Re: OCI vs. Pro*C

From: Al Parker <al.parker_at_canrem.com>
Date: Sun, 9 Jan 94 18:47:00 -0400
Message-ID: <60.8198.3082.0N190827_at_canrem.com>


-> Why would you NOT want to use Pro*C? The OCI calls are quite
-> low-level. Using Pro*C just saves a lot of grunt work.
->
-> --
-> Bruce MacDonald
-> Database Systems Consultant
->
-> Email: Bruce_MacDonald_at_mindlink.bc

Part of the reason may be speed - from what I've seen, the OCI version of equivalent software uses fewer function calls that Pro*C and can be better optimized by the programmer. Does anyone have any statistics comparing the two to bear this out ?

Also, it seems to me that it is less convenient to handle LONG datatypes in Pro*C. I believe you can't retrieve a portion of a LONG datatype which you can do in OCI. A typical strategy is to store the length of the LONG data in its first few bytes so that arbitrary size (up to 2 Gig) storage can be allocated based on reading just the first few bytes.

If this is a non-issue I would be interested to hear why...

Al Parker, Andersen Consulting (TIS Etobicoke, Canada) al.parker_at_canrem.com Received on Sun Jan 09 1994 - 23:47:00 CET

Original text of this message