Flow: Find / Isolate Value within a String

[CodyS]

Hello All! 

I'm trying to leverage our use of ASANA as a project management tool with our clients through FLOW to send out an e-mail when projects are completed asking for the client to complete a survey / ranking.  Problem is - the information I need to send to an individual is located in a single attribute ("note").  In order to set up the pretty simple flow, I need to isolate the 'email' value provided in this document.  I can't wrap my brain around which direction to lean...

Currently the string looks like this:

----

Project Name:

My really cool project

Name:

First: John

Last: Smith

Organization Name:

Super Cool Awesome Company

Email:

johnsmith@test.com

Phone Number:

999-999-9999

Completed electronic file is due to you by::

2019-11-15

----

Basically I just need to capture the value of e-mail from this whole string using an expression.  Possible? Any advice?

[ScottShearer]

@CodyS I believe that you could do the following:

1. Use the indexOf expression to find the location of the string Email:
2. Use the indexOf expression to find the location of the string PhoneNumber:
3. Use the substringOf expression to grab all text between the two
4. Use the trim expression on the result

You may/may not need to use the HTML to text action on your content before your start.

Let me know if you need an example.

[CodyS]

Thanks @ScottShearer  - that explanation totally makes sense. Thanks for listing it out.  As it's my first time building an advanced expression in FLOW, I'm having a hard time conceptualizing how I would nest a lot of these expressions - would you suggest working from the bottom up and layering the pieces in?

I tried putting them together that way and wound up with:

trim(substring(triggerBody()?['notes'],indexOf(triggerBody()?['notes'], 'Email:'),indexOf(triggerBody()?['notes'], 'Phone Number:')))

but it didn't return just the e-mail - it returned a longer string, akin to:

"Email: jsmith@uvm.edu Phone Number: 9999999999 Completed electronic file is due to you by:: 2019-11-04 "

Close...but my brain just isn't quite there.  Any advice?

[ScottShearer]

@CodyS 

To simply things, you might try using variables or compose actions for each piece and doing it one step at a time.  When you have that working, then put it all together into a single expression and remove the variables.  So, create a variable for indexOf(Start) and another for indexOf(End), etc..  Then, reference those two variables in the substring action.  It is easier to get it working that way when you are first starting with expressions.  One thing to note, the indexOf will give you the position where a string starts - you'll need to do some math and add to that index so that you end up at the end of the string (if that makes sense...).

If you are having trouble, post what you have and we'll go from there.

[ScottShearer]

@CodyS 

UPDATE:  In my screen shot below in the Find Email: String compose action, remove the : after Email in the expression.

Also, please note that if the text I am searching for appears elsewhere, then the results will not be what is expected.

Please see my example below.  All the expressions that I used can be seen in the action comments.  I started by putting your sample text in a compose action.  After doing this, I think it would be a really bad idea to have this all in one expression - leave it broken out per my example.  Post here if you need more help.

[v-bacao-msft · Accepted solution]

Hi @CodyS ,

Please try the following method, using the fixed field name as the split point to extract the email.

Use the split() function to split the email body with "Email:" and "Phone Number:" and extract the email.

Expression reference:

trim(split(split(body('Html_to_text'),'Email:')[1],'Phone Number:')[0])

Image reference:

Hope it helps.

Best Regards, 

[ScottShearer]

@CodyS 

The solution provided by @v-bacao-msft is obviously a much better solution.  I keep forgetting to use split() …..

[CodyS]

@v-bacao-msft  - That worked perfectly with the slight alteration so that it can worth with the ASANA connection:

trim(split(split(triggerBody()?['notes'],'Email:')[1],'Phone Number:')[0])

Much thanks to both you and @ScottShearer for your help.  EXTREMELY helpful - both in finding the answer and in understanding how these expressions work. 🙂