In Microsoft Flow it is possible to use the Excel connectors to read data form a spreadsheet. there are however a few major issues with doing this:

  1.  You data needs to exist within Excel Tables.
  2. Large tables take long to load. (4000 rows takes about 2 minutes)
  3. The Excel connector likes hard coded names for sites, libraries and Excel files.
  4. Excel files get locked as you configure flows. And it can take a while for the locks to be removed.

If the above is not a problem please feel free to create a flow like this:

 

If you want more however you might want to look into the graph API. With the help of Flow Ninja, John Liu‘s post on twitter I got this all working.

I started with setting up my app permissions which gave me a Tenant ID, Client Id and Secret which I then used in 3 variables. If you need some help setting this up you could look at the infoPulse site.

Now I’m having to Collect the site details form graph. This can be done by configuring an HTTP action as shown below. After the v1.0/sites bit simply sepcifiy the url of your subsite.

 

This now gives the ID of the site that we need to get the list of the lists an libraries that we are interested in. Remember you will need to specify the same authentication bits as in the previous HTTP call.

 

Now I can use the ID returned for the library (I used a compose in between these actions to hold the library details.)

This now give me the fully generic solution that helps me copy the flow between my Dev, UAT and production environment

 

Then finally I can run through the Excel data with an Apply to each step. In this case I only did a Compose to collect each row but most likely you would now do something with the data like updating a SharePoint list.

 

  1. No need for Excel Tables.
  2. Large tables are quick to load. (4000 rows takes about 4 seconds)
  3. My flow is a generic process that is easy to copy between environments
  4. No file locking on my Excel files.
Advertisements