Towards Quarkus 2.x

[Guillaume Smet]

Hello,

Quarkus was announced to the world 2 years ago. During these 2 years, a lot of things happened, and we can’t thank you enough for all the support and contributions you have all provided. These two years have been an incredible journey, and this mail is about the future.

At the moment Quarkus 1.13 is getting released. You should be able to use it very soon. It has a lot of cool features such as DevServices, Kubernetes Service Binding, and OpenTelemetry.

So what’s next?

Over the last few months, we have been working on the integration of MicroProfile 4 and Vert.x 4. MicroProfile 4 is the latest version of the MicroProfile specifications. It comes with plenty of novelties, such as an improved configuration, fault tolerance, OpenAPI, GraphQL, and so on. Vert.x 4 is the latest version of Vert.x containing lots of API improvements, traceability support, and plenty of other novelties. As Quarkus is based on Vert.x and relies heavily on MicroProfile specifications, such switches are important.

That’s why we decided to bump the Quarkus version to 2.x once these features will be integrated.

Is that all in 2.x?

Oh no! These are mostly internal changes. It should be noted that the migration of most applications will be a no-brainer. We will describe the few breaking changes and how to update your applications if you use some incompatible code in a migration guide as usual.

But, we have more to offer in 2.x. First, Stuart recently started working on Continuous Testing. We plan to include it in 2.x.

qs, the new Quarkus command-line tool is going to be included in 2.x too.

We are still evaluating if we could drop Java 8 support in 2.x (1.x will still be compatible with Java 8).

We are working on other features, but we don’t want to tell you everything now ;-) Follow the pull requests on GitHub to know more.

Is 1.13 really the last 1.x one?

We will still backport bug fixes in the 1.13 branch and release 1.13.z versions periodically.

If we have features for a 1.14, we may even consider releasing a 1.14. However, the main development will focus on 2.x and our main branch will be 2.x soon.

What about the schedule?

We plan to cut a few alpha versions of 2.x in April and May to collect feedback, for a final release in June (this is not a commitment, just a rough plan).

Quarkus 2.x is the continuation of 1.x. It’s not a big refactor, it’s not something totally new. The focus will still be the same: supersonic, subatomic, fun!

This is a proposal, Quarkus is an open project so we are open to feedback, discussions, and ideas from the community.

Thanks!

For the Quarkus Team
Clement and Guillaume

[Manyanda Chitimbo]

Good news!! Thanks for the update! 

--
You received this message because you are subscribed to the Google Groups "Quarkus Development mailing list" group.
To unsubscribe from this group and stop receiving emails from it, send an email to quarkus-dev...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/quarkus-dev/CALt0%2Bo_1m3adZ5KBmVZautJXRh_uZYNgg-_%2BYUBnHVM3pLiq6A%40mail.gmail.com.

--

Manyanda Chitimbo.

[Loïc MATHIEU]

Good news!

> We are still evaluating if we could drop Java 8 support in 2.x (1.x will still be compatible with Java 8).

Where will the fight occurs ;)

I think the community is ready for 11 now, so you have my vote on this one ;)

> If we have features for a 1.14, we may even consider releasing a 1.14

As 1.13 was forked from mainline two weeks ago I think we should deliver a 1.14 (there are already important features on it), but it's up to you as it will mean more work for you not for me ;)

> Quarkus 2.x is the continuation of 1.x. It’s not a big refactor, it’s not something totally new. The focus will still be the same: supersonic, subatomic, fun!

As I understand it, Quarkus 2 will bake for more time than usuall releases (3 months instead of 1), and as its major version increment suggest, may break some things, as little as possible, but still we, as extension authors, car take the opportunity to make some big refactoring.

For exemple:

- Splitting MongoDB extensions in two: an imperative and a reactive extension (we already discussed it and it'll be more natural as all other extensions do this)

- Moving an extension to/from the quarkiverse

- Others should have other ideas ;)

Regards,

Loïc

To view this discussion on the web visit https://groups.google.com/d/msgid/quarkus-dev/CANwCmXKAtySrdLZpLAMKhi2q45OO1LcXNhyTGvzGZiFs17DTYA%40mail.gmail.com.

