(j3.2006) Chaining type-bound procedure references

[Damian Rouson]

On May 1, 2014, at 10:48 AM, Tom Clune <Thomas.L.Clune at nasa.gov> wrote:
> 
> I want to back off from my conservative statements yesterday.   I am finding a significant category of use cases where the ASSOCIATE construct will definitely help.    I was blinded by the simplicity of my synthetic test cases, but real use cases seem to take enough advantage of the association to amortize the extra lines of code.

In the hopes that this thread won?t drop, I?d just like to say that the ASSOCIATE approach, while it?s a reasonable workaround if it doesn?t get too deeply nested, it will likely get very deeply nested in some of my use cases and, in those cases, it will greatly reduce the code clarity because I?m trying to write a formula that would easily recognizable if it weren?t broken up in this way.  I still hope Tom?s original suggestion becomes possible at some point.

Damian