What to Expect in Ext JS 4.1
We’ve been hard at work on Ext JS; both working on bug fixes and improvements in 4.0.x as well as in 4.1. The first beta of 4.1 is around the corner so wanted to share the details of what to expect in Ext JS 4.1.
In talking with the community, we heard loud and clear the two areas we needed to spend the most time: performance and documentation. Indeed that has been our focus for 4.1. Let’s look at each of them individually.
Performance
Many of you rely on Ext JS for your mission critical projects, which are often highly complex apps with hundreds of layouts, views and configurations. We’ve found that for certain classes of applications, Ext JS 4.0.x could be a lot faster in how it deals with the DOM. Specifically, for applications that have complex layouts, Ext JS 4.0.x reads and writes to the DOM frequently causing latency in the layout process. In Ext JS 4.1 we’ve changed this to be far less ‘chatty’. Instead of re-reading and re-writing the DOM during the entire layout process, 4.1 will now buffer and batch reads and writes.
We’ve also made further changes in the component rendering: instead of each component rendering itself directly to the DOM, its writes are now batched up into a single DOM insert per top-level container, which means that any app using a Viewport will have all of its components rendered in a single DOM write. This should significantly increase the performance of both simple and complex layouts across browsers.
In addition to the new DOM management strategy, Ext JS 4.1 also speeds up component rendering. In 4.0.x, when a component was rendered we used CSS selectors to find elements. We’ve updated that to use IDs, which are faster when performing a component layout. We have also updated the Grid component to use browser-native scrollbars and have moved away from virtualized scrolling. This offers two significant benefits. First, the grid scrolls faster and offers a more ‘browser-native’ scrolling feel. Second, for Mac OS X Lion users, the scrolling behavior is now Lion-compatible.
Not specifically performance related, we’ve also fixed bugs across the board that have been reported in our forums, through our support channels, as well as through customer conversations. We’ve fixed bugs in the data package (making sure esoteric combinations of proxies, readers and writers work as expected), the charting package (performance and edge case improvements). We’re also hoping to reintroduce the Calendar component with 4.1.
You’ll also notice in the Ext JS Forums, bugs that have been linked to our internal bug reporting system now show the issue number so we can better track and respond to community issues.
Many of these fixes are already available to support subscribers and we’ll be happy to make them publicly available when Ext JS 4.1 ships.
Documentation
The documentation tool that was released with Ext JS 4.0 provided a fresh new look at all the documentation that Ext JS had to offer. While documentation is never done, we’ve spent a good amount of improving two areas of the docs: the actual docs, as well as the doc tool. In 4.1 we’ll be including new content as well as refreshes to the existing content for clarity and consistency. The upgrades we’ll be making to the docs tool will make it easier to navigate and explore the content and allow it to become a better reference guide for working on Ext JS. We’ll be sharing more details about documentation in a blog post next week.
When to Expect Ext JS 4.1
We’re looking to deliver the first beta of Ext JS 4.1 in mid to late September. While that may feel like some time from now, the overhaul of the layout mechanics have required a significant amount of engineering and testing and we want to ensure that we work through as many of the ‘gotchas’ as possible so our community doesn’t have to.
Ext JS 4.0 introduced a huge number of new features, but also turned out to have some performance and documentation holes and we hope that Ext JS 4.1 solves as many issues and concerns that you’ve had. Also, we originally had RTL and ARIA support scheduled for 4.1, but with the emphasis on performance improvements for 4.1, we’ve had to push these features out to a 4.2 release. We can never thank you enough for all the help and attention you give to Ext JS, whether it’s through reports on the forums, conversations on Twitter, finding us at conferences, or wherever. We’re committed to making Ext JS hands-down the best choice for desktop web application projects!

There are 72 responses. Add yours.
Michael Camden
9 months agoIn the 4.0.x releases you introduced support for locked grids, but that support was little more than allowing basic grids to be locked. It’s a struggle to implement customized locked grids with the current system. Does 4.1 address any of the number of issues people have had with locked grids?
Sebastien Tardif
9 months agoCould I beta test this ?
I used extensively IE 8 using complex layout, large buffered grid, and chart with lot of points.