[Georgios Andrianakis]

On Wed, Mar 31, 2021 at 11:03 AM Loïc MATHIEU <loik...@gmail.com> wrote:

Good news!

> We are still evaluating if we could drop Java 8 support in 2.x (1.x will still be compatible with Java 8).

Where will the fight occurs ;)

I think the community is ready for 11 now, so you have my vote on this one ;)

> If we have features for a 1.14, we may even consider releasing a 1.14

As 1.13 was forked from mainline two weeks ago I think we should deliver a 1.14 (there are already important features on it), but it's up to you as it will mean more work for you not for me ;)

> Quarkus 2.x is the continuation of 1.x. It’s not a big refactor, it’s not something totally new. The focus will still be the same: supersonic, subatomic, fun!

As I understand it, Quarkus 2 will bake for more time than usuall releases (3 months instead of 1), and as its major version increment suggest, may break some things, as little as possible, but still we, as extension authors, car take the opportunity to make some big refactoring.

I should note that we are thinking of delivering Alphas (and Betas if necessary) so users will be able to track the progress and help identify issues with 2.x early on and also not have to wait 3 months for a release.

 

To view this discussion on the web visit https://groups.google.com/d/msgid/quarkus-dev/CAJLxjVFm3gSPb8Ymkmau%3DjHyZOg2GmqWiV34fV2i4w1P1_0xpA%40mail.gmail.com.

[Alexey Loubyansky]

On Wed, Mar 31, 2021 at 10:03 AM Loïc MATHIEU <loik...@gmail.com> wrote:

Good news!

> We are still evaluating if we could drop Java 8 support in 2.x (1.x will still be compatible with Java 8).

Where will the fight occurs ;)

I think the community is ready for 11 now, so you have my vote on this one ;)

> If we have features for a 1.14, we may even consider releasing a 1.14

As 1.13 was forked from mainline two weeks ago I think we should deliver a 1.14 (there are already important features on it), but it's up to you as it will mean more work for you not for me ;)

> Quarkus 2.x is the continuation of 1.x. It’s not a big refactor, it’s not something totally new. The focus will still be the same: supersonic, subatomic, fun!

As I understand it, Quarkus 2 will bake for more time than usuall releases (3 months instead of 1), and as its major version increment suggest, may break some things, as little as possible, but still we, as extension authors, car take the opportunity to make some big refactoring.

For exemple:

- Splitting MongoDB extensions in two: an imperative and a reactive extension (we already discussed it and it'll be more natural as all other extensions do this)

- Moving an extension to/from the quarkiverse

Right. We didn't have the Quarkiverse org from the start, so the core repo was used as the ecosystem nest. Now is a very good time to review what is hosted where.

 

To view this discussion on the web visit https://groups.google.com/d/msgid/quarkus-dev/CAJLxjVFm3gSPb8Ymkmau%3DjHyZOg2GmqWiV34fV2i4w1P1_0xpA%40mail.gmail.com.

[Max Rydahl Andersen]

> On Wed, Mar 31, 2021 at 10:03 AM Loïc MATHIEU <loik...@gmail.com>
> wrote:
>
>> Good news!
>>
>>> We are still evaluating if we could drop Java 8 support in 2.x (1.x
>>> will
>> still be compatible with Java 8).
>>
>> Where will the fight occurs ;)
>> I think the community is ready for 11 now, so you have my vote on
>> this one
>> ;)

Yeah, last time it was the community voices/asks that was pulling for
keeping 8 as
their existing environment prohibited use of Java 11+.

We expected hard push back from commercial industry but thus far we
haven't seen Java 8 asks
from customers. Would be curious to hear what others seen.

Thus it looks like those who actually do start with Quarkus are willing
to use newer Java,
and those who are in areas where Java 8 is the newest version they can
use probably also won't start on
a new project using newer frameworks and approaches (even though they
have teams that would love to do so)

I also think its time to rip the bandaid - otherwise we'll stay stuck.

Yes, time is right to do such moves but I do want to underline we should
do what we can
(within reason) to enable easy transition. i.e. thinking out loud about
the mongodb as an example if possible
to keep current extension name for easy migration but have it include
the two split ones enabling an easy upgrade path.

