Gregory,

If this were 40 years ago I would right away jump to every suggestion (or idea) brought about and start coding. But now that I am older and wiser (some may disagree though) I am thinking of all possibilities and mainly about creating a maintenance nightmare (which I am very good at). There is one more option that we didn't discuss; I could say to customer, Sorry, I can't do it. I am learning to say No; it is a work in progress and painful but in the end very rewarding <g>.

I will consider, of course, everything you suggest. Thank you very much!

>Answers in line
>
>
>>Gregory, see my replied under your comments:
>>
>>>Dmitry,
>>>
>>>Your test shows or demonstrates the normal behaviour
>>>
>>>A couple of possibilities
>>>
>>>(1) Change the properties of the controls you want in the form's init. I do not see a problem there
>>
>>This is true. But I wanted the property set before a control's Init fires.
>
>That is difficult - maybe addObject can
>
>
>>
>>>
>>>(2) You could make some classes of controls to look at form properties, ie Adding properties to the form in the form's load(), then each of those controls could find out whether a certain property exists and use it in its init
>>
>>This is a possibility. And I already tested this approach although Naomi found some holes in it.
>
>No holes
>
>
>>>
>>>(3) BindEvent the init of the controls (post flag of bindevent) to a form's method, ie
>>>Change each control control class to bindEvent itself in the init to a form's method (not tested)
>>>Then in the form's method you can find out what control it is and act accordingly
>>
>>I have never used BindEvent yet. And I am concerned that I will do something wrong (due to inexperience) and break the app. I prefer to spend some time to learn it first.
>
>There's always a first time. It's worth the time investigating
>
>ps: Add this to the destroy of all your control classes + form class
>[ I have once read - but cannot confirm it - that the system may become unstable if you do not unbind - better safe than sorry]
>
>
>

>try
>	=UnbindEvents(m.this)
>catch
>	assert .f.
>
>endtry
>

>
>
>>So I will probably go with approach 2 unless I find other issues (on top of what Naomi already suggested).
>>
>>Thank you very much.