Skip to main content
cancel
Showing results for 
Search instead for 
Did you mean: 

Be one of the first to start using Fabric Databases. View on-demand sessions with database experts and the Microsoft product team to learn just how easy it is to get started. Watch now

Date format assumed earlier than expected in Power Query

When getting data from Excel source, the date format is assumed when the format of the column is set to Any (early stage of data preparation, see printscreen below) and the header of the column is "Date".

 

Something system wise happening during header promotion. When we want to replace incorrect values (in this case 00.00.0000 generated for empty date by SAP) by null or another number, it returns an error stating 00.00.0000 is not a date even so the column type is set to any.

 

Enter a Date Value messageEnter a Date Value message

 

 

 

Could the logic be fixed so when the column type is set to any, we are not obliged to add the step before the header promotion generated automatically by the system ?

 

David

Status: New
Comments
v-qiuyu-msft
Community Support

Hi @Anonymous, 

 

Please upload your screenshots to clarify the issue. Please do mask sensitive information in the screenshots before uploading them. 

 

Best Regards,
Qiuyun Yu 

Anonymous
Not applicable

Hi @v-qiuyu-msft ,

 

I have successfully uploaded the picture (strangely enough, I was not able to do so yesterday).

Hope this will help to understand the point.

 

Best Regards,

David

v-qiuyu-msft
Community Support

Hi @Anonymous, 

 

I tested on my side with Power BI desktop 2.76.5678.782 64-bit (December 2019) but not able to reproduce the issue. Please run the same Power BI desktop version as ours then test again. 

 

w1.PNG

 

Best Regards,
Qiuyun Yu

Anonymous
Not applicable

Dear @v-qiuyu-msft ,

 

Thank you for your message. I checked that my version is the latest available (2.76.5678.782 64-bit);

2020-01-16_09-38-26 (pbi desktop version).jpg

 

I noticed that your Replace value screen is different than the one I have, is there a reason? (you have the option to choose the data format on the left, which I don't have).

I am anyway getting the error using the "Replace Values" button in the ribbon and it is easily reproductible (dataset coming from SAP). Even with a datatable, it happens (and it is even worse as there is no way to avoid the "auto-conversion-to-date-format");

2020-01-16_10-00-07.jpg

 

2020-01-16_10-00-44.jpg

 

2020-01-16_10-02-18.jpg

 

2020-01-16_10-02-50.jpg

 

Thank you for your support,

David

v-qiuyu-msft
Community Support

Hi @Anonymous, 

 

I would suggest you create a support ticket to let engineers look into the issue on your side.

 

Support Ticket.gif

 

Best Regards,
Qiuyun Yu