Ricarde Pereti
9 months agoPlease fix the ExtJS behaviour on the Ipad. Currently is unusable.
Neil Grover
9 months agoRe: “We have also updated the Grid component to use browser-native scrollbars and have moved away from virtualized scrolling.”
Does that mean there is no virtualized scrolling or does it must mean that you use a native scroll-bar to achieve virtualized scrolling?
Aditya Bansod Sencha Employee
9 months ago@Ricarde—Ext JS is not designed for mobile devices; it is only designed to work on desktop browsers. You should look at Sencha Touch for mobile.
Steffen Hiller
9 months agoI like those what to expect posts!!
dbrin
9 months agoLooking forward to it!
Jay Garcia
9 months agoI can totally tell that the team is working hard on performance and stability. I think that the work and efforts placed into 4.1 will be applauded by many devs.
diego
9 months agowould be good if this changes are reflected in gxt.

Conor
9 months agoLooking forward to the first beta.
Really hope the calendar component makes it into 4.1 So many apps use it as an integral part.
Noam
9 months agoIt’s not OK! we really need the RTL support. we buy a license because you told us by email that you will support RTL in version 4.1!
It’s really not OK!!!
ordinary_guy
9 months agoMany thanks for the update Aditya! With scrollbars sporadically dying in grids and trees and general tab panel slowness, has been a hard time convincing our team to choose Ext JS over Java Server Faces for our Java Swing app re-write.
Definitely prefer performance and stability over new features.
Cheers!
dd
9 months agoWill the inline grid filter row be included?
How about the new themes?
Aditya Bansod Sencha Employee
9 months ago@Noam—what we’ve effectively done is split 4.1 in to two release. 4.1 focused on performance and docs, and 4.2 the features. We remain committed to delivering on ARIA/508 and RTL as before but we wanted to ensure we got the performance fixes out to people as soon as possible.
frank
9 months agoLet the code smaller
AwesomeBob
9 months agoWhy do we have to keep asking for Neptune? Give us Neptune.
jamie Nicholson
9 months agoRicarde Pereti - I’ve got some working code that adjusts the layout properly for ExtJS in the iPad/iPhone/iPod. Some of the more detailed features don’t work too well though. Flick me an instant message on the Forums for the code - I think it’s in the forums/web somewhere too.
michael
9 months agois faster than 3.x?
Ajay Mali
9 months agoHi Aditya
I am implementing ExtJS in lotus notes domino, facing with some issues in address picker. not received ay response from long time.
http://www.sencha.com/forum/showthread.php?77156-2r2-PickList-Names-Issues
May I know, who will be the right person to contact for this issue.
Thank you in advance
Girtab76
9 months agoGreat. Keep up the good work. imho a more extensive examples section would be really appreciated.
It makes you better and quicker understand the framework.
Chris
9 months agoWhy aren’t you rendering in Canvas yet?
I didn’t expect the DOM to be slow. Even not with PreLayout-Properties.
bad extjs !
9 months agoI am going to changed our project to Dojo, Because of extjs4 ‘s bugs!
I think sencha is not a responsible company !!!
The extjs 4.1 release date was changed from mid of August to late in September ,I think the extjs 4.1 will be release in the next centrury !!!
I suggest the user of extjs4 , be ready to change your javascript framework ,
Niklas
9 months agoNeptune got pushed from 4.0 to 4.1 to..?
Anyways, great that this blog gets updated so often! Always worth the reading.
john
9 months agoEver since the introduction of the buggy Extjs 3.0, followed by the shamelessly buggy initial release of Designer, I have told our management to not even consider using any Sencha products before a version x.2 is released. Untill then.
bobp
9 months agoAditya,
Can you pls address the important question re grids asked previously:
“Does that mean there is no virtualized scrolling or does it must mean that you use a native scroll-bar to achieve virtualized scrolling?”
We don’t want to embed virtual scrolling in our product now under development, only to have to remove it later if the feature is withdrawn from Ext… tks.
Steve Howe
9 months agoNice news. I hope you didn’t forget about core functionality missing, like this:
http://www.sencha.com/forum/showthread.php?137311-processResponse-is-wrong-in-Ext.data.proxy.Server&p=618877
Thanks !
