That sounds cool, I'll check it out at some point.  Note though that with Jelly 
you could in theory drop in/override virtually any tag and not just entire 
widgets i.e. a custom field type for forms (although I'm not sure that what 
you've described doesn't allow for something like that).

Regards
Scott

On 17/01/2011, at 12:41 PM, Adrian Crum wrote:

> Cool - thanks.
> 
> Btw, we can drop in jars for screen widget extensions now. I haven't made the 
> change necessary for the other widgets (forms, menus, trees), but it would be 
> pretty easy to do by following the same factory pattern.
> 
> -Adrian
> 
> --- On Sun, 1/16/11, Scott Gray <[email protected]> wrote:
>> My current leaning would be to
>> replace them with widgets implemented with Jelly and I would
>> like to do a PoC "single" form widget replacement when I
>> have a chance.  The idea of being able to drop in jars
>> for instant widget extension really appeals to me and may
>> help foster new widgets and extensions outside of our code
>> base.
>> 
>> Regards
>> Scott
>> 
>> HotWax Media
>> http://www.hotwaxmedia.com
>> 
>> On 10/01/2011, at 5:11 AM, Adrian Crum wrote:
>> 
>>> Scott,
>>> 
>>> Any more thoughts on this?
>>> 
>>> -Adrian
>>> 
>>> --- On Mon, 5/3/10, Adrian Crum <[email protected]>
>> wrote:
>>>> --- On Mon, 5/3/10, Scott Gray <[email protected]>
>>>> wrote:
>>>>> On 4/05/2010, at 4:23 PM, Adrian Crum
>>>>> wrote:
>>>>> 
>>>>>> --- On Mon, 5/3/10, Adrian Crum <[email protected]>
>>>>> wrote:
>>>>>>> --- On Mon, 5/3/10, Scott Gray <[email protected]>
>>>>>>> wrote:
>>>>>>>> On 4/05/2010, at 11:32 AM, Adrian
>>>>>>>> Crum wrote:
>>>>>>>> 
>>>>>>>>> On 5/3/2010 4:25 PM, Scott
>> Gray
>>>> wrote:
>>>>>>>>>> Sometimes I think using
>> the same
>>>>> schema and
>>>>>>> class
>>>>>>>> for single forms and list forms
>> holds us
>>>> back
>>>>> from
>>>>>>>> implementing features specific to
>> one or
>>>> the
>>>>> other and
>>>>>>> even
>>>>>>>> when we do it results in a messy
>> schema.
>>>>>>>>>> 
>>>>>>>>>> What if we were to
>> separate the
>>>> two
>>>>> but have
>>>>>>> them
>>>>>>>> share a common base?  We
>> could start by
>>>>> separating
>>>>>>> the
>>>>>>>> schemas and then work on the
>> code.
>>>>>>>>>> 
>>>>>>>>>> Thoughts?
>>>>>>>>> 
>>>>>>>>> Any effort to clean up and
>> improve
>>>> widget
>>>>> code
>>>>>>> gets a
>>>>>>>> big thumbs-up from me.
>>>>>>>>> 
>>>>>>>>> While we're at it, could we
>> separate
>>>> the
>>>>> styling
>>>>>>> into
>>>>>>>> a separate element and Java class?
>> Chris
>>>> Howe
>>>>> had
>>>>>>> suggested
>>>>>>>> that some years ago.
>>>>>>>> 
>>>>>>>> I'm not entirely sure what you
>> mean on
>>>> this
>>>>> one? 
>>>>>>>> Could you give me an example of
>> what the
>>>>> problem is
>>>>>>> and how
>>>>>>>> this would solve it?
>>>>>>> 
>>>>>>> I think it was around the beginning of
>> 2007.
>>>> The
>>>>> idea was
>>>>>>> to move all of the form widget
>> styling
>>>> attributes
>>>>> to a style
>>>>>>> element - to cut down on the amount of
>> form
>>>>> attributes. We
>>>>>>> made a preliminary move in that
>> direction by
>>>>> putting all of
>>>>>>> the form styling attributes on their
>> own line
>>>> - so
>>>>> that
>>>>>>> later they could be contained in a
>> separate
>>>>> element.
>>>>>> 
>>>>>> I can't find the ml discussion, but here
>> is the
>>>>> related Jira issue:
>>>>>> 
>>>>>> https://issues.apache.org/jira/browse/OFBIZ-760
>>>>>> 
>>>>>> -Adrian
>>>>> 
>>>>> Okay thanks, I see what you mean now. 
>> I'll stew on
>>>> it
>>>>> for a bit and see what I can add to the
>> discussion.
>>>> 
>>>> There is a chance it isn't needed any more. The
>> problem 3
>>>> years ago was that there was a form attribute to
>> style
>>>> nearly every element in the form. I introduced the
>> concept
>>>> of having a container CSS class that would style
>> the form
>>>> and all of its child elements (using descendant
>> selectors).
>>>> After that the effort to move styling to a
>> separate element
>>>> fizzled out. I don't know if that was due to lack
>> of
>>>> interest or the new container method of styling.
>>>> 
>>>> -Adrian
>>> 
>>> 
>>> 
>>> 
>> 
>> 
> 
> 
> 

Attachment: smime.p7s
Description: S/MIME cryptographic signature

Reply via email to