Replace and concatenate domain name

[Anonymous]

Is there a expression to remove the @ sign at the front and then replace the space between two names to a '.' and at the end of the name add @ domain name. Also to trim out anything after the last name.

eg- 

@ Testing Person this is a test comment -> Testing.Person@gmail.com

[eliotcole]

Assuming that your email is in a variable called emailVAR, then:

trim(replace(first(split(variables('emailVAR'), '@')), '.', ' '))

If you wanted to create that whole line, then:

concat('@ ', trim(replace(first(split(variables('emailVAR'), '@')), '.', ' ')), ' this is a test comment -> ', variables('emailVAR'))

If all the emails were definitely split by given and family name respectively, then you could define a givenVAR string variable with:

first(split(first(split(variables('emailVAR'), '@')), '.'))

And you could define family with

last(split(first(split(variables('emailVAR'), '@')), '.'))

[Anonymous]

I think there is a misunderstanding, what I am trying to achieve is that, assuming that you only know persons first name and last name, create an email address from it. Hard coding the domain is fine.

This is the output of the HTML to Text 

@ Testing Person this is a test comment 

I want to convert the above to 

testing.person@gmail.com  

[eliotcole]

Ah, my mistake, that's easier, then.

Solution

If we can assume:

1. The input text is always validatedly set in the the above forma
2. The following variables exist
   1. domainVAR - string - has the domain
   2. inputStringVAR - string - has the input
   3. emailVAR - string - is empty

... then we can use the following to set the emailVAR variable:

concat(first(split(substring(variables('inputStringVAR'), 2), ' ')), '.', first(skip(split(substring(variables('inputStringVAR'), 2), ' '), 1)), '@', variables('domainVAR'))

More Options

If you want it in lower case, then use:

concat(first(split(substring(toLower(variables('inputStringVAR')), 2), ' ')), '.', first(skip(split(substring(toLower(variables('inputStringVAR')), 2), ' '), 1)), '@', variables('domainVAR'))

If you also have empty givenVAR and familyVAR string variables set for the names (using the parts above):

concat(toLower(variables('givenVAR')), '.', toLower(variables('familyVAR')), '@', variables('domainVAR'))

You may wish to do this to make the logic more apparent to future users, and/or to re-use the names elsewhere in the flow.

[Anonymous]

@eliotcole

unfortunately this is the output

testing.person this@gmail.com

input 

@Testing Person this is just a test

[Anonymous]

You can make this work by splitting up your input string and putting it into an array.

I did this using an array called 'words', created with the expression:

split(substring(variables('inputStringVAR'), 2), ' ')

Once you've done that you can isolate firstname 'Testing':

first(variables('words'))

and the lastname 'Person':

first(skip(variables('words'), 1))

if you had the domain in a string variable, you can now concatinate:

concat( first(variables('words')), '.', first(skip(variables('words'), 1)), '@', variables('domainVAR'))

And the output:

Hope this helps.

Please flag as solution if this works for you. It'll help others find this.

[Anonymous]

@Anonymous  This is the output I get

[

  "esting",

  "Person this",

  "is",

  "a",

  "test",

  "comment\n\n"

]

esting.Person this@gmail.com

[eliotcole]

@Anonymous, huh, that's odd. Let's try something.

These are the exact actions that I just took from my text above.

1. Create a new flow.
2. Define a string variable: inputStringVAR
3. Enter the exact text you gave me earlier with no space at the front in the inputStringVAR value:
   @ Testing Person this is a test comment
4. Define a string variable: domainVAR
5. Enter this in the domainVAR value:
   gmail.com
6. Add a Compose action
7. Create an expression in the Compose action and paste in the below code
8. Save then run this new flow.

concat(first(split(substring(toLower(variables('inputStringVAR')), 2), ' ')), '.', first(skip(split(substring(toLower(variables('inputStringVAR')), 2), ' '), 1)), '@', variables('domainVAR'))

Just report back with what happens if you can? 🙂

---

FYI 1 - Avoiding Leading Spaces

If we want to eliminate the chance that the inputStringVAR has erroneous spaces before and after we can simply add a couple of trim() functions, like so:

concat(first(split(substring(toLower(trim(variables('inputStringVAR'))), 2), ' ')), '.', first(skip(split(substring(toLower(trim(variables('inputStringVAR'))), 2), ' '), 1)), '@', variables('domainVAR'))

FYI 2 - Avoiding Substring

Substring is fine, but finickity, it means that lengths need to be exact each time, and often you'll end up doing more work in the long run. That doesn't mean it won't work, but it'll have more upkeep.