AwesomeBob
9 months agoWow Sencha. Looks like you shouldn’t have blogged this one. Most of the commendatory is displeased.
mdmadph
9 months agoSo, can we get back our tabs in the documentation? Please?
Les
9 months agoSencha documentation is the best I have seen, so I’m not sure what the complaining is about. I don’t use it that much because I prefer to open the source, which also is very nicely put together compared to other libraries that I’m familiar with (jQuery and Dojo).
I hope many bugs in the Ext.draw package will be fixed in 4.1. Overall, 4.0 is a very impressive release.
Michael
9 months agoTo the guy planning to switch to Dojo:
Dojo has many more bugs and much less active development. I use both and each has their place. Just don’t think that the particular problems of bugginess, responsibility or release dates are going to be better elsewhere. They won’t.
Ed Spencer
9 months ago@Steve Howe: we fixed that one in 4.0.5
Will be present in 4.1.0 also
@bobp the virtualized scrolling was not able to adequately address all of the use cases we threw at it. We also found a way to use normal native scrolling and achieve all of our goals with fewer bugs and the same API
omegafox
9 months agoI am wondering if there is major improvements with MVC architecture and MVC’s documentation. Few new examples for MVC willl be huge plus.
Crysfel
9 months ago@mdmadph : So, can we get back our tabs in the documentation? Please?
I agreed with that! please tabs in the docs!!
ykey
9 months agoI do not miss the tabs at all. The history support and caching more than make up for it.
slemmon
9 months agoAgree with ykey. At first missed the tabs. But, don’t anymore.
BTW, is the list of open bugs out there somewhere visible? Or a ‘fixed’ list between versions? I was wondering why combo’s select event wasn’t picked up by my controller and finally saw in the forums that there’s a bug open on that. Would be nice to get to that info more quickly if it’s out there for the community to see.
bclinton
9 months agoWhile I agree that performance and documentation are two of the three major areas that needed to be addressed, I am disappointed that the third area, “stuff you’ve been promising for a year for the 4.0 version but have never delivered” seems like it is being passed to an undefined future release.
When will the Neptune theme, SASS and Compass stuff that you previewed at SenchaCon 2010 be available? You guess realize you still have this video in the Theming section listed on the first page of the API docs right? Maybe a large red text “NOT AVAILABLE - NO ETA” disclaimer above the video would be appropriate?
The other thing I was counting on for the 4.1 release was a functioning MVC dispatcher. I became comfortable using this with Sencha Touch and then it seems it was quietly removed from ExtJS sometime after the 4.0 release. Hopefully this will still be part of the 4.1 release but it is impossible to know from this blog post.
I appreciate this blog post since communication has been somewhat less than optimal, but the pessimist in me reads it as “4.1 is coming out much later than you expected and it still won’t contain many of the features you were expecting it to have.”
I love ExtJS and I’m a huge fan of the people that put it out, but it really feels like y’all screwed the pooch by releasing 4.0 well before it was ready and now you are desperately trying to play catch up and are losing the battle. I’ll stick with it because ExtJS is without a doubt the best JavaScript application framework available, but in the future I hope you correct the problems that led to this fiasco. You are all too good to look this crappy.
mehdi7033
9 months agoRTL????????????????????????????????????????????????
why not support this version.
Arjen
9 months agoThanks for the update! Fine-tuning the performance is always a good thing.
What about Sencha Touch? Late summer is mid-September?
christian
9 months agoI’d rather know what’s going on with Sencha Touch. 5 months without even a minor bug fix release is meh.
Dawesi
9 months agoas long as you put tabs (or something just as useful) back in the docs, the rest of what you mention has already made me happy.
Also bringing Sencha ExtJS and Sencha Touch to be in unity would be great. It would even better to get to a point where ExtJS would dynamically load ExtJS or touch extra libraries on top of core libs on detection, rather than having to have two code bases.
Dawesi
9 months agoand also Fix the differences between the MVC features of touch and extjs!!!!!!!!!
