Efficient COBOL coding

kalaignar_S · New User Joined: 06 Jan 2006 Posts: 21

Which one is the efficient COBOL coding, has been mentioned below??
Why??

MOVE A TO B.

COMPUTE B = A.

If anybodies know this let me know the reason..

Thanks
Kalaignar

ralph_v · New User Joined: 07 Jul 2005 Posts: 27

COMPUTE B = A is efficient coding because you can check for truncation using ONSIZE ERROR clause.

parikshit123 · Active User Joined: 01 Jul 2005 Posts: 269 Location: India

Well,

When you say 'efficient coding', offcourse COMPUTE should be a better choice. You can do more with that. for example you can check for size or even overflow....

But, when it comes to "efficient code", if you ignore exceptions, MOVE is efficient. as it maps to less number of instructions.

I hope the difference between "efficient coding" and "efficient code" is clear.

parikshit123 · Active User Joined: 01 Jul 2005 Posts: 269 Location: India

Also,
I would suggest you to choose a good title for the topic.

I feel that "Efficient Coding" is too general for this topic.

However, it was really a nice question.

Thanks,
Parikshit

DavidatK · Posted: Wed May 10, 2006 10:05 pm

Kalaignar,

It's interesting that in the specific example you give, the compiler generates the exact same code for move and compute.

parikshit123 · Active User Joined: 01 Jul 2005 Posts: 269 Location: India

Hi David,

Thats correct!
the specific example generates the same code.

Now, suppose if you are using ONSIZE ERROR clause then compiler will definitely generate more number of instructions. isn't it?

So, only an assignment is your need, offcourse simple move is a better choice.

Any further ideas.. seems to be a nice topic isn't it?

Thanks,
Parikshit.