extracting text from email with no structure

(0) Share

Report

Posted on by um09

I am try to extract three things from a email into a list, i was wondering if some could guide me how to do it ?

All want to do is take the order number from email,name of person,their problem.

Thanks

Email address

Name

Order

Problem

I have tried to use :

to grab the text BEFORE the # symbol : first(split(body('Html_to_text'),'#'))

to grab the text AFTER the # symbol : last(split(body('Html_to_text'),'#'))

but this doesn't seem to work i have even tried to convert into lines but i am struggling.

Example of an email 1

Hi,

We have received the below from a customer. Please can you confirm what actions need to be taken so we can relay this to the customer or if able please can you contact the customer directly.

Firstname Lastname
###@gmail.com
W12433311-1M-V

----------------------------------------------

!*** Firstname Lastname was not signed in when this comment was submitted. Learn more ***!

Firstname Lastname, 14 Dec 2022, 09:42

I still haven’t received my order can you pls update me

‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ Conversation ID: 42581102-8a97-11ed-9d64-7d7cb637a4b4 42581102-8a97-11ed-9d64-7d7cb637a4b4 49be6de

Example 2

Hi,

We have received the below from a customer. Please can you confirm what actions need to be taken so we can relay this to the customer or if able please can you contact the customer directly.

Firstname Lastname
#####@yahoo.co.uk
W12617455-2M-V

----------------------------------------------

!*** Firstname Lastname was not signed in when this comment was submitted. Learn more ***!

Firstname Lastname, 2 Jan 2023, 05:36

I ordered this item weeks ago and it still hasn't arrived. I would like to be refunded please.

------------------
Submitted from:

‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ Conversation ID: 48d7040b-8a7c-11ed-a8ee-96d2e547bdbc 48d7040b-8a7c-11ed-a8ee-96d2e547bdbc 49be6de

Categories:

Building flows

extract.png

I have the same question (0)

All responses (3)

Answers (3)

Verified answer

v-liwei-msft on at

Like (0)

Report
Hi @um09

Based on the first email example, I made a test for your reference:
1. Email address:

split(trim(last(split(first(split(outputs('Html_to_text')?['body'],'-----')),'directly.'))),decodeUriComponent('%0A'))[1]

2. Name:

split(trim(last(split(first(split(outputs('Html_to_text')?['body'],'-----')),'directly.'))),decodeUriComponent('%0A'))[0]

3. Order number:

split(trim(last(split(first(split(outputs('Html_to_text')?['body'],'-----')),'directly.'))),decodeUriComponent('%0A'))[2]

4. For customer questions:
Add an additional compose, enter two carriage returns in it

then use the formula

split(first(split(outputs('Html_to_text')?['body'],'Conversation ID')),outputs('Compose_5'))[add(length(split(first(split(outputs('Html_to_text')?['body'],'Conversation ID')),outputs('Compose_5'))),-2)]

Best Regards,

Levi

Was this reply helpful? Yes No
Verified answer

grantjenkins 11,063 Moderator on at

Like (0)

Report
Assuming the initial body of the email is the same (except for the data you're trying to retrieve), you can try the following.

See full flow below. I'll go into each of the actions.

When a new email arrives is what you already have.

Html to text takes in the body of the email as you have already done.

Filter array extracts the data into an array, removing any empty lines. The expressions used are:
//Body (input) split(outputs('Html_to_text')?['body'], decodeUriComponent('%0A')) //Condition trim(item())

We can then extract out the Name, Email, and Order Number using the following expressions:
//Name trim(body('Filter_array')[4]) //Email trim(body('Filter_array')[5]) //Order Number trim(body('Filter_array')[6])

In my example flow, I've created three variables to hold the values.

Or, if you just wanted an object with the values, we could create a variable of type object with the following.
{ "Name": @{trim(body('Filter_array')[4])}, "Email": @{trim(body('Filter_array')[5])}, "Order Number": @{trim(body('Filter_array')[6])} }

This would give us the following output.

----------------------------------------------------------------------
If I've answered your question, please mark the post as Solved.
If you like my response, please consider giving it a Thumbs Up.

Was this reply helpful? Yes No
Verified answer

grantjenkins 11,063 Moderator on at

Like (0)

Report
Another option is to use XPath expressions. This is assuming the email address will always contain #@

See full flow below. I'll go into each of the actions.

XML is a Compose that takes in the data from Html to text, split by new line and converts it to XML. The expression used is:
xml(json(concat('{"root": { value:', split(outputs('Html_to_text')?['body'], decodeUriComponent('%0A')), '}}')))

We can then use the following expressions to get the Name, Email, and Order Number.
//Name xpath(outputs('XML'), 'string(//root/value[contains(text(), "#@")]/preceding-sibling::value[1]/text())') //Email xpath(outputs('XML'), 'string(//root/value[contains(text(), "#@")]/text())') //Order Number xpath(outputs('XML'), 'string(//root/value[contains(text(), "#@")]/following-sibling::value[1]/text())')

In my example flow, I've created three variables to hold the values.

----------------------------------------------------------------------
If I've answered your question, please mark the post as Solved.
If you like my response, please consider giving it a Thumbs Up.

Was this reply helpful? Yes No