Dataverse List Rows "row count" advanced option question

[b_tomp]

How does it select the records to count when I set a number in row count? For instance, if I set row count = 50, does it take the last 50 records created, or just 50 random records? I am not sure - I assumed it was the last 50 but I am getting some behavior that doesn't match up w/ that so I wanted to double check

[brricard · Accepted solution]

Hello,

Created a simple table with a primary key "Number" and an additional field called "Name".

The result of listing the first 5 records is (checking the JSON output of the "List rows" of dataverse):

Basically, seems ordered by PrimaryKey or Date (modified or created); However...

If I change the record Number=1 to Number=7 (edit the entry):

When I run the flow, I still get the same entries; I can conclude that is not ordered by Primary Key; Also, from other test seems not sorted by Modified Date; 

It seems ordered by createdOn;

---

Even for a table with +500 entries created automatically with a flow in seconds, this field seems to be one that sorts the entries for the "row count".
---------------

Nevertheless, my recommendation would be to always use the "Filter rows" or "Sort By" options in the "List Rows" to avoid unexpected outputs.

Thank you,

BR

[b_tomp]

Thank you, that was very helpful. The issue I'm having is that I am looping through an entire table checking for duplicate records, and I need to loop through the entire table the first time the flow runs. That seems okay, and then after that I need to loop through the table just enough to catch the new duplicates for the day or hour. That's why I was thinking once I run it through the whole table, I can set the row count to 1000 or 2000, just to make sure I got all the recent ones. Looping to the whole table each time it takes forever, so my thought was to speed it up I would set a row count value to get the most recently created records, which it seems like it will do. What do you think? @brricard 

[brricard]

Hi,

To detect duplicates you can set up the duplicate detection job:

https://docs.microsoft.com/en-us/power-platform/admin/run-bulk-system-jobs-detect-duplicate-records

You can select an environment/entity/set the column values you want to compare to check if its a duplicate (by e-mail, user id, name,.. or even all of them. Very useful.

I believe the best way to achieve this (if you want to use automate) would be to define an action plan and apply it:

1) Create a flow just to detect all duplicates and apply an action; This flow would run just once.

2) Define what columns would be used to define if an entry is duplicated;

3) Create a new flow that runs when a entry is created or modified and applies some action if it detects a duplicated entry.

2) If you define that a duplicate is 2 entries in a table that have the column "Number" and "Name" that are equal then;

3) 

List only Rows where Number and Name = the one that was modified:

crcc0_number eq '@{triggerOutputs()?['body']?['crcc0_number']}' and crcc0_name eq '@{triggerOutputs()?['body']?['crcc0_name']}'

And then you would apply some action to the duplicates.

Thank you.

BR

[b_tomp]

@brricard thank you that is very helpful. For the initial flow that runs once through the entire table...is there a way to set it to do this. Default is 5000 records. I am not sure how many records are in the table, should I just set the value as a high as the system will accept and hope it gets them all? 

[brricard]

Hello,

That is documented here:

https://docs.microsoft.com/en-us/power-automate/dataverse/list-rows

And if you want a full post on how to use the Skip Token with more details you can follow the above link or you can check a full solution in this post:

https://community.dynamics.com/365/b/linn-s-power-platform-notebook/posts/retrieve-more-than-100-000...

Note: I have not tested this solution.

Thank you,

BR