04-13-2021 19:12 PM
This Flow deletes Excel Rows Older than X amount of days based on a Date column in Excel. The Excel file can be in SharePoint or OneDrive. Very little configuration is needed by the user to make this work.
When importing the Flow, you might get an error message. This is because the Excel file that the Flow is being pointed to does not exist in your Tenant. Click 'Save as new flow' if you get this message.
I have a Excel file that has 2 columns,
My Name Field is my unique Key Column.
Date Hired, I will be checking and deleting all records that have a Date Hired older than 7 days.
** Your Date column MUST be in Date format in Excel **
Here is what my Excel file looks like.
The trigger can be anything you want:
Button, Schedule, Anything!
All 3 Variables need to be inputted by the user:
Days: Enter number of days you want the Flow to check against the date column to delete. For example, -7 will delete all records 8 days old
** Don't forget to put the - **
KeyColumn: Enter your Key Column from Excel. This column MUST have unique values.
In my Excel file this column name is Name Field
DateColumn: Enter the name of your Date Column from Excel
In my Excel file this column name is Date Hired
Past Date: Only change the Destination time zone field. Change to your time zone
List rows present in table: Only fill in required information marked with red *
** If your Excel file has more than 1500 records. Click the 3 dots on the List rows action and select settings. Than up the value on the Threshold and click done. **
Filter array - Records older than X: Do not touch anything in here
For each record older than X: The loop itself do not touch
Delete a row: Fill in ONLY -- Location, Document Library, File, Table --
That is it, some important things to consider:
- Make sure your Key Column has unique values
- Change the trigger to your liking
- Don't forget to add the minus symbol in the Days Variable
- Only change the values highlighted in Yellow
- Make sure to check the amount of records in your Excel file and configure Threshold values as needed
- Your Date column in Excel MUST be formatted as Date
We get todays date using utcNow() expression, than add days based on the amount of days entered into the Days variable. This date is converted to local time, and formatted like yyyy/MM/dd. This is to do easy comparison/evaluation.
The expression I use is:
addDays(utcNow(),variables('Days'))
We list ALL rows in Excel because Excel connector has limitations on the Filters we can use. Also the Excel Connector does not like spaces in the Column names. Which is why Filter array is used instead.
The Filter array grabs the Converted date from the first step and checks if the date is greater than the Excel date.
Note the Excel date comes over as a serial date in Power Automate. This is a long Integer, so to convert the serial date into a date we can read and use, we use this expression:
addDays('1899-12-30', int(item()?[variables('DateColumn')]), 'yyyy/MM/dd')
Now in the Apply to each loop, we use the values returned from the Filter array (Which is only records older than X date)
We use a Delete a row action to delete the record in Excel.
I am using the item() expression to select the Key Value from the row inside the Loop.
The syntax of this is:
item()?['<Excel Column Name>']
item()?[variables('KeyColumn')]
Done!!
If you have any questions let me know. Thanks!
Proud to be a Flownaut!
Hey @Dsym
Can you give me a sample of your excel file?
Proud to be a Flownaut!
Hi @Jcook See the below screenshot oh how the excel file is set up. Date column is in date format (short date). This is just a test at the moment hence the made up names and numbers etc.
Thank you @Dsym
Can you send me the outputs of the List Rows action in Power Automate in Run history.
Proud to be a Flownaut!
{"body":{"value":[{"@odata.etag":"","ItemInternalId":"a5c29481-a829-43ff-a433-b2d7d876cec6","Date":"44228","Name":"a","Phone Number":"","email address":"","Destination":"","International travel (14 days)":"","Close contact (48Hrs)":"","Symptoms":""},{"@odata.etag":"","ItemInternalId":"c8550d5c-f023-420e-9c53-0f9208ed3df4","Date":"44229","Name":"s","Phone Number":"","email address":"","Destination":"","International travel (14 days)":"","Close contact (48Hrs)":"","Symptoms":""},{"@odata.etag":"","ItemInternalId":"d0973eed-06e3-4402-888e-f2d2399bc2a8","Date":"44231","Name":"d","Phone Number":"","email address":"","Destination":"","International travel (14 days)":"","Close contact (48Hrs)":"","Symptoms":""},{"@odata.etag":"","ItemInternalId":"35897161-6051-4d3a-ad0c-173edfc1ba55","Date":"44232","Name":"f","Phone Number":"","email address":"","Destination":"","International travel (14 days)":"","Close contact (48Hrs)":"","Symptoms":""},{"@odata.etag":"","ItemInternalId":"e4256e2e-7d8d-44a9-8d10-e0a7a7cfce75","Date":"44230","Name":"g","Phone Number":"","email address":"","Destination":"","International travel (14 days)":"","Close contact (48Hrs)":"","Symptoms":""},{"@odata.etag":"","ItemInternalId":"fd63fd99-1849-4b61-81e1-b7d5da349401","Date":"44233","Name":"h","Phone Number":"","email address":"","Destination":"","International travel (14 days)":"","Close contact (48Hrs)":"","Symptoms":""},{"@odata.etag":"","ItemInternalId":"0d8b635f-c880-4865-8335-2beea35142ef","Date":"44234","Name":"j","Phone Number":"","email address":"","Destination":"","International travel (14 days)":"","Close contact (48Hrs)":"","Symptoms":""},{"@odata.etag":"","ItemInternalId":"ce2dd1ef-ca6d-484e-a07a-541ed18c8a0e","Date":"44235","Name":"k","Phone Number":"","email address":"","Destination":"","International travel (14 days)":"","Close contact (48Hrs)":"","Symptoms":""},{"@odata.etag":"","ItemInternalId":"8b195829-791c-493a-bbbd-e5115babe1ca","Date":"44236.5030555556","Name":"ef","Phone Number":"ef","email address":"fe","Destination":"Sheringham Shoal Offshore","International travel (14 days)":"No","Close contact (48Hrs)":"No","Symptoms":"No"}]}}
Hey @Dsym
Thank you for the outputs, that is what I was looking for,
In the outputs, the dates from excel come across like this:
OR
2) Try to replace the expression in the Filter Array action.
Instead of using addDays('1899-12-30', int(item()?[variables('DateColumn')]), 'yyyy/MM/dd')
USE
addseconds('1899-12-30',int(formatnumber(mul(float(item()?[variables('DateColumn')]),86400),'0')))
Let me know if this works,
I am working on an updated version of this Flow which would make things way easier, which I will be posting on my blog this weekend.
Proud to be a Flownaut!
Hi @Jcook Many thanks for the explanation. I tried changing the texts as suggested and although the test ran successfully, the excel sheet did not update. Reading through your explanation, I went back to the flow which pulled the data into the excel sheet from forms. The submission date carries over the time to, so I replaced this with
Hi JCook,
Firstly thanks for a great article. My team and I are based in Australia and are having a bit of trouble with this one. After a bit of tweaking (also reading all the comments) we can get the flow to run without errors. But the filter for a long time wouldn't return any rows legible for deletion. After changing the filter to 'less than' it now actually deletes, hurray!
However it is deleting everything and not excluding rows that it should be excluding. We think it might be to do with our time and date format in Australia. While we do use short date format, it's dd/MM/yyyy here and not MM/dd/yyyy. We think this might be the cause of our issue but haven't been able to nail it. Could you take a look at these screenshots and let me know what you think?
Thanks!
Hey @Anonymous ,
Than you!
Could you send me a sample of your excel file?
Proud to be a Flownaut!
Hi @Anonymous
It just dawned on me, Can you try with the format 'yyyy/MM/dd' this is how dates are compared. My Flow converts both values to this date format before doing the compare.
Proud to be a Flownaut!