Mark
9 months ago508/ARIA delayed yet again. It’s been promised repeatedly since 2.x. I’m tired of being strung along with this “next release” crap. I don’t feel we can trust anything out of Sencha’s mouth regarding accessibility. If it really was a focus, it would have been done by now. If you do actually get it done.. how about this—you give everybody who’s ever paid for a license a free upgrade because of your repeated failure in this regard.
as
9 months agoRTL support !!!
Monaco
9 months agoTrying to please everyone regarding extended interoperability and features without ironing out performance issues and bugs first, would eventually completely water down the quality of the framework.
Thanks Sencha team for prioritizing responsibly and for not letting the feature race cloud your judgement.
Viacheslav
9 months agoRequesting range slider support with several knobs.
Ajsie
9 months agoI agree with a comment above that ExtJS 4 was delivered way too soon before it was even a ready product. The MVC framework is a total disaster.
You should look at Sproutcore and learn about how proper MVC should be done on the client side. At least have a Statechart implemented. That should be one of you priority feature when it comes to MVC imo.
Anyway, I love ExtJS and I hope you will learn from your mistakes and the community feedbacks, even if you are getting a lot of comments you don’t like to hear
Reza
9 months ago1- Please if it’s possible combine ExtJS and Sencha Touch API and just mention their differences. So their learn stuff would be usable for both developers group.
( core ExtJS [data,store, ...]
+
Destktop ExtJS [desktop specific components]
+
Mobile ExtJS [mobile specific component] )
2- RTL is all about a CSS file and probably a small Js file. Its implementation should not be very hard (in fact it’s so easy for you!) so please be RTLful!
3- Please embed popular ux in ExtJS (plugins that are distributed with ExtJS)
Jared
9 months agoCan you tell us if there will be any enhancements to 4.1? The items addressed in the blog seem to be improvements (ie. no new functionality). Will 4.1 be bring anything functionally new?
makiss
9 months agoThanks for laying out guys and telling it like it is. It is very important to keep communication open with your customers and admit when things could be better and comfort us that they are going to get better
Cheers!
Louis
9 months agoEven a hint about the imminent appearance of Neptune would be appreciated. Our decision to begin theme development rests on whether or not Neptune will actually ever be delivered. We have, after all, been waiting since February.
Jonathan
9 months agoRTL support is very important to several of our customers. We set expectations with them based on Sencha’s statement that RTL was coming in 4.1. Now RTL has been delayed (again), which will cause us significant difficulties with those customers.
I appreciate the heads-up that RTL will not be in 4.1, but we need more clarity on when that feature will actually be released. I understand that development is not a science and releases cannot be precisely predicted. However, we need to know at least in which quarter we can safely expect RTL support.
Thanks!
LoreZyra
9 months agoI agree with mdmadph on the tabs in Documentation… What happened to that idea???! Now, it feels like the functionality took a step backwards as it (re)loads a new page for every function we look at!
LoreZyra
9 months agoNeptune theme…
508/ARIA support…
Delayed to future undefined release…
Many great features of ExtJS 3.3.x series lost in release of ExtJS 4.0
Then you say: let’s focus on performance and documentation. I agree that you must focus in these areas… But if it forces your other promises of what we should expect in the next release to be further delayed, then there must be something in the management that needs to be adjusted.
If there was ever something that Sencha should improve, it would be Sencha itself. You have provided great products in the past. I’ve been a fan ever since the release of ExtJS 2.0. I wonder if Jack Slocum had stayed would the company be moving as a single unit.
Roadmaps are even less detailed since Jack left and this leaves many wondering the future value of the framework. Many, including myself, make tentative plans based on your announcements. So, when I read “or late September,” I instantly re-read it as “late October…” So, when you guys said late-Summer for the release of 4.1. I told myself and team to expect end of the year. My point here is that your communication about your products leaves a LOT of room for improvement.
BTW, can’t wait to see ExtJS 4.2 released.
Davi
9 months ago@LoreZyla I Hope Late summer doesnt mean at the end of the Summer. I’m Brazilling so the Summer will end at 2012’s March…. (hahah) And I hope the Maia guys are wrong and the EOE (End of Earth) will not be in December 21th 2012 (probable when 4.2 or 4.3 will be released).
