extensive overhaul of logging including per run diagnostic consumer. Huge thanks @syjer, a lot of work in this PR. See logging page on wiki for more info.
- #501 Upgrade PDFBOX to 2.0.20 and PDFBox-Graphics2D to 0.26. Thanks for PR @rototor.
- #490 Fix for NPE when decoding image data url fails. Thanks for PR @syjer and reporting @AlexisCothenet.
- #516 Add OSGI bundle metadata to MANIFEST.MFs. Thanks for requesting and investigating @zspitzer.
- IMPORTANT: This release contains fixes for two bugs that may result in endless loops/denial of service when using
word-wrap: break-word. If you are using this feature, please upgrade promptly.
- #483 Fix for endless loop bug with
word-wrap: break-wordand soft hyphens. Thanks @rototor for PR, @syjer for analysis and @swarl for reporting.
- #466 Fix for endless loop bug with
word-wrap: break-wordand zero width boxes. Thanks @syjer for analysis and @AlexisCothenet for reporting.
- #486 SVG plugin can now provide a list of allowed protocols for external resources and any configured uri resolver/stream handlers will be used. Thanks @syjer for PR and @ieugen for reporting.
- #480 Fix for link shapes being returned from custom object drawers. Thanks @rototor for PR and @hbergmey for reporting.
- #485 Implement support for SVG data uris. Thanks @syjer for PR and @adrianrodfer for reporting.
- #470 Allow
mailto:links or any other valid link. Thanks @syjer for PR and @mndzielski for reporting.
- #464 Honor the
directionCSS property. Thanks @AnanasPizza for reporting.
- #460 Change thrown exception class to more specific
IOException. Thanks for PR @leonorader.
- #459 Implement the
remCSS unit. Thanks to @leonorader for reporting.
- #211 Images can now be used in the CSS
contentproperty. Thanks for requesting @Kuhlware.
- #445 Fix for not picking up attribute values in Jsoup converted documents. Thanks for reporting @testinfected.
- #450 Java2D output only: Ability to add fonts via code. Also environment fonts will no longer be used by default. To use environment fonts:
- SECURITY Removed Log4J 1.x adaptor as it had CVE-2019-17571 with no updated version available.
- #448 Implement
background-imageproperty. By @danfickle. Requested by @rja907.
- #429 Major overhaul of
word-wrap: break-word. Now a word will not be broken unless it is too big for a line by itself. By @danfickle. Thanks for reporting and testing @mndzielski.
- #433 Do not justify lines ending with
<br/>tag. Thanks for reporting @fcorneli.
- #440 Remove trailing white space for right aligned text to avoid jagged appearance. Thanks for reporting @AnanasPizza.
- #446 Look for lang attribute on ancestor elements when using
lang()selector. Thanks for reporting and tracking down the bug @fungc.
- #430 Use relative path to license in source jars instead of absolute path. Thanks for reporting @gabro and fixing via PR @syjer.
- #417 Keep aspect ratio of images with width/height properties as well as min/max width/height properties. Thanks for reporting and basis for fix @swarl.
- #423 Allow multiple font sources to be specified with
formattags. Only use
format(truetype). Thanks for requesting @MichaelZaleskovsky and basis for implementation @syjer.
- #415 Avoid class cast exception if user tries to float table cell. Thanks for reporting @dmartineau99 and PR @syjer.
- #421 Avoid NPE when justified text is mixed with unjustifiable content. Thanks for reporting @Megingjard and PR @syjer.
- Updated PDFBOX 2.0.17 to 2.0.19.
- #413 Handle form problems such as no name on input element without throwing a NPE. Thanks @syjer for PR and @mmatecki for reporting.
- #412 Add HTML block level elements usch as
sectionto default CSS. Thanks @syjer.
- #339 Remove the JSoup to DOM converter module. Thanks @kewilson.
- 0cd098 Fix for letter-spacing support on last line of block with trailing space. Also performance improvements and refactoring. By @danfickle.
- #410 Fix for wrong bold setting on list item counters. Thanks @syjer for PR fix (and test!) and @acieplinski for reporting.
- Wiki Configurable text justification settings as part of a justification overhaul that also allows more space to be used inter-char when there are no spaces on the line. By @danfickle. Commits listed in #403.
- #403 Soft hyphen support. Soft hyphens are now replaced with hard hyphens when used as line ending character. Thanks @sbrunecker.
- #408 Fix for bookmarks not working with HTML5 parsers such as JSoup. Thanks @syjer for investigating and fixing and @Milchreis for reporting.
- #404 Upgrade Batik to 1.12 and xmlgraphics-common to 2.4 (both used in SVG module) to avoid CVE in one or both. Thanks @avoiculet.
- #396 Much faster rendering of boxes using border-radius properties. Thanks @mndzielski.
- #400 Support for
abbrtag for accessible PDFs. Thanks @Ignaciort91.
- #394, #395 Upgrade PDFBOX to 2.0.17 and pdfbox-graphics2d to 0.25. Thanks @cristan, @rototor.
- #384 Allow user to provide PDFont supplier. Thanks @DSW-PS.
- #373 Fix regression where both max-width and max-height are provided for images with certain aspect ratios. Thanks @rototor.
- #380 Much better support for flowing columns including explicit column breaks, floating content, block level nested content. By @danfickle.
- #372 Much improved sizing support for
- #344 Use PDFs in
<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:
Finally, a big thanks to all issue reporters and extra thanks to those who help out in issues.