Number of pages in a PDF file

[nana001]

In a Flow I am building I would like to obtain the number of pages in a PDF document. Is there anyway?

[abm]

Hi @nana001 

Unfortunately using Power Automate this won't be possible to do.

Thanks

[nana001]

Abm, thanks for your message. 

The only other potential solution that I could find is reading the PDF file in plain text as suggested here: Solved: Re: Extracting PDF meta data and document info - Power Platform Community (microsoft.com)

However, I cannot seem to decipher the "/Pages" metadata for a PDF file. I looked through the ISO reference. Any idea how to potentially read a PDF file in plain text to get the total number of pages in it?

[seewo]

Dear @abm @nana001 

         I found an imperfect solution for counting pdf pages in Power Automate: Here is the procedure for your reference:

1: extract your pdf file and set "the index numbers of the pages " to large quantity which is clearly pages out of bounds to make error-"Page out of bounds"  happen .  😀

2: Make sure continue flow run in Advanced: Page out of bounds and set " Go to next action" only.

3: Get the last error and capture the error message, then split text by "," and get the textlist[1],  Convert to number, then -2 and get the right page number now.😀

[Paulie78 · Accepted solution]

@nana001  You can do this in a cloud flow quite easily:

Simply get the file content of a PDF file. And then create a compose action with the following expression:

sub(length(split(replace(base64ToString(outputs('Get_file_content')?['body']['$content']), 'Type/Pages', ''), 'Type/Page')),1)

This will return you the count of pages within the PDF file.

[Anonymous]

It works. Thank you.

[SandraSoniec]

@seewo Couldn't get the proposed solution using regex running. Thanks for the imperfect alternative, which works perfectly for my need.

[Frederico]

I like the approach of @Paulie78 , and I'm trying to use it to solve my problem as well. I need to insert a page number at the bottom of my pdfs pages. 

Is there a way (without using the enconde connector) to insert the pages number (and be incremental as word does) in a pdf or a HTML file using Power Automate?

The following is my current flow (canvas app send to the flow a HTML code --> flow creates a file with the HTML ---> convert this file to pdf

Thanks

[JimJ]

Thank you, this was a lifesaver!
For those unfamiliar, in Step 5  - Set Variable, you can expand the variable name to get 'Message.Trimmed.Trimmed' as shown below
For Step 7 - Set Variable I manually modified the variable to add "[1]" to get textlist[1]

[SCGC_Developer]

This one is not work for some types of pdf . For ex, Pdf which was built from scan to pdf ( scan from machine)  anyone has ideas ?  

[minhvo]

I found a solution when using AI model for extracting information.

From AI model, 'predictionOutput' variable contains the "pageCount".

For example, my pdf file has 2 pages. So, 'predictionOutput' value likes this:

{"@odata.type":"#Microsoft.Dynamics.CRM.expando","pageCount":2,"layoutName",....}

My idea is to create a variable 'predOutput' and assign its values from AI model. 

Then the number of page will be equal to:

split(split(variables('predOutput'),',"layoutName"')[0],'pageCount":')[1]

Best,

[JustDabbling]

Hi there. What changes would need to be made in your expression to personalize it for my library? I'm getting an error saying:  "cannot be evaluated because property '$content' cannot be selected." What am I doing wrong? Thanks in advance!

[vitormatos]

Worked for me as well! Thanks.

[powerwood]

Thanks for the response. This doesn't seem to work for me, and I'm not sure why. The output consistently shows as "0". 

Here is a screenshot of my flow: 

Whereas the Compose input formula is: 

sub(length(split(replace(base64ToString(outputs('Get_file_content')?['body']['$content']), 'Type/Pages', ''), 'Type/Page')),1)

Perhaps I'm doing something wrong?

[powerwood]

While trying to get to the bottom of this, I also created a Compose step to only convert base64toString and pointed it to the same Get File Content/body/content. When the automation was triggered, I searched the results and could not find "Type/Pages" anywhere in the output. I only found one occasion of "Page" in the output.

[nana001]

Thank you, this worked for most (but not all) of my PDFs 🙂

[powerwood]

@nana001 , I agree. I tested this on a number of PDFs. Some worked and some did not.

I would be interested if anyone has some insight into any additional document processing that might need to happen to make this work more consistently. I think the issue is that not every PDF has "Type/Pages" in the file content.

[jokemanJ]

Smart solution, it is also pitty that PAD has no direct solution for this question

[PTadrous]

Hi @Frederico  
Did you find a solution for that?
Thanks

[xhevahires1]

Hi @Paulie78 , I tried using the expression above but it's not working for me. Is there anything that need to be changed in the expression? Otherwise, the number of pages that the compose action it's giving me back all the time is 0. Has anything changed about the expression or something like that,  since I see that this is a accepted solution for this case.
Thank you