Developer Annoyance #1

[Arthur Urban]

Thanx for the alternate examples you provided. I addressed their 
shortcomings in other posts in the thread, so I won't reiterate. :)

While I will continue to entertain other "solutions", the main thrust of 
my initial post was to propose a new, clearer syntax for message 
calling. One I think Rev could handle easily.

As function:

  put calc(varA,varB) of btn "source" into localResult

As command/message

  send calc varA,varB to btn "source"

Clearly, everything between 'send' and 'to' is the message followed by it's arguments; very simple to parse. Why send appears to be limited to a single string argument (under the hood), is odd to me. Even if varA was a chunk expression (char 1 to 2), Rev is smart enough to know that you can't send a message to a chunk expression, ergo keep looking for an object reference when parsing.

I think if developers and casual coders could easily assemble complex messages, they would be more inclined to keep the behavior with the data, leading to greater reusability. I imagine I'm a voice in the wilderness here.


Jim Ault wrote:
> Ooops, I got the wrong idea from your use of value(merge... etc
>
> You can use the 'send' or 'call' commands to use a function outside the
> message hierarchy, even in another stack.
>
> As Andre has shown
>
> send "calc VarA, VarB" to btn "source"
> put the result into localResult
>
> where...
> on calc VarA, VarB
> ...do work here
> return theVal
> end calc
>
> or
>
> insert the script of btn "source" into back
> --now the handler(s) are in the message hierarchy
>  put calc(varA,varB) into localResult
> remove the script of btn "source" from back --now or later
>
> Jim Ault
> Las Vegas
>
> On 3/16/06 9:43 PM, "Arthur Urban" <aturban at qwest.net> wrote:
>
>   
>> I think a new syntax needs to be made available for calling functions
>> outside the message hierarchy. The existing mechanism does not lend
>> itself to high readability or maintainability. Consider the following,
>> if you will:
>>
>>   put 1 into varA
>>   put 2 into varB
>>   put value( merge("calc([[varA]],[[varB]]))", btn "source" ) into
>> localResult
>>
>> further, if I need to protect the integrity of the variable passed it
>> begins to look like this:
>>
>>   put "1,2,3" into varA
>>   put 4 into varB
>>   put value( merge("calc([[quote&varA&quote]],[[varB]]))", btn "source"
>> ) into localResult
>>
>> I'm not sure that this is the right direction for any language to be
>> headed. Please explain why we can't have the following syntax:
>>
>>   put calc(varA,varB) of btn "source" into localResult
>>
>> _______________________________________________
>> use-revolution mailing list
>> use-revolution at lists.runrev.com
>> Please visit this url to subscribe, unsubscribe and manage your subscription
>> preferences:
>> http://lists.runrev.com/mailman/listinfo/use-revolution
>>     
>
>
> _______________________________________________
> use-revolution mailing list
> use-revolution at lists.runrev.com
> Please visit this url to subscribe, unsubscribe and manage your subscription preferences:
> http://lists.runrev.com/mailman/listinfo/use-revolution
>
>
>