Messages in MQ Series

[DSmamta]

Hello:

Having froblems in MQ series stage. Would appreciate some help.

INPUT Data in the Message:

11|CONT|1|16|511|PNR|OAS|I2|OAS|20020617|2004-08-10|P2|75|IL|20020613|004.25||02|01|CONT_COM|1|10|51136|PNR|BSC|001|OAS|SP3Q3|2004-08-10|84|20020617|11|

Message queue format is as follows:
CONTRACT(table neame)| Contract Row Count| Contract Col Count| Contract Data|

Actual Message
11|CONT|1|16|511|PNR|OAS|I2|OAS|20020617|2004-08-10|P2|75|IL|20020613|004.25||02|01|CONT_COM|1|10|51136|PNR|BSC|001|OAS|SP3Q3|2004-08-10|84|20020617|11|

In Designer

MQ Series -------> Row Splitter --------> TRANSFORMER ---------> Sequential Flat File

PROBLEM:

Run and compile is successful. But when I go to target Seq File and click on "view Data" get the following error:

MQ_TEST_2a.SEQ_FILE.DSLINK24: read_delimited() - row 1, too many column in record.

Then in the Output fields the the delimiter "|" is read as a char rather than a delimiter that seperates the data in the message.

Would like to know why the delimiter "|" is not being recognized. Cant find in the MQ Websphere stage parameters where to define the delimiter "|". In the target Sequential file stage have specified the delimiter.

Output looks like this:

TABLECOUT | Table Name| ROW COPUNT| CNT ID | CNT_JNH_IUR|

6 ABC 2 12345678 ERHJ


CNT_EFF_DT| CNT_CD |

|2003-5- 26|345000

Thanks,
MJ

[ray.wurlod]

The message queue format specifies four pipe-delimited fields. The data contain more than four pipe-delimited fields.

Message queue format is just like a table definition. It must match the data, otherwise errors or warnings will be generated. Set up a message queue definition that describes ALL the fields in the data, and all should be well.

[DSmamta]

Hello Ray:

Thank you very much for your advice.

I have one more question.

Who should create the message queue definition...? Is it the perosn who created the message queues or me who is only extracting the data from the messages.

Would appreciate if you could reply at your earliest convenience.

Once again thank you very very much.

Regards,
MJ

[ray.wurlod]

You create it. If the other person already has one, certainly use their work, but ultimately you need the collection of column definitions in your job design.

Best practice is to create it in the Repository first, then load it into your job design.

It must have exactly the same number of columns as you expect to process.