Data Enrichment in Data Preparation in SAP Data Intelligence.

[amish1980_95]

Hello Folks,

I am trying to use inner join feature of data enrichment while joining a csv and a table or two tables based on a Primary key. Everytime I am getting zero rows as a result , however If I perform the same join on same tables in SAP HANA DB, it gives me the correct result.

Can somebody put some light on it, why inner join in data preparation is not giving correct result?

AM I missing soemthing?

Thanks,

Indu Khurana.

[jimgiffin]

One thing to clarify - when on the 'preview' mode of the data preparation, it only pulls the first 'n' records from each object (in your scenario the CSV and HANA table) - if these two data preview sets don't have matching data for the join, the resulting UI view of the preview will not contain data. You have to execute the full data preparation to see the full results.

Can you run a data preview or a data profile in the metadata browser for the target object of the data preparation? If there is still no data, it could be the data types/lengths inferred by DI from the csv object as mentioned by Simon Ye previously. When working with files, SAP DI does try to give metadata, but inherently files have no true metadata so when joining with tables it can lead to confusion technically.

As another test, can you temporarily load the csv data to a HANA table and attempt the join again?

[Simon_Ye]

try left/right outer join first, if you can get the data, data from the .csv file may have a different data type with data from HANA DB.

[amish1980_95]

Hello Simon,

The data which is present in the sample set (default 1000) is being considered for the joins, is it expected behaviour?

I created two smaller CSVs/tables, it is giving me some output but that's not the case with larger datasets.

Are the joins performed only on a subset/sample of the data or the entire dataset?

Thanks,

Indu

[Simon_Ye]

I don't think the join can only be performed on a subset.

I'm not very clear about your scenario, could you paste a screenshot here?

[amish1980_95]

Hello Simon Thank you for your response.

So, my requirement is to perform inner join on a csv uploaded in s3 bucket and a table in SAP HANA.

The csv has 90k records and Hana table has 1348437 records.

I am trying to run the inner join in Data enrichment in data preparation in ME in SAP DI on these two data sources based on a P-key, but it is generating 0 records. However, when I run the inner join query in HANA DB explorer, I get 20K records as a result.

To troubleshoot further, I created two smaller tables with 6 records each and ran the inner join, it is working as expected.

Is it like the joins in data enrichment work on the smaller data set, do we need to make changes in default configurations to handle this?

Thanks,

Indu Khurana.