Re: Suggestion for improving loop performance

On 19-May-2015 14:40 -0500, Singh, Ashwani (Lake Mary) wrote:

On 19-May-2015 14:30 -0500, Vinay Gavankar wrote:

Also, what is the maximum number of | you are expecting? You could
try reducing the Array size. I think Scan will have to initialize
the remaining elements of array for every record. <<SNIP>>

the number of | may differ with each sub-record within a record.

We have one control record running into multiple records of flat file
i.e.

X,1:01 A|B|C"
X,1:02 D|E|F"
X,1:03 G|H|I

X,2:01 1|2|3|4|5
X,2:02 ......

And i know that this record X:01 should contain 5 field's but have 2
missing and rest all are fine (It could vary for each record)

The output becomes like this

X,1: A|B|C||D|E|E|G|H|I

This whole process is repeated like this...

But if the count of pipe characters must only ever reach 99, then do not declare an array of 999, because the SCAN instruction will reinitialize 900 more elements than necessary each iteration.

Again, the described output seems askew from the description.

I am unsure how the new description of the record data with a prefix of "X,1:01" is any clearer than "X:01"? If the data is described more explicitly in a structure language [so unlike English], e.g. as SQL data using DDL and DML, there is so little speculation as to what is the meaning attempted to be conveyed. If the data is in a stream file, then after the SQL code showing the DDL creating the TABLE into which the INSERT populates that file, then show the CPYTOSTMF as the final step; i.e. use the Control Language (CL) as the structured [non-English] means to describe the scenario.