DSXchange

I defined one of the output columns as key and mapped that to the output along with the rest of the pivot columns. The output is coming as :-

col1 col2 col3 col4 col 5
1 1 1 1 1
2 2 2 2 2
3 3 3 3 3
4 4 4 4 4

Sainath.Srinivasan wrote:Read the source and write an output with record delimiter as comma

Thanks for the reply. tried that it splits the records into 4 rows as :-

1,
2,
3,
4,

The output should be in single row.

That would be a "vertical pivot", you can search the forums for the tecniques involved. Only thing "extra" you'd need to do is take the separate columns the pivout would generate and concatenate them ... I was able to configure the pivot stage but the output is not correct, it s...

That would be a "vertical pivot", you can search the forums for the tecniques involved. Only thing "extra" you'd need to do is take the separate columns the pivout would generate and concatenate them ... Thanks for the reply. I tried the pivot stage with the vertical pivot prope...

Hi All, I have a requirement which is something like this. The input column will have values like :- 1 2 3 4 that is input column values are in single column which needs to be output as comma separated values in single column and single row like :- 1,2,3,4 Please help me in achieving this logic. I h...

I added the $OSH_PRINT_SCHEMAS env variable to the job but there was no difference for the key column in the schema definition in the OSH script generated by the job.

Actually I am running a upsert query on the table using schema file, hence I need to define the key in the schema file.

Any replies or suggestions.

In the parameter's default value put $PROJDEF.

{Key} ... Thanks for the reply. Can you please give an example, the way we were trying it was we picked the parallel layout of the table definition, defined one of the columns in the table as key but there was no change in the layout of the table definition and we searched the documentation on the ...

Hi,

How to define a key column in schema file.

Thanks in advance.

ray.wurlod wrote:What is the data type of your source and of your target? If they are Date, then format is irrelevant. Dates are stored as binary quantities. ...

My source column is date and my target column is also date. The target database stores date in YY/MM/DD format only.

Kryt0n wrote:F1 (Help files) -> Search (Date formats)

The option that I have tried is first converted date to string and got the output as YYYY/MM/DD and then tried converting string to date giving format as %YY/%MM/%DD but it is not working.

Hi All,

We have date as YYYY-MM-DD and we that to be converted into YY/MM/DD. Please let me know the procedure.

Thanks in advance.

ray.wurlod wrote:Hence "as far as I am aware". I've learned something today.

To answer the original question, ODBC would seem to be your pathway.

Seems a bit odd to max out at 96TB though - or is this just the ...

Will ODBC driver be able to achieve the parallelism in loading.