"Get file content" changes the file encoding. How to fix?

[iroprish]

"Get file content" and "Get file content with path" actions (SharePoint based) became to change the encoding of the file. It wasn't like this 2 weeks ago. Looks like a bug to me.
Original file: 

Flow: 

Result:

 😞

[iroprish]

I tried to use .xlsx output type instead of .csv. Also didn't work.

[v-bofeng-msft]

Hi @iroprish :

This Post may be similar to the situation you encountered, you may need to Use UTF-8 mode to re-import the file:

1. Open Microsoft Excel
2. Click on the Data menu bar option.

3. Select the file

4.Choose UTF-8 Mode

The Result:

I think these links will help you a lot:

Create a .csv file that uses UTF-8 character encoding 
How to import a .csv file that uses UTF-8 character encoding 

If you want this problem to be completely resolved, I suggest you consider posting your needs in this forum:

https://powerusers.microsoft.com/t5/Power-Automate-Ideas/idb-p/MPAIdeas 

Best Regards,

Bof

[ingmar]

I'm having the same issue when getting .txt files from an FTP and save it to Sharepoint
the file on the FTP is encoded UTF-8 BOM when Power Automate get's the file content it shows the file content Binary

so now characters like é or ü or ï are now �

[Seabass]

Same issue here!

[Anonymous]

It doesn't solve the problema in an automated way. I have a flow that read a csv file generated by a system, I cannot manually fix  the file every time.

Is there another way to fix this?

[murbanik]

Anyone?

[rommelrueco]

You have to update the value of the File Content field in the "Upload File" task.

From: File Content

To: concat(uriComponentToString('%EF%BB%BF'),body('Get_file_content'))

Note: 

• You can type the updated value by clicking the "Expression" tab.
• You can get the value of body('Get_file_content') by clicking the "Dynamic Content" tab and selecting the "File Content"
  

Explanation:

.csv file is a kind of text file. If you encode it in UTF-8 format, there are two ways to save it.

- Text only

- Byte Order Mark(BOM)+Text

BOM is 3 characters (EF BB BF) to mark the file is encoded as UTF-8.

When you generate .csv file in Power Automate, it does not include BOM.

So, when you open the .csv file in excel, special characters (i.e. Korean) will be broken.

The solution is to add 3 BOM characters in front of the .csv file content before saving it.

Let me know if this works. It is working fine using the "Create File" task. I haven't tested using the "Upload File" task.

Rommel Rueco

[Linn9000]

I had the same issue in a somewhat different case (copying a CSV file from SFTP to Sharepoint).  CSV file wasn't the same after copying.

To fix the CSV issues, i set "Infer Content Type" to "No".  After this, my CSV encoding stayed the same