Don't see 2.0 to be a place to break things for users - the 2.0 is more
about extension writers having
to update to vert.x 4 and other cleanup. We should still deprecate and
provide transitions where we can.


/max

>
>
>> - Others should have other ideas ;)
>>
>> Regards,
>>
>> Loïc
>>
>> Le mar. 30 mars 2021 à 17:40, Manyanda Chitimbo <
>> manyanda...@gmail.com> a écrit :
>>
>>> Good news!! Thanks for the update!
>>>
>>> On Tue 30 Mar 2021 at 17:06, Guillaume Smet
>>> <guillau...@gmail.com>
>>> wrote:
>>>
>>>> Hello,
>>>>
>>>> Quarkus was announced to the world 2 years ago. During these 2
>>>> years, a
>>>> lot of things happened, and we can’t thank you enough for all the
>>>> support
>>>> and contributions you have all provided. These two years have been
>>>> an
>>>> incredible journey, and this mail is about the future.
>>>>
>>>> At the moment Quarkus 1.13 is getting released. You should be able
>>>> to
>>>> use it very soon. It has a lot of cool features such as
>>>> DevServices,
>>>> Kubernetes Service Binding, and OpenTelemetry.
>>>>

>>>> *So what’s next?*

>>>>
>>>> Over the last few months, we have been working on the integration
>>>> of
>>>> MicroProfile 4 and Vert.x 4. MicroProfile 4 is the latest version
>>>> of the
>>>> MicroProfile specifications. It comes with plenty of novelties,
>>>> such as an
>>>> improved configuration, fault tolerance, OpenAPI, GraphQL, and so
>>>> on.
>>>> Vert.x 4 is the latest version of Vert.x containing lots of API
>>>> improvements, traceability support, and plenty of other novelties.
>>>> As
>>>> Quarkus is based on Vert.x and relies heavily on MicroProfile
>>>> specifications, such switches are important.
>>>>
>>>> That’s why we decided to bump the Quarkus version to 2.x once
>>>> these
>>>> features will be integrated.
>>>>

>>>> *Is that all in 2.x?*
>>>>
>>>> Oh no! These are mostly internal changes.* It should be noted that
>>>> the
>>>> migration of most applications will be a no-brainer.* We will

>>>> describe
>>>> the few breaking changes and how to update your applications if you
>>>> use
>>>> some incompatible code in a migration guide as usual.
>>>>
>>>> But, we have more to offer in 2.x. First, Stuart recently started
>>>> working on Continuous Testing. We plan to include it in 2.x.
>>>>
>>>> qs, the new Quarkus command-line tool is going to be included in
>>>> 2.x too.
>>>>
>>>> We are still evaluating if we could drop Java 8 support in 2.x (1.x
>>>> will
>>>> still be compatible with Java 8).
>>>>
>>>> We are working on other features, but we don’t want to tell you
>>>> everything now ;-) Follow the pull requests on GitHub to know more.
>>>>

>>>> *Is 1.13 really the last 1.x one?*

>>>>
>>>> We will still backport bug fixes in the 1.13 branch and release
>>>> 1.13.z
>>>> versions periodically.
>>>>
>>>> If we have features for a 1.14, we may even consider releasing a
>>>> 1.14.
>>>> However, the main development will focus on 2.x and our main branch
>>>> will be
>>>> 2.x soon.
>>>>

>>>> *What about the schedule?*

>>>>
>>>> We plan to cut a few alpha versions of 2.x in April and May to
>>>> collect
>>>> feedback, for a final release in June (this is not a commitment,
>>>> just a
>>>> rough plan).
>>>>

>>>> *Quarkus 2.x is the continuation of 1.x. It’s not a big refactor,

>>>> it’s
>>>> not something totally new. The focus will still be the same:
>>>> supersonic,

