Great! With this we would be able to support Rx programming model and Stream
communication.
I’d like to give a review on it.
Jun
> On Mar 7, 2019, at 3:05 PM, yuneng xie wrote:
>
> Hi folks,
>
> the first stage of the reative-stream support has been finished. we now can
> using Mono/Flux as
Hi folks,
the first stage of the reative-stream support has been finished. we now can
using Mono/Flux as return value. i have raised a pr :
https://github.com/apache/incubator-dubbo/pull/3609 . all code is in the
module called dubbo-rpc-rsocket.
i would try to complete the feature so that we
Agree. Pls. keep discussing the wish list for dubbo 3.0. I will init a
project on GitHub, and keep collecting good ideas into it.
-Ian.
On Sun, Feb 3, 2019 at 6:34 PM Kun Song wrote:
> An issue will be fine, maybe we can discuss features on mail list, and add
> more serious conclusions in that
I guess what Taosheng propose here is a DSL to build REST apis, DSL is good,
confluent and concise.
> 在 2019年2月3日，下午3:24，jun liu 写道：
>
> Hi, Taosheng
>
>> I think we can provide functional api to users for configuring dubbo,
>> comparable to xml or annotation configuration. Just like spring
An issue will be fine, maybe we can discuss features on mail list, and add more
serious conclusions in that issue. What others think?
> 在 2019年2月2日，下午7:10，Imteyaz Khan 写道：
>
> +1 in the suggestion.
>
> On Saturday, February 2, 2019, Ian Luo wrote:
>
>> I am thinking of firing an issue on
Hi, Taosheng
> I think we can provide functional api to users for configuring dubbo,
> comparable to xml or annotation configuration. Just like spring 5 provide
> both controller configuration and functional api for http
Sounds pretty good, only functional bean registration is a new concept
I think we can provide functional api to users for configuring dubbo,
comparable to xml or annotation configuration. Just like spring 5 provide both
controller configuration and functional api for http.
-- Original --
From: Kun Song
Date: Sat,Feb 2,2019 6:12
+1 in the suggestion.
On Saturday, February 2, 2019, Ian Luo wrote:
> I am thinking of firing an issue on Github to collect 3.0 wishlist from the
> community. What do you say?
>
> -Ian.
>
>
>
>
> On Sat, Feb 2, 2019 at 6:12 AM Kun Song wrote:
>
> > I especially interested in reactive
I am thinking of firing an issue on Github to collect 3.0 wishlist from the
community. What do you say?
-Ian.
On Sat, Feb 2, 2019 at 6:12 AM Kun Song wrote:
> I especially interested in reactive programming and cloud native support.
>
> A reactive Dubbo will be attractive, as the community
Hi,
On Sat, Feb 2, 2019 at 6:12 AM Kun Song wrote:
>
> I especially interested in reactive programming and cloud native support.
>
> A reactive Dubbo will be attractive, as the community point out. I even
> propose to make Dubbo implements the Reactive Streams specification, which
> will
I especially interested in reactive programming and cloud native support.
A reactive Dubbo will be attractive, as the community point out. I even propose
to make Dubbo implements the Reactive Streams specification, which will
integrate back pressure(flow control) and make Dubbo a choice for
I especially interested in reactive programming and cloud native support.
A reactive Dubbo will be attractive, as the community point out. I even propose
to make Dubbo implements the Reactive Streams specification, which will
integrate back pressure(flow control) and make Dubbo a choice for
Agree, we should consider seriously both HTTP/2 and rsocket.
Thanks,
-Ian.
On Thu, Jan 24, 2019 at 10:59 AM Taosheng, Wei
wrote:
> Lan,
> Yes, I think http2 and some new protocols such as rsocket can be
> considered.
> I will spend some time to study this issue.
>
>
> Warm regards,
> Taosheng
Lan,
Yes, I think http2 and some new protocols such as rsocket can be considered.
I will spend some time to study this issue.
