Data in Preview is OK, but not in the real result

[pyguillou]

Hi folks

I was not able to find a similar issue recently, so I am wondering if this is new or not.

The context is as follows: 

• we use native ODBC connection to a SAP HANA system. the source is a calculation view of type cube with a star join
• We retrieve 4 fields, and filter on 2 specific items
• In the preview of the data, everything is fine: each of the 4 columns retrieve what the source gives/sends
• but, when putting the result into a visual, or looking at the data in the table view, then the data is different
  • The data we retrieve is the article number, the alternative external article number, the warehouse, and then a conactenation of the alternative number and the warehouse. 
  • It happens that the alternative numbers are very similar, the only difference is that we have 2 letters in lower case, whereas they are in upper case for the other line...

An image is worth thousands words: 

in the Preview: 

exactly what we expect

in the Table view after forcing the data refresh: 

The source of the data is of type NVARCHAR, that PBI interprets as Text. So far so good...

Anyone would have any idea ?

Thanks

PY

[v-sathmakuri]

Hi @pyguillou ,

I hope this information provided is helpful. Feel free to reach out if you have any further questions or would like to discuss this in more detail. If responses provided answers your question, please accept it as a solution so other community members with similar problems can find a solution faster.

Thank you!!

[v-sathmakuri]

hI @pyguillou ,

I wanted to check if you had the opportunity to review the information provided. Please feel free to contact us if you have any further questions. If the responses has addressed your query, please accept it as a solution and give a 'Kudos' so other members can easily find it.

Thank you!!

[pyguillou]

Hello,

This topic fell under the pile

Will update you as soon as I can (may be beginning of next week) 

Sorry for the delay

[v-sathmakuri]

Hi @pyguillou ,

Thank you for the response.We will be waiting for your reply.

[v-sathmakuri]

Hi @pyguillou ,

May I ask if you have resolved this issue? If so, please mark the helpful reply and accept it as the solution. This will be helpful for other community members who have similar problems to solve it faster.

Thank you!!

[pyguillou]

@v-sathmakuri , thanks for the update.

So, there is no way to make the PowerBI models case sensitive ? We have to recerate a column where we decode in UNICODE, and then how to get the lower case character ?

Thanks

[v-sathmakuri · Accepted solution]

Hi @pyguillou ,

I have combined 2 columns and then i am able to reproduce the case sensitivity. Combined Text column and a unique number column which is like identitty column. Please find below screenshot for reference, also attached pbix file for reference.

Below is the M Query :

let
Source = Excel.Workbook(File.Contents(source), null, true),
Sheet1_Sheet = Source{[Item="Sheet1",Kind="Sheet"]}[Data],
#"Changed Type" = Table.TransformColumnTypes(Sheet1_Sheet,{{"Column1", type text}}),

AddCombined = Table.AddColumn(#"Changed Type", "Combined", each [Column1] & Text.From([Column2]))
in
AddCombined

If this post helps, then please consider Accepting as solution to help the other members find it more quickly, don't forget to give a "Kudos" – I’d truly appreciate it! 

Thank you!

[v-sathmakuri]

Hi @pyguillou ,

Thank you for reaching out to Microsoft Fabric Community.

This behavior is due to Power BI's data model being case-insensitive for text fields. While the Power Query preview respects case differences, the model merges values like AIP and AiP as the same.

• Create a new column encoding each character using UNICODE() to preserve case.
• Keep the table in DirectQuery mode as SAP HANA respects case.
• If case sensitivity isn’t important, you can standardize the values by converting them to uppercase using Text.Upper

If this post helps, then please consider Accepting as solution to help the other members find it more quickly, don't forget to give a "Kudos" – I’d truly appreciate it! 

Thanks & Regards,

Rekha,

CustomerSupportTeam.