Unanswered

Find phrase in a text file, then perform an action if that text is found

(0) Share

Report

Posted on by CDowlin

Hello all,

Very new to Power Automate, and am attempting to create a flow that will work with a daily text document release.

I can get that file and save it to sharepoint, but I wish to scan the file afterwards for a specific line of text, and then do an action for every line of text after that line until the automation sees a line break.

Example: I want to look for 'Material lists", and if it finds that phrase, it will create a To Do task for each subsequent line of text ("Bill of material 1", "Auxiliary Material 4", etc) until a line break is found in order.

How would I go about doing this?

Thanks!

Categories:

Building flows

I have the same question (0)

All responses (9)

Answers (0)

Rhiassuring 8,692 Moderator on at

Like (0)

Report
Copy link

Link copied!

What format are the documents? Word? Or Excel sheets? Or .txt?

Depending - if you can extract the text from the file - you can do this with Find Text / Substring and a variable set with a New Line. Let me know about the format, maybe post an anonymized example of the output, and we'll figure it out.

Was this reply helpful? Yes No
CDowlin 8 on at

Like (0)

Report
Copy link

Link copied!

The export from the email is currently a .txt document. Each daily release follows a certain structure, and this would be an example:

Material Lists
Project1-BOM.xls
Project2-ANCILLARY_HW.xls

Revisions
Project3-Piece1REV02.dxf
Project3-Piece6REV01.dxf
Project4-Piece14REV01.dxf
Project1-Piece11REV03.dxf
Project3-Piece11REV05.dxf

Remedials
Project3-remedial001-issue.pdf
project3-remedial002-issue.pdf
project5-remedial011-issue.pdf

Each group of items has a line of 'header' text, followed by the names of the files generated on that day, and each group is separated by a line break. Not every section appears on every release, and the number of items in each group can vary.

I'm using .txt with sharepoint currently because it seems to segregate everything into lines as they appear in the attachment, but can convert to another document type for greater utility if necessary.

Thanks for the help!

Was this reply helpful? Yes No
AndreasN 318 Super User 2024 Season 1 on at

Like (0)

Report
Copy link

Link copied!

I tried recreating the flow with a split function, but I am having issues with the formatting of the TXT.

The delimiter to split each string, would be ENTER, so I've initialized a variable and simply just pressed ENTER and then use that string variable to define the delimiter in the split compose.

Doing so gives me a weird result, but you might be able to do a successful apply to each on this.

Honestly, I would suggest saving the TXT as a PDF, use the already trained OCR AI in the AI builder which extracts text in PDF documents. You can simply customize the template they have already created, this will create an apply to each for all strings in the document, and then you can create a SWITCH or a CONDITION to apply the logic you need for each string.

Was this reply helpful? Yes No
CDowlin 8 on at

Like (0)

Report
Copy link

Link copied!

Would there be a way to export it in a similar fashion to an excel spreadsheet? I may be able to utilize the data for more than just this particular automation , if so.

Was this reply helpful? Yes No
Matthy79 4,188 Super User 2024 Season 1 on at

Like (2)

Report
Copy link

Link copied!

Hello @CDowlin

Try this:

skip(split(split(split(YOURFILECONTENT, 'Material Lists')?[1], decodeUriComponent('%0D%0A%0D%0A'))?[0], decodeUriComponent('%0D%0A')), 1)

Instead of "YOURFILECONTENT" you have to put in the Text you want to search.

Was this reply helpful? Yes No
CDowlin 8 on at

Like (0)

Report
Copy link

Link copied!

Can you break down what this line of code is doing? apologies, am still very unfamiliar.

Was this reply helpful? Yes No
Matthy79 4,188 Super User 2024 Season 1 on at

Like (0)

Report
Copy link

Link copied!

split separates a string by using another string. I used decodeUriComponent to split by crlf (‘%0D%0A’ is the Uri decoded string for that). The rest is just picking the right values of the array created by split.

You have to read it from the inside out.

1.) Split by ’Material Lists’ and take the string after (If ‘Material Lists’ would be inside the whole string more often, then only the string from the first to the second would be recognized).

2.) Split the string from first step by 2 crlf and take everything before.

3.) Split the string from 2. Step by crlf and take everything except the first line which would be empty (You could change that if you use ‘Material Lists’ + crlf to split in step 1).

Was this reply helpful? Yes No
CDowlin 8 on at

Like (0)

Report
Copy link

Link copied!

Apologies for the radio silence... had a thought. would it be easier to load these into an excel spreadsheet , and then have them sorted to make the tasks?

and i also apologize for having a hell of a time trying to decipher the code here... i'm not making the connection in my head.

Was this reply helpful? Yes No
Matthy79 4,188 Super User 2024 Season 1 on at

Like (1)

Report
Copy link

Link copied!

Hello @CDowlin

This wouldn’t be easier. I also would try to avoid excel whenever possible.

Have you tried the formula. Doesn’t it work? Did you receive an error. Send a screenshot of where you have the problem.

Was this reply helpful? Yes No