In my defense I'd say neither Jersey nor Resteasy has as many many
modules as CXF has, lol :-)
Sergey
On 16/11/17 13:55, Andriy Redko wrote:
+1 to that, also Jersey has RxJava and RxJava2 modules (at least for
the client side).
Thursday, November 16, 2017, 8:51:25 AM, you wrote:
SB> Hi Andriy
SB> Yeah, that is true. The only indirect reference to the fact CXF +
SB> RxJava1 might be combined somehow is that the initial RxJava1 code was
SB> added after a JIRA request was opened.
SB> By the way I've browsed around and found out ReastEasy friends have
SB> RxJava and RxJava2 modules :-).
SB> I guess the only prob with splitting it into tow modules in CXF is that
SB> CXF 3.2.1 is known to ship RxJava2 code in the cxf-rt-rs-extension-rx,
SB> so I guess it would have to be moved to cxf-rt-rs-extension-rx2, and I'd
SB> not be surprised if it would actually be noticed by CXF 3.2.2 users,
SB> given that users like trying newer things...
SB> So perhaps keeping things as is in 3.2.x is the best compromize
SB> Cheers. Sergey
SB> On 16/11/17 13:41, Andriy Redko wrote:
Let's do what is really the best for CXF in short term (long term is obviously
dropping RxJava 1.x). I saw and still see RxJava 1.x in the field, BUT I
haven't
seen the CXF + RxJava 1.x in use yet :) So my arguments are purely based on
assupmtions, not the real facts :-D
SB> It's obviously not only my decision what to do with this code, you are
SB> right it's only my opinion (which will stay non-binding) which is to
SB> keep where it is now just in case and drop it once the new master opens.
SB> To be honest, it does not matter much to me :-), so if few more PMCs say
SB> yes, def has to be a new module - then I'll give my +1 and move on, as I
SB> said purely from a tech point of view a dedicated module without
SB> optional deps is better.
SB> I'm simply hesitating, given how much effort went into dropping some old
SB> modules from 3.2.x, to start with another module with precisely 4 files
SB> (3 in .client subpackage, 1 in .server) with us (me definitely) unlikely
SB> contributing to it at this stage. I'd rather spend the limited amount of
SB> time I have now on growing the small (but with the prospect of growth)
SB> reactivestreams lib we've discussed with John which can be used by
SB> RxJava2 and Reactor code...
SB> Cheers, Sergey
SB> On 16/11/17 12:02, Andriy Redko wrote:
Fair enough, if we the new module is not a option (in your opinion),
I would vote to remove the RxJava 1.x integration and dependency.
SB> As I said, as far as CXF is concerned, there's no prospect of RxJava
SB> related code growing, and contributing to a CXF module noise to support
SB> a legacy library (I know I have to be careful now about the wording:-),
SB> I'm meaning here RxJava2 embracing org.ractivestreams) is not worth it IMHO.
SB> If you check my earlier reply, I suggested to keep it where it is now
SB> after all. So if we have some users somewhere deciding to stay with
SB> RxJava then they'd have the support they need.
SB> Cheers, SErgey
SB> On 16/11/17 11:45, Andriy Redko wrote:
Got it, so "legacy" part is questionable here. Check out the releases page,
https://github.com/ReactiveX/RxJava/releases, the 1.x is still being
actively
supported and maintained (and there are reasons for that, as I
mentioned). So
it is really up to us to decide, should we support it or not, but with
the new
module we could get the stats and make the decision not based on
"legacy" but
if it is used or not. I don't have particular attachments to RxJava 1.x so
if you are confident no one is relying on this integration, I would
agree with
you and we should better remove this code.
*SB> The problem is not about a new module, but about RxJava is a legacy
lib,
SB> and having a module with 2/3 files with no prospect of going beyond
this
SB> number is not worth it IMHO
SB> Sergey
SB> On 16/11/17 11:15, Andrey Redko wrote:
Hey Sergey,
I think the "ideal" in this case depends on whom to ask. For us - yet
another module to support, for users - out of the box integration. With new
module we could collect a bit more insights if people use it or not. No use
- drop in next releases. Thanks.
Best Regards,
Andriy Redko
On Nov 16, 2017 4:42 AM, "Sergey Beryozkin" <*sberyoz...@gmail.com
<mailto:sberyoz...@gmail.com>*> wrote:
Hi Andriy
As I said, introducing a dedicated support for a legacy library in the
form of a new module would not be ideal IMHO
Cheers, Sergey
On 15/11/17 23:53, Andriy Redko wrote:
Hey Sergey,
That would be ideal I think (move RxJava into separate module). RxJava2
and
RxJava are quite different frameworks, some people just stuck with RxJava
so
we could support them there. Thanks.
Best Regards,
Andriy Redko
JDA> What about just leaving the old RxJava code in a module by itself
(when I
JDA> was looking recently, it didn't make much sense to see both RxJava
and
JDA> RxJava2 in one module).
JDA> On Wed, Nov 15, 2017 at 10:56 AM Sergey Beryozkin <
*>>>> sberyoz...@gmail.com <mailto:sberyoz...@gmail.com>*>
JDA> wrote:
Hi
cxf-rt-rs-extension-rx ships the code for both (old) RxJava and RxJava2
code. It supports returning RxJava2 Flowable and Observable on the
server and accepting it on the client, and the same for the (old) RxJava
Observable...
While even the (old) RxJava code is very new for CXF, the reality is
that RxJava has been around for a while now and with RxJava2 embracing
org.reactivestreams, it's hard to see CXF users preferring to start with
the (old) RxJava.
The other minor problem is that cxf-rt-rs-extension-rx has optional
RxJava and RxJava2 deps to be able to ship the relevant code in the same
module and splitting it into 2 modules will be too much at this point.
I suggest that unless some users confirm (I CC to the users) that they
need to use the (old) RxJava code, then we just remove it and make
things much simpler...
Thanks, Sergey
*
