Help Parsing Email Body

[bryanshort12]

Sorry, I know this has been posted about ad nauseum, but I'm a stupid attorney and I'm just trying to get this done! I have created a flow such that each time I receive an email with certain strings in its Subject line, it will add an item to a sharepoint list. Problem is, I need to parse text from the email body in order to be able to generate strings for the column data for my sharepoint item. 

I've gotten this far: 

I understand that in the Compose item, I need to actually manipulate the strings parsing out the plain text to arrive at the wanted strings to then set those strings as my variables and then finally use those variables as the items to include in the Create Sharepoint Item. 

That said, I can't seem to figure out how to trim the output text to arrive at the strings I want. Here is some sample plain text (aka after the HTML to Text Conversion. 

Your Agreement Has Been Sent for Signature!

•

•Adobe Sign has sent Engagement Letter - John Doe to johndoe@gmail.com for signature. 

•


•When johndoe@gmail.com completes Engagement Letter - John Doe all parties will receive a final PDF copy by email. 

•

Click here to view this document online in your Adobe Sign account.

You can always Login to Adobe Sign to:

•Check the status of this document

•Set up a reminder informing another party that you are waiting for their signature

•Share your agreements with a colleague

So you can obviously see that I am trying to parse out the client name and the email address out of this text. 

Please help me generate the text parsing. Once I'm able to figure that out, I can set those into the appropriate variables and then use the variables to generate my Sharepoint list item.

I promise I can do the rest! Thanks and apologies if this has been answered a million times, but I tried finding useful examples in the support, but I couldn't figure out how to use prior examples in this case. 

[abm]

Hi @bryanshort12 

May I ask you the following text is always going to be static except the bold characters (Name & Email)

Adobe Sign has sent Engagement Letter - John Doe to johndoe@gmail.com for signature

Thanks

[bryanshort12]

Yes, the text is identical in each email other than the variables, obviously (being email and client name).

[bryanshort12]

@abm yes, the text of the email body is always identical except for the unique values of client name and email address. Thanks!

[abm]

Hi @bryanshort12

Thanks for your quick reply. There might be different ways to do this. This is one way of doing this.

Here it is

The expression for the above compose are as follows:

last(split(outputs('Compose_2'),'Adobe Sign has sent Engagement Letter - '))

first(split(outputs('Compose_3'),' for signature.'))

first(split(outputs('Compose'),'to'))

last(split(outputs('Compose'),'to'))

Here are my test results

Thanks

[bryanshort12]

@abm fantastic, thanks so much! I'll mark as solution

[bryanshort12]

@abm I'm still having trouble with this. Sorry. So, I've got everything working almost perfectly, but the last parsing operation yields an error. The data I'm using is clientName = Testy McTesterton email=johndoe@gmail.com

Running all the operations as you've generated yields the correct clientName (actually I needed to fix it up a bit), but in the last parse operation, this yields as a string email= 'Testy McTesterton to johndoe@gmail.com'

The issue is obviously with the last(split(outputs('Compose_3'), 'to')) operation. I've tried last(split(outputs('Compose_3'), 'to ')).

I think there's some sort of issue with the assumptions as to how "last" parses data as it seems to not be performing the operation correctly. 

Here's the output from the test: 

Any ideas?

[abm]

Hi @bryanshort12 

Looks like its a data issue. What's the third compose result you have?

Thanks

[bryanshort12]

Here is the output in Compose_3:

Maybe there's a \n in there? It keeps showing some kind of non-string character there prior to the email

[bryanshort12]

sorry @abm I Forgot to @mention you on my response. It's above.

[abm]

Hi @bryanshort12 

Yes there is a \n there. So try split using the to and \n

[bryanshort12]

@abm sorry to abuse your good graces. I've changed this a bit to make the data capture easier. Now I'm going to grab text from the email Subject. It will be easier, I think. 

Subject = 'Engagement Letter - Testy McTesterton has been sent out for signature to johndoe@icloud.com'

So, I will do

1. last(split(outputs(Compose),'Engagement Letter - '))
2. first(split(outputs(Compose_2),'has been sent out for signature to'))

So we've got our client name parsed (only thought is that I might need a wildcard at the beginning of number 2 string there, not sure how to do that). 

Then I think I need to :

last(split(outputs(Compose),'has been sent out for signature to'))

That should get me the email address, I'm hopeful. 

Thoughts?

[abm · Accepted solution]

Hi @bryanshort12 

No worries. Here it is.

Above expressions are as follows:

replace(replace(outputs('Compose_2'),'Engagement Letter - ',''),' has been sent out for signature ','')

first(split(outputs('Compose_3'),'to '))

last(split(outputs('Compose_3'),'to '))

My test result

[bryanshort12]

@abm  Sorry, this is driving me nuts. 

I've set this up and tested it 100 times and I keep getting some errors in the data. If I test the way you did, the result is perfect, but if I use my flow, the data has some sort of issue. 

This is the result of both of the last and first operations:

The only thing that I can come up with is that there is a carriage return in there. I don't know how to call it out in the expression. I've tried \n but nothing changes. Any help is greatly appreciated!

[abm]

Hi @bryanshort12 

Check your subject text contains 'to' with spaces between. I can't see it here.

[bryanshort12]

@abm I think that I've figured out what is going on. The input text has a \n in it. 

Engagement Letter - Testy McTesterton has been sent out for signature to
johndoe@icloud.com

This leads the first compose function to output: 

You can see that the text spans two lines. I've tried everything I can think of to try and trim that \n, but I can't figure out how to do it. Any suggestions? 

[abm]

Hi 

Glad that it resolved. My suggested answer was correct based on what you said originally.

Thanks

Thanks