Flat file adapter - sequence of records

, ,
1

Guys,

I created a FF schema to parse a record delimited file. convertToValues works fine if all the record types appear in sequence in the file. I mean if Record Type 1, Record Type 2 and Record Type 3 are the three types of records that can occur any number of times in the file, then the FF schema works as expected (it creates Record Type 1, Record Type 2, Record Type 3 one after the other) if the file looks like this:

Record Type 1
Record Type 1
Record Type 1
Record Type 2
Record Type 2
Record Type 3
Record Type 3
Record Type 3
Record Type 3
…

It comes up with a strange hierarchical document (and ffvalues) if the records appear in a mixed order, like:

Record Type 1
Record Type 2
Record Type 3
Record Type 1
Record Type 2
Record Type 3

Is there a setting somewhere that I am missing? Would much appreciate any advice on this. Thanks in advance!

2

hi sunil,
if u need the RecordTypte to repeat any no of times,no need to create RecordType1,Recordype1…like that.
just create a RecordType1 select the maxRepaet as ‘unlimited’ for that record.which u can see in the properties.

just try this.

Sree.

3

I can get upto 10 different record types in this file. I used 3 above for simplicity. I created 10 different record definitions with max repeat set to ‘Unlimited’ in my FF schema. Each of the record types has its own identifier which is char 14 and 15.

4

The sequence of the data being processed by the schema is going to influence this.
Can you post a sample?

5

ok if the sequence is right what u specified …then its such a simple scenario…
so u get
record 1
record 2
record 3
record 1
record 2
record 3
------- unlimited times
so u create ur schema this way
record 1(unlimited)
—record 2
—record 3

and run ur schema it should work fine …let me know if the scenario is different from what i imagined…
HTH
sri