Power Platform Community Forum Thread Details

Hi all, I'm having some trouble with a flow and hoping to get some input.

What I want to achieve is fairly simple. When a certain email arrives, it saves the whole email into a specific SharePoint folder. But what I want to do, is with the naming of the file, I want to extract some information from the body of the email and save this in the file name. It just doesn't seem to be working. So far my flow looks like this:

1. When a new email arrives - recipient in there as well as a filter by the word 'Confirmation' in the subject line

2. Condition action - filter the work Confirmation in the subject - if yes, goes to step 3

3. This is where I run into trouble because I tried just saving the file, but in order to name the file correctly I have to extract the data I need from the email body. I've followed some instructions and used a HTML to text action which does what it says, but when I do a compose action after this to get the information it's failing.

4. So this step is 'Compose' to extract the first peice of information which is the booking reference number. The second compose is the 'Title' thats in the body.

I'm getting the following error:

Unable to process template language expressions in action 'Compose_-_reference' inputs at line '0' and column '0': 'The template language function 'indexOf' expects two parameters: the string as the first parameter, the value to find the index of as the second parameter. The function was invoked with '3' parameter(s). Please see https://aka.ms/logicexpressions#indexof for usage details.'.

Below is the outputs from the HTML to text action: (i've replaced some confidential information with ***'s

{
    "statusCode": 200,
    "headers": {
        "Cache-Control": "no-store, no-cache",
        "Pragma": "no-cache",
        "Transfer-Encoding": "chunked",
        "Vary": "Accept-Encoding",
        "Set-Cookie": "ARRAffinity=c85838fcf7ee5278cd880b7886ac6709900e8967f551e575ce4342d3d993c719;Path=/;HttpOnly;Secure;Domain=conversionservice-eus.******-eus-002.p.azurewebsites.net,ARRAffinitySameSite=c85838fcf7ee5278cd880b7886ac6709900e8967f551e575ce4342d3d993c719;Path=/;HttpOnly;SameSite=None;Secure;Domain=conversionservice-eus.******-eus-002.p.azurewebsites.net",
        "Strict-Transport-Security": "max-age=31536000; includeSubDomains",
        "x-ms-request-id": "49621741-bc4c-46a5-a682-6764917e759d",
        "X-Content-Type-Options": "nosniff",
        "X-Frame-Options": "DENY",
        "x-ms-environment-id": "default-568a5434-7d3f-4714-b824-fe722e2748c0",
        "x-ms-tenant-id": "568a5434-7d3f-4714-b824-fe722e2748c0",
        "x-ms-dlp-re": "-|-|-",
        "x-ms-dlp-gu": "-|-",
        "Timing-Allow-Origin": "*",
        "x-ms-apihub-cached-response": "false",
        "x-ms-apihub-obo": "false",
        "Date": "Wed, 27 Nov 2024 15:57:11 GMT",
        "Content-Type": "text/html; charset=utf-8",
        "Content-Length": "2336",
        "Expires": "-1"
    },
    "body": "********* Meeting Room Booking System (Do Not Reply)\n\nBooking Details\n\n \n\nBooking Reference:\n\n1137009 [https://mqb.*******-app.com.**/dwsapps-v2/ghw?meetingKey=1137009]\n\nTitle:\n\n********* | *** meeting\n\nStatus:\n\nBooked\n\nRoom:\n\n***\n\nDate:\n\n15/01/2025\n\nTime:\n\n11:00-12:00\n\nBooked for:\n\n******, ******\n\nBooked by:\n\n******, Stephanie\n\nType:\n\nExternal-Room Only\n\nAttendees:\n\n5\n\nLayout:\n\n*Standard*\n\nState:\n\nConfirmed-External\n\nCost Code:\n\n****************\n\nSite:\n\nLONDON - ** ********** Street\n\n \n\nService Item\n\nTime\n\nQuantity\n\nCost\n\nNotes\n\nApproval Status\n\nWater Complimentary\n\n11:00\n\n5\n\n0.00\n\nPre-approved\n\n 
}

Basically, I need to extract the booking reference number 1137009 and the meeting title (in stars). If I can extract them as separate bits of data, I can then use them as part of the title for my saved file. Whats the best way to tackle this? I've used Chat GPT and I'm just going around in circles. Thanks!

Categories:

General topics

Your post intrigued me because I need to do something similar to get information from FedEx shipment notifications. So, I tried to get the data out of the body of your HTML to text action since that is what I will need to do. I cannot guarantee this will work, but I managed to get the value you wanted:

This is how I went about it, which assumes the e-mails are in a consistent format since the body of the message says do not reply:

Return just the body value from what you show above. For my test, I just copied and pasted your text into a compose action.
The first thing we then need to do once we have the body of the e-mail is to replace the double returns (\n\n) with a single return (\n) and get rid of all the asterisks and trim any leading and trailing spaces. We can do that with this expression: trim(replace(replace(outputs('Compose'),'\n\n','\n'),'*',''))
Next, we want to split the output on the returns, which can be done with this expression: split(outputs('Compose_Replace_Double_Returns'),'\n')
This will give us array of items with one item for each sentence in the e-mail.
Next, I filter the array to only return the row with the booking reference number. I used a filter where the item (row) contains "https://mqb.-app.com".
Now that I have the row with the data I want, it is rather easy if this is always formatted as shown above in the screen shot (the booking number, followed by a space and then the URL). I can use this expression to extract the value: first(split(first(body('Filter_array')),' '))
1. Working inward to outward on the expression:
  1. first(body('Filter_array') gets the first item from the filter.
  2. split creates a new array on the space between the booking number and the URL.
  3. first then gets the first value from the split, which will be the booking reference number.