>>>> subatomic, fun!*

>>>>
>>>> This is a proposal, Quarkus is an open project so we are open to
>>>> feedback, discussions, and ideas from the community.
>>>>
>>>> Thanks!
>>>>
>>>> For the Quarkus Team
>>>> Clement and Guillaume
>>>>
>>>> --
>>>> You received this message because you are subscribed to the Google
>>>> Groups "Quarkus Development mailing list" group.
>>>> To unsubscribe from this group and stop receiving emails from it,
>>>> send
>>>> an email to quarkus-dev...@googlegroups.com.
>>>> To view this discussion on the web visit
>>>> https://groups.google.com/d/msgid/quarkus-dev/CALt0%2Bo_1m3adZ5KBmVZautJXRh_uZYNgg-_%2BYUBnHVM3pLiq6A%40mail.gmail.com

>>>> <https://groups.google.com/d/msgid/quarkus-dev/CALt0%2Bo_1m3adZ5KBmVZautJXRh_uZYNgg-_%2BYUBnHVM3pLiq6A%40mail.gmail.com?utm_medium=email&utm_source=footer>
>>>> .

>>>>
>>> --
>>> Manyanda Chitimbo.
>>>
>>> --
>>> You received this message because you are subscribed to the Google
>>> Groups
>>> "Quarkus Development mailing list" group.
>>> To unsubscribe from this group and stop receiving emails from it,
>>> send an
>>> email to quarkus-dev...@googlegroups.com.
>>> To view this discussion on the web visit
>>> https://groups.google.com/d/msgid/quarkus-dev/CANwCmXKAtySrdLZpLAMKhi2q45OO1LcXNhyTGvzGZiFs17DTYA%40mail.gmail.com

>>> <https://groups.google.com/d/msgid/quarkus-dev/CANwCmXKAtySrdLZpLAMKhi2q45OO1LcXNhyTGvzGZiFs17DTYA%40mail.gmail.com?utm_medium=email&utm_source=footer>
>>> .

>>>
>> --
>> You received this message because you are subscribed to the Google
>> Groups
>> "Quarkus Development mailing list" group.
>> To unsubscribe from this group and stop receiving emails from it,
>> send an
>> email to quarkus-dev...@googlegroups.com.
>> To view this discussion on the web visit

>> https://groups.google.com/d/msgid/quarkus-dev/CAJLxjVFm3gSPb8Ymkmau%3DjHyZOg2GmqWiV34fV2i4w1P1_0xpA%40mail.gmail.com
>> <https://groups.google.com/d/msgid/quarkus-dev/CAJLxjVFm3gSPb8Ymkmau%3DjHyZOg2GmqWiV34fV2i4w1P1_0xpA%40mail.gmail.com?utm_medium=email&utm_source=footer>
>> .

>>
>
> --
> You received this message because you are subscribed to the Google
> Groups "Quarkus Development mailing list" group.
> To unsubscribe from this group and stop receiving emails from it, send
> an email to quarkus-dev...@googlegroups.com.
> To view this discussion on the web visit

> https://groups.google.com/d/msgid/quarkus-dev/CAJ97idE4NoCRkOjW3AncaL9Fc5Yy78jENK9kWqjyXTcrWQnyxw%40mail.gmail.com.

[Jorge Solórzano]

That’s why we decided to bump the Quarkus version to 2.x once these features will be integrated.

Great news!

We are still evaluating if we could drop Java 8 support in 2.x (1.x will still be compatible with Java 8).

This definitely should be done, drop Java 8 and target 11, this year will be released Java 17, and Mandrel doesn't even support Java 8, so it makes sense to move forward, and what a better time to do this than a major release.

 

[Loïc MATHIEU]

> thinking out loud about
the mongodb as an example if possible
to keep current extension name for easy migration but have it include
the two split ones enabling an easy upgrade path.

@Max Rydahl Andersen yeah, I was thinking about something like this also. I'll open a thread on the mailing list to discuss the MongoDB split.

--

You received this message because you are subscribed to the Google Groups "Quarkus Development mailing list" group.
To unsubscribe from this group and stop receiving emails from it, send an email to quarkus-dev...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/quarkus-dev/2e06c773-8c35-4692-a9cb-84665f11b47bn%40googlegroups.com.

[Kristijan Rusu]

Just sharing my two cents:
If java 8 is dropped, I think that Quarkus should invest in supporting all of Java 11 features. Java modules come to mind, which are currently not supported.
I support the moving forward part and dropping Java 8, but to make a simple ServiceLoader implementation, the implementation would need to use java modules, but the main project should not.

