Understanding the long ID?

[Richard Gaskin]

Mark Talluto wrote:

> On May 15, 2007, at 3:55 PM, Chipp Walters wrote: 
>> I'm using the latest release of Rev 2.8.1
>> I create a new stack, add a single fld and a single button. In the  
>> button script I add:
>>
>> on mouseUp
>>  put the long ID of fld 1 into tID
>>  set the disabled of tID to not the disabled of tID
>>  put "fred" into tID
>> end mouseUp
>>
>> When I press it, the field becomes disabled alternately (as  
>> expected). The disabled is only here to show that tID is 
>> referenced correctly.
>>
>> But, "fred" never gets put in the field? I wonder why? Is this a bug?
>>
>> BTW, this does work:
>>
>> set the text of tID to "fred"
> 
> 
> I think it is a bug.

If Rev variables were types and tID was typed as an object reference it 
might be a bug.

But as it is, Rev vars are just buckets into which you can put anything 
you want.  One minute you can store a string that refers to an object, 
the next minute you can store some arbitrary text -- how is Rev to know 
you mean otherwise?

Rev vars don't have properties, but objects do, so when you say:

   set the text of tID to "fred"

...it knows that you're setting a property so it first evaluates tID to 
get the object reference, then sets that object's property.

The ambiguity is -- once again -- HyperCard compatibility.  In most 
languages, properties are set in a consistent way, and in Rev the text 
property can be set using "set the text of...".

But HyperCard allowed the convenience of setting the text property 
without having to name the property by using "put <val> into <obj>".

While it's convenient to treat the text property of objects as an 
implied container, the ambiguity inherent in HyperTalk's design has led 
to a great many confusing moments for many people over the last couple 
decades.

-- 
  Richard Gaskin
  Fourth World Media Corporation
  ___________________________________________________________
  Ambassador at FourthWorld.com       http://www.FourthWorld.com