Warm regards,
Taosheng
-- Original --
From: Ian Luo
Date: Thu,Jan 24,2019 9:58 AM
To: dev
Subject: Re: [DISCUSS]: Some thoughts on
Hi Zhiguang,
> In this condition we use dubbo on internet and keep security by security
> rules that only allow the servers connect to each other.
I want to know how do you apply the security rules to Dubbo. Can you provide
more detail on this?
Jun
> On Jan 23, 2019, at 5:47 PM,
Taosheng,
In this scenario, it looks like we should use http2 to transport the
payload, what do you think?
Thanks,
-Ian.
On Wed, Jan 23, 2019 at 10:35 PM Taosheng Wei
wrote:
> I think we can find a binary protocol with strong potential to be a public
> application protocol like http, and
I think we can find a binary protocol with strong potential to be a public
application protocol like http, and extend it with security function. Or if
there aren't such suitable protocols, we can try to formulate a new
protocol. Then make Dubbo support it.
In my opinion, this way may not only
If the RT is acceptable between the data center, this usage is perfectly
fine.
-Ian.
On Wed, Jan 23, 2019 at 5:47 PM zhi_guang_...@163.com
wrote:
> I have a similar Question as this mail:
> Is Dubbo designed for use on internet?
> I have just join a company last year and our business is all
I have a similar Question as this mail:
Is Dubbo designed for use on internet?
I have just join a company last year and our business is all around the world.
So we have servers on US and ASIA and EU.
In this condition we use dubbo on internet and keep security by security rules
that only allow
> - Reative programming and its fundamental FP start to get adopted. We
should consider to support it.
It looks like we're thinking about "full stack asynchronization" , I think
the Spring4 shoud be upgraded to spring5
Although Spring is an optional component in Dubbo, but spring 5 provides
Thanks for Yuhao raising this good point. In my opinion, both ACL and
transaction are important to Dubbo, and the community keep demanding the
support on these areas. We should put them in our roadmap.
Thanks,
-Ian.
On Tue, Jan 22, 2019 at 10:55 PM Yuhao Bi wrote:
> Hi lan and community,
>
>
Hi lan and community,
Although I have already heard "Dubbo" a few years ago,
but I just started to learn dubbo after the meetup last year in Chengdu
after it became the Apache Dubbo.
Maybe I'm not such that familiar with the underlying details, but after
the continuous participated
I feel like a
> I think the online integration test and performance test environment should
> be set up for the new features.
Agree! We should start as soon as possible, from 2.7.x.
Jun
> On Jan 22, 2019, at 5:43 PM, Xin Wang wrote:
>
> I think the online integration test and performance test environment
I think the online integration test and performance test environment should
be set up for the new features.
Ian Luo 于2019年1月22日周二 下午3:04写道：
> Yuhao, good idea.
>
> BTW, do you have any thought on what Dubbo 3.0 should be?
>
> Thanks,
> -Ian.
>
>
> On Tue, Jan 22, 2019 at 2:39 PM Yuhao Bi
Hi Yuneng,
Sounds interesting. I am especially interested in reactive programming
support. Pls. go ahead to try implement it on 3.x branch.
Thanks,
-Ian.
On Thu, Jan 17, 2019 at 11:03 AM yuneng xie wrote:
> Hi folks,
>
> I agreed with Ian Luo on the improvement list. I also got some idea in
Hi folks,
I agreed with Ian Luo on the improvement list. I also got some idea in my
mind. I'd just share with you two points below in detail which i'm most
interested in right now.
1. Upgrade the core abstraction "Invoker", which works in sync mode, to an
abstraction works in async mode. then
What I am thinking about the next generation of Dubbo should be:
- cloud native: should be well integrated into Kubernetes, and Dubbo
protocol should be seamlessly go through service mesh sidecar like
Envoy.
- reactive: should keep the RPC invocation completely reactive
- polyglot: Dubbo service
27 matches
Mail list logo
