We are happy to announce that the development efforts for Thymeleaf 3.0 have not only already started, but in fact have already gone a long way! Thymeleaf 3 is going to be a huge leap for Thymeleaf, and we are really excited about the features we plan to include in this new major release.
Just to give you a taste, these will be Thymeleaf 3's main goals:
* Better performance. Faster template engine execution.
* Smaller memory footprint. Reduced latency in highload servers.
* Full support of (nonXMLized) HTML code, esp. HTML5.
* No disruption of existing template code base. Seamless migration of templates from v2.
* Maximum ability to create HTML templates as pure HTML. Not only Natural, but also Decoupled Templates: HTML/XML templates with no Thymeleaf code at all.
How will Thymeleaf v3 achieve this? We will need two new things:
* A new parser. Lighter, more efficient, purpose-built. With much better HTML support.
* A new engine. More modular, heavily optimized. Reduced parse-to-output time.
The good news is, we already have such new parser: AttoParser v2. AttoParser was born in late 2012 with the aim of being the markup parser for future versions of Thymeleaf, and we have now given it such a huge refactoring in order to support even better all the things we need for Thymeleaf 3 that it jumped to major version v2 (still in beta, anyway, but very stable). It is now the perfect parser for a markup-savvy template engine like Thymeleaf :-)
And the even better news is, we also have the foundations for the new engine. Many of the new concepts have already been tested (and implemented!) and we are looking really good so far.
There are lots of things to do yet, but we hope that we will have Thymeleaf 3 at least in BETA stage somewhere in the Summer of 2015, and a stable release soon after.
Thanks to all of you for having allowed this project reach where it is now.