Sencha, I respect very much your product, but there are frameworks growing out there so you guys must keep the “Tundercats’ Eye” to keep you on top. Senchas is going to be a HUGE framework when it would be the BEST framework…
Cheers,
Davi
Late summer
Swarnendu De
9 months agoI hope to see a TabPanel in ExtJS 4 docs similar to what we had in Ext 3.3 and before. It really annoys a lot when I need to check 3-4 components together and every time going back to the menu.
I found the SASS themeing with ExtJS 4 is giving some trouble (Its good for Sencha Touch though). Anyone else getting this issue?
Jalal
9 months agoRTL support was promised on EXTJS 4.1 !!
Wemerson
9 months agoNeptune theme please!
John Wilander
9 months agoSo you’re postponing WAI-ARIA again. “Mid to late September” means we can expect 4.1 in October-November. If WAI-ARIA is actually included in 4.2 I guess that’ll be around March 2012. Am I right?
daiei27
9 months ago@Monaco: I totally agree!
It’s very important to get the framework up to par first and make sure 4.x starts out as good as the latest 3.x release before starting down other rabbit holes.
Thanks for all the hard work!
EnterpriseClient
8 months agoHello there,
just to let you know how serious this is. We are currently having 3 large ExtJS projects. Large means business-critical for the client and involving developer teams of about 10 people each. Thus in total about 30 people are working on them. The delay we had to justify to the clients due to the buggyness and underperformance of 4.0 was so big that I had to make concessions worth 75.000$ to one of the clients, on top of which we lost a new contract that was considered safe. That is damage, indeed!
Specific features such as ARIA: As much as they may be nice-to-haves, as long as the core functionality of ExtJS 4.0 is not satisfactory, there is no point whatsoever to waste time on them. Please fix the ExtJS 4.0 core first, because you and your clients, such as we, otherwise lose marketshare and a lot of money if you don’t and will eventually have to switch.
Best regards.
nikee69
8 months agoEnterpriseClient is absolutely right, the ExtJS 4.0 debacle is costing a f*cking lot of money. For us it’s about $35’000 spent in an enterprise project to migrate to version 4.0 that is now absolutely useless as we were forced to stick to ExtJS 2.1 because of intolerable performance issues.
Anil Bhagwat
8 months agoHi,
We have been seeing really bad performance on IE7 with ExtJS 3.2.2. Compared with Ext 2x it looks to be about 50% slower. Millions of our corporate client user base is still on IE7 whether we like it or not. We provided some sample code to Sencha a few weeks ago demonstrating the big performance issue in Ext 3.2.2 against IE7.
Does Ext 4.1 address performance issues on older browsers like IE7?
EnterpriseClient
8 months ago@Anil: At least you should not use Ext4.0 if you still have a large IE7 user base. Regarding 4.1: We won’t trust it until we see it.
jony
8 months agoIt is nearly Mid_Sep. now ,where is 4.1.0???????
Sebastien Tardif
8 months ago@jony You should remember how the release of 4.0 was managed. The “release” date had quick changed to “preview 1” when the promised date was near, then the real “release” was months after. Since we didn’t yet see or heard about any 4.1 preview or beta, and knowing that each preview or beta iteration are probably at least 2 weeks apart, that should give you an idea how optimistist you should be…
Diandra
8 months agoThis ?free sihrang? of information seems too good to be true. Like communism.
Michael Camden
8 months agoIt is now the end of September, any chance 4.1 will be released today? Do you intend on delaying this release further?
kprk
8 months agoHi I am looking forward for EXTJS 4.1 when is next release?
Les
8 months agoThere was a post by Ed Spencer in the Sencha Forum… expect 4.1 beta early in October.
Michael Mullany
8 months agoWe’ll have a complete blog update on the 4.1 release hopefully tomorrow.
UPDATE: We’re hoping to get a blog post update on Monday. Sorry for the delay, but we want to have good confidence about the plan
Lars Andersson
8 months agoAny news about the update?
Thomas
8 months agoHi There,
I read in the forums that there were plans for integration tristate-functionality for checkboxes/trees in 4.1…
Any news about that? Will this make it in the final release?
Comments are Gravatar enabled. Your email address will not be shown.
Commenting is not available in this channel entry.