FYI 3 - Shortening Using take() and skip()

There is a shorter way of doing it using take() and skip() functions in the expression. The take() function will return the first few items of an array, and the skip() function will jump over them, you define how many this is after the comma:

:

concat(join(take(skip(split(toLower(trim(variables('inputStringVAR'))), ' '), 1), 2), '.'), '@', variables('domainVAR'))

[Anonymous]

@eliotcole I converted the input to an array to see what the output look like in the following, as you can see the second item in the array has " this" where it should be "Person"

[

  "Testing",

  "Person this",

  "is",

  "a",

  "test",

  "comment\n\n"

]

[eliotcole]

Have you followed the steps that I just posted above?

Because when I paste that text, as written, on my instructions above, with the code, above, in a new flow, I get different results.

Could you show me a screenshot of your new flow, please? 🙂

[Anonymous]

@eliotcole  

[eliotcole]

OK, thanks, this is what your main issue is (it would seem) it's that the text that you're putting in your inputStringVAR is not the exact text that we mentioned above, it has different characters in it.

You'd mentioned that the text was going to be consistent, so we worked with that, however what we're seeing is not the text mentioned.

If you remove that 'The plain text c...' and entered what I mention above, you'll see that it works.

Could you please send me the data from that 'System_History' field that you're sending into the HTML to text action?

[Anonymous]

When I replace the Html to text with "@ Testing Person this is a test comment" the output shows the correct email.

<p><div><a href="#" data-vss-mention="version:2.0,c2fc253f-6644-6f52-9366-318d9b406142">@Testing Person</a>&nbsp;this is a test</div></p>

output when converted 

@testing Person this is  a test

[eliotcole · Accepted solution]

OK, there you can see the href link closure and the &nbsp; (which isn't required to be honest) which are likely causing you issues, here.

I would raise that as an issue, separately, to the Power Automate team elsewhere in this forum.

Link back to this post to help them get context on why it's not presenting the right data. 🙂

I'm seeing what you're seeing based on that HTML to plain text thing, and it's clearly not working right.

Here's the HTTP codes that it's got there instead of spaces:

testing.person%C2%A0this%40gmail.com%20space

Workaround & Fix

I have done a quick check and I think this should fix it for you, in your HTML to text action delete the System_History item that is there, and replace it with an expression, and use this:

replace(body('Parse_JSON')?['System_History'], ' ', ' ')

What that does is still use the same field, but if there is any usage of the &nbsp; HTML that is used to make spaces, then it will simply replace those characters with an actual space.

Also, I've somewhat optimised the emailVAR expression. You should be fine with what you have, but you can try this, too (maybe in a separate place?):

concat(join(take(split(substring(toLower(variables('inputStringVAR')), add(indexOf(toLower(variables('inputStringVAR')), '@'), 1)), ' '), 2), '.'), '@', variables('domainVAR'))

[Anonymous]

@eliotcole When the Html to text is replaced with the given expression I get this error 

Unable to process template language expressions in action 'Initialize_variable_4' inputs at line '1' and column '20025': 'The template language expression 'body('Html_to_text')?['body']' cannot be evaluated because property 'body' cannot be selected. Property selection is not supported on values of type 'String'.

[eliotcole]

I'd guess that looks like your Html to text didn't receive any data on that run, mate.

[Anonymous]

just realized it was the missing ['fields']? that was causing it. 

replace(body('Parse_JSON')?['fields']?['System_History'], '&nbsp;', ' ')

In your new concat expression this is the new output any way to trim out the </a> tag

testing.person</a>@gmail.com