[Davide Di Domenico]

What about possibility to have Kotlin in stable stage? I'm going to create a brand new project where I want use massively reactive stack, with kotlin and the package mutiny-kotlin is possible to write mutiny functions under imperative way, but my thoughts go to the fact that in Quarkus the kotlin support is in previes stage.

[Justin Lee]

We can probably promote the kotlin support for Panache out of experimental at this point.  That's something we'll want to discuss.  I'm currently working on coroutine support for resteasy reactive that will allow you define 'suspend' functions for your endpoints.  That's currently working today in JVM mode but fails in native mode due to missing information for graalvm.  That's what I'm currently fixing and hope to have that resolved soon.  In general, if there's a missing kotlin piece you're running in to, please file an issue and we'll see what can be done.

This isn't me committing anyone to anything, necessarily, but I did at least want to let you know where things stand since you explicitly asked for kotlin support.

To view this discussion on the web visit https://groups.google.com/d/msgid/quarkus-dev/3180e427-e5ad-4047-a6e9-7b025d47eccdn%40googlegroups.com.

[Davide Di Domenico]

Thank you for the response. For sure there are so many pieces of kotlin that is now missing and watching the rare support to kotlin in microprofile world I say WoW Quarkus is doing a very good work. But my question looks forward the actual problems and I'm searching if there is an "insurrance" that kotlin will stay stable in Quarkus world or in some point will be droped because is expensive to maintain? My project will be a long term project in industry and home automation via IoT and I would stard with well established techologies. I choose Quarkus because I love standards and I love how Quarkus implements them, Kotlin is in the some shade of gray between standard and non-standard in my point of view

[Max Rydahl Andersen]

On 7 Apr 2021, at 23:27, Davide Di Domenico wrote:

> Thank you for the response. For sure there are so many pieces of
> kotlin
> that is now missing and watching the rare support to kotlin in
> microprofile
> world I say WoW Quarkus is doing a very good work. But my question
> looks
> forward the actual problems and I'm searching if there is an
> "insurrance"
> that kotlin will stay stable in Quarkus world or in some point will be
> droped because is expensive to maintain? My project will be a long
> term
> project in industry and home automation via IoT and I would stard with
> well
> established techologies. I choose Quarkus because I love standards and
> I
> love how Quarkus implements them, Kotlin is in the some shade of gray
> between standard and non-standard in my point of view

We can't guarantee anything will continue to always work - we can just
say that
we do want to see good Kotlin support and do our best to fix and improve
it.

Justin's work on coroutines should help lot here and I believe many of
the devmode
issues been found.

If you have particular issues in mind let us know the specifics.
Best approach is to open issues and help out.

/max

>>>>>> <https://groups.google.com/d/msgid/quarkus-dev/2e06c773-8c35-4692-a9cb-84665f11b47bn%40googlegroups.com?utm_medium=email&utm_source=footer>
>>>>>> .

>>>>>>
>>>>> --
>>> You received this message because you are subscribed to the Google
>>> Groups
>>> "Quarkus Development mailing list" group.
>>> To unsubscribe from this group and stop receiving emails from it,
>>> send an
>>> email to quarkus-dev...@googlegroups.com.
>>>
>> To view this discussion on the web visit

>>> https://groups.google.com/d/msgid/quarkus-dev/3180e427-e5ad-4047-a6e9-7b025d47eccdn%40googlegroups.com
>>> <https://groups.google.com/d/msgid/quarkus-dev/3180e427-e5ad-4047-a6e9-7b025d47eccdn%40googlegroups.com?utm_medium=email&utm_source=footer>
>>> .

>>>
>>
>
> --
> You received this message because you are subscribed to the Google
> Groups "Quarkus Development mailing list" group.
> To unsubscribe from this group and stop receiving emails from it, send
> an email to quarkus-dev...@googlegroups.com.
> To view this discussion on the web visit

> https://groups.google.com/d/msgid/quarkus-dev/561227c1-ffb8-4f1e-afe0-c156430a7acan%40googlegroups.com.


/max
https://xam.dk/about