Maybe you could share your ideas, and even work, in a Jira?


Le 01/12/2016 à 10:53, Taher Alkhateeb a écrit :
I am working on this

On Dec 1, 2016 12:48 PM, "Jacques Le Roux" <jacques.le.r...@les7arts.com>

Totally agreed!

BTW Gareth Cater told me he did a work related to "custom-widget" (he used
the same term but not sure it's the same thing). I'll try to contact him
about that. Has anybody else begun to work on that?


Le 01/12/2016 à 10:38, Taher Alkhateeb a écrit :

Hi Jacques,

That was already discussed. My opinion is still not to allow it BUT, you
can create a DSL for something, let's call it custom-widget. Then, all
you need to do to drop down to FTL is to create macros in the theme to
implement your special widget.

This way, you can maintain purity of the widgets while at the same time
allowing you to muck with HTML. In a sense, we tell our developers, do
whatever you want, OUTSIDE. So this custom-widget becomes the gateway to

On Thu, Dec 1, 2016 at 12:28 PM, Jacques Le Roux <
jacques.le.r...@les7arts.com> wrote:

That would be good, I know we can do a lot with form widgets backed by js
scripts and I'm always been a form widgets enthusiast (if not fanatic

But there should be also a way to allow to call FTL from screens because,
in an ecommerce alike situation, it's not realistic to do it all with
widgets (at least as is now). Could be allowed on certain component for
instance or using properties.


Le 01/12/2016 à 09:56, Taher Alkhateeb a écrit :

I think by introducing a new DSL we can enforce no leakage of HTML / FTL
into any widgets (I'm assuming this is what you guys are talking about)

On Thu, Dec 1, 2016 at 11:49 AM, Jacques Le Roux <
jacques.le.r...@les7arts.com> wrote:

Nicely said, thanks Julien :)


Le 01/12/2016 à 09:46, Julien NICOLAS a écrit :

Hi Pierre,

I hope that, like code source convention, people will respect the work
done and respect people behind them.

I think that I'll do my best to explain the reason of the work I'll
begin, hope that it will be accepted by the community, hope that it
will be
implemented in all OFBiz screens. I know that the OFBiz community are
people and respect that.

But you true, it could happens that somebody fail to follow rules, I
I see it in code review and ask for an update ;)

I'm quite sure that when you'll be charmed by this UI standard, you'll
also a rules keeper <3

Have a nice day,


On 30/11/2016 21:16, Pierre Smits wrote:

So when you speak of

a super-structure that will be used in place of currently conventions
are not always respected

how do you envision that with that new 'super-structure' conventions
be respected?

Best regards,

Pierre Smits

ORRTIZ.COM <http://www.orrtiz.com>
OFBiz based solutions & services

OFBiz Extensions Marketplace

On Wed, Nov 30, 2016 at 9:00 PM, Jacques Le Roux <
jacques.le.r...@les7arts.com> wrote:


Inline ...
Le 30/11/2016 à 10:02, Julien NICOLAS a écrit :

Hi Jacques,

On 30/11/2016 08:51, Jacques Le Roux wrote:
     - Each screen must be linked to a screen structure.

What would be this screen structure? You don't need to develop much
this stage, just that I can't vision what it would be.

This structure is to follow the a UI standard that can be managed

theme. For example, the find screen can be define as :
     - A research field area
     - A result area

Ah, I see, we have already this concept in screen widget.

If all the find screen could be linked to this structure, it will

easier for theme to manage it's own template of search screen.
You mean that it would be a super-structure that will be used in

currently conventions which are not always respected, I see.

It will be included in the main decorator that will also linked to a

structure, so theme can manage to change the template. And when you

the theme, it could be a completely different look and feel :)
I hope I explain well my thought.

Got it, thanks :)

Why we need a new component to test new theme ?
When I start working with OFBiz, I was so surprised that the UI is
heavy. Then I was thinking that I have to improve the UI to
a best
one. After several try I understand that the actual UI is not a
final user
interface. It is a developer one. It's a developer UI because it
all the features developed. But definitively, we can't provide
kind of
UI to final users, we have to simplify it. In the same times, we
delete the current UI because developers need to improve it with
features that will help us to deploy new features to our final

For this new component, we can implement an existing component
simplified and ready for the new theme(s).

You mean we could take and existing component, say example for

and would simply it at the UI level. I picked example because it's
rather simple and contains demonstration of features.

No, I mean to define a component (party, product, facility, etc.)

that we
start to re-implement (using the existing services) but in a more
way (without all the features, selecting only the main ones).

I see

In conclusion, if the new component dedicated for test a new theme
match to the community needs, Taher think to a super simplified
user interface that facilitate developers to improve the
interface without any constraint that allow developers to develop
new features.

Another thing I can't vision at this stage, no hurry, I guess


Yes, too many thing to explain, I have to add details about this

I'll do it soon ^^

I did not get a chance to look yet at the POC Nicolas, Gil and you

working on. I guess I'd get the ideas from there then?


Reply via email to