Yes I think I confused myself on this one (for my defense I got a *strong* cold 
recently and I'm just recovering :p)

Inline...

Le 14/10/2015 14:05, Ron Wheeler a écrit :
+1 on Tahar Alkhateeb's points.

I don't think that anyone is proposing to convert all existing code to use the 
latest Java 8 features.

I think that once you force people to run the Java 8 JVM by doing the compile 
with Java 8, you are pretty much done.

If someone adds a Stream function in their code and the code is validated to 
work correctly, that should be the end of the story.
We don't vote on other code changes.

Who is going to check new patches to ensure that they do not depend on Java 8 
features or bugs fixed in Java 8?

As explained in my answer to Taher, we need to be sure we don't backport Java 8 features in released branches which are supposed to work with Java 7, that's all.


What was the process when Java 7 features where allowed into the trunk? Was a 2 
part vote required?

Nope, mistake on my side

Jacques


Ron

On 14/10/2015 6:17 AM, Taher Alkhateeb wrote:
Hi Jacques,

I am a bit confused? Why wait for a frozen branch for anything? I thought trunk is your development/unstable bleeding-edge part of the code. This is where you introduce changes without worrying about people's stable systems because they base their systems on release branches. In fact, branches exist only to stabilize trunk at certain points of time.

Furthermore, if we introduce Java 8 into trunk, what possible logical reason exists for _voting_ to use the new features of the language? Why did you upgrade in the first place if you do not intend to use the newer features?

Taher Alkhateeb

----- Original Message -----

From: "Jacques Le Roux" <jacques.le.r...@les7arts.com>
To: dev@ofbiz.apache.org
Sent: Wednesday, 14 October, 2015 12:23:31 PM
Subject: Re: Java 8 and functional programming in trunk

I tend to agree. Actually there are 2 points,
1) use Java 8 to compile
2) use new Java 8 features

We can already do 1, but we should maybe wait the next freezed branch to do 2. 
That should be discussed but does not prevent 1. Of course any commit
with 2 will be rejected as long as an agreement on 2 will not be reached

Jacques

Le 11/10/2015 19:09, Ron Wheeler a écrit :
I am not sure why backwards compatibility is an issue with moving to Java 8.

It works well,
it is the supported version of Java.
As far as I have heard, the Java 8 JVM runs code compiled with earlier versions 
of Java. (My own experience is uniformly positive).

I can see no reason to think that anyone who is developing or using OFBiz would 
need to compile or run with an older version of Java.

In this day of VM and Docker Containers, it is hard to imagine that someone 
would be stuck on an older JVM in a production environment and unable to
come up with a configuration that would allow them to run JVM 8 to support 
OFBiz.

IMHO, we should just move to requiring the supported version of Java as soon as 
possible.

Ron


On 11/10/2015 4:14 AM, Taher Alkhateeb wrote:
Hello Everyone,

We created and provided the patches for this issue since June and OFBiz is 
pretty much ready to move to JDK 8. I'm not sure if lazy consensus is
enough or whether a vote is warranted to move this issue forward?

Cheers,

Taher Alkhateeb

----- Original Message -----

From: "Pierre Smits" <pierre.sm...@gmail.com>
To: dev@ofbiz.apache.org
Sent: Thursday, 7 May, 2015 10:52:00 AM
Subject: Re: Java 8 and functional programming in trunk

I don't think that it is a question of using or not using J8. It is more
about when to move it into trunk, as from that moment on there will be
backward compatibility issues.
A helpful solution in this respect could be to designate a specific release
branch now (e.g. r15 or r16), create the version in JIRA and setup the dev
branch in svn. Doing it that way J8 related issues can get registered,
changes can get implemented and assessed (in the dev branch) without
breaking current stuff in trunk.

But it also ensures that we can create awareness prior to releasing stuff.

Best regards,



Pierre Smits

*ORRTIZ.COM <http://www.orrtiz.com>*
Services & Solutions for Cloud-
Based Manufacturing, Professional
Services and Retail & Trade
http://www.orrtiz.com

On Thu, May 7, 2015 at 9:30 AM, Nicolas Malin <nicolas.ma...@nereide.fr>
wrote:

Ok Thanks Scott and Jacques.

So Who is against use java 8 and more on trunk ?

:)

Nicolas


Le 07/05/2015 08:46, Jacques Le Roux a écrit :

Yes (lazy) consensus over vote ;)

Jacques

Le 07/05/2015 05:02, Scott Gray a écrit :

I'm not sure if a vote is necessary, particularly if no one has any
objections.

Regards
Scott
On 7 May 2015 07:44, "Nicolas Malin" <nicolas.ma...@nereide.fr> wrote:

I'm favorable to use java 8.
I think it's will be pretty fin if you can support oracle jdk8 and
openjdk8 also.

I propose to organize a vote to validate or not this proposition

Nicolas

Le 03/05/2015 11:52, Jacques Le Roux a écrit :

Hi Taher,
Yes I think so. For now well known (I hope ;)) security reasons, if
people want to use Oracle JDK they need to use Java 8.
So implementing with new Java 8 features now in trunk sounds good to
me.
BTW this is only my opinion...

Note that our demos are still using OpenJDK 1.7 I'm not quite sure of
the
policy

http://www.cvedetails.com/vulnerability-list/vendor_id-93/product_id-23642/version_id-138218/Oracle-Openjdk-1.7.0.html



Jacques


Le 03/05/2015 10:52, Taher Alkhateeb a écrit :

Hi everyone,
I would like to work on a few JIRAs and at the same refactor some
existing
code to utilize Lambdas and Streams utilizing Java 8 features.


Is it acceptable to develop with JDK 8 features into trunk?

Taher Alkhateeb







Reply via email to