PDF Corrupted in Sharepoint

[Anonymous]

I created a flow to grab a PDF attachment from an email when one arrives and save it in a Sharepoint document library. The file is created correctly but it cannot be previewed or opened in Adobe Acrobat. The attachments can be opened just fine when I open them through the email. When I try to open it in Adobe, it gives this error:

Here are screenshots of my flow. I omitted the middle section but it is just variable manipulation and shouldn't have any impact on the file content. Any help is appreciated!

[abm]

Hi @Anonymous 

I tested this and it is working as expected for me.

Please make sure you have another loop above the attachments loop. I couldn't see that from your screenshot you provided.

Thanks

[chsanche]

Hi @Anonymous 

I tried to reproduce this in my environment but it seems to be working fine. I created a simplified version of the flow:

Can you please remove the variables from the create file action to test and let me know the result?

[ScottShearer]

@Anonymous 

As a test, can you saving the file from Outlook directly to the SharePoint library or download/upload the file to the library.

Try to open it from SharePoint and see if you experience the same issue.  The reason is to determine if the issue is with your Flow or with your workstation or the SharePoint library.

[yashag2255]

Hi @Anonymous 

Can you also check if the email has an attachment of type pdf only. Note that images in the body of the email or the signatures containing images area also pulled along with attachments and those might be the issue as well. 

Hope this Helps!

If this reply has answered your question or solved your issue, please mark this question as answered. Answered questions helps users in the future who may have the same issue or question quickly find a resolution via search. If you liked my response, please consider giving it a thumbs up. THANKS!

[Anonymous]

Thanks for your help everyone. @abm @chsanche @yashag2255 @ScottShearer 

I tested with a different PDF and after running it through the flow, it could be opened in Sharepoint so it is an issue specific to the PDF I'm using. I then uploaded the "broken" PDF directly into the Sharepoint library and it could be opened just fine. So the PDF is fine as is but somehow gets corrupted when it is run through the flow. Any ideas?

Maybe these screenshots of the Create File action when I run the flow with the corrupted PDF will help?

[yashag2255]

Hi @Anonymous 

For some reason the attachment type is getting converted to octet stream. Can you try wrapping the content in the base64 function and try again. 

Hope this Helps!

If this reply has answered your question or solved your issue, please mark this question as answered. Answered questions helps users in the future who may have the same issue or question quickly find a resolution via search. If you liked my response, please consider giving it a thumbs up. THANKS!

[abm]

Hi @Anonymous 

Another thought related to upload. Try uploading the file to OneDrive or FileSystem and see if that works or not? This will give some indication where the problem is only for SharePoint or the PDF file.

Thanks

[Anonymous]

Thanks @abm 

I switched out the Sharepoint Create File action for a Onedrive for Business Create File action and I still have the same problem--the PDF I need to use is corrupted after being run through the flow and a different PDF that I tested works fine. Any other suggestions? Thanks

[Anonymous]

Thanks @yashag2255 

Both the "corrupted" PDF and my other test PDF that works fine are being converted to "octet-stream" according the input data in the Create File action. If you think wrapping the content in the base64 function may still be the solution, I will give it a try. I will need some clarity on how to do that though. Thanks

[abm · Accepted solution]

Hi @Anonymous 

Did you try to use the same filename when you create via Flow? 

The test reveals that you have some specific content which is inside the PDF doc. Look the content whether its mapping anything specific. How it gets generated compared to other PDF docs?

Thanks

[Anonymous]

@abm I changed the file name from my custom name to the original file name using dynamic content and this time it created 3 files at once in Sharepoint--the PDF (which can now be opened!) and two images which were part of the signatures in the test email I was forwarding.

I deleted the signatures in the test email and tried again with my custom file name and the images are no longer created and the PDF can be opened now.

I guess if you're unknowingly capturing multiple attachments at once, you won't know it unless you keep all dynamic content the same in the Create File action. 

[abm]

Hi @Anonymous 

Thanks for the update. Glad that you found the issue. The signature file names are fixed so you could check that in an IF condition to exclude those.

Thanks

[ZERAREZ]

Hi,

I believe I am encountering the same issue; however, in all file types. 

My initial goal is to simply upload copies of all email attachments to a specific SharePoint Document Library. I was able to do that but the files being created automatically are all corrupted/non-functional and 4 bytes in size.

I have tried a work around re: base64ToBinary(items('Apply_to_each')?['contentBytes']); however, it's giving me a 'Null' error. I'm not sure where to go from here after searching the web for possible fixes and failing.

Appreciate any help,

Neil

[traceyannspence]

Did you resolve this. I'm having similar issue. the PDF file uploads through flow from onedrive and the file is fine on onedrive and is corrupt on SharePoint.

[abm]

Hi @traceyannspence 

Try the below

base64ToBinary(items('Apply_to_each_2')?['contentBytes'])

[traceyannspence]

I got this error message when I put that formula in (changed "apply_to_each_2" to "apply_to_each")

InvalidTemplate. Unable to process template language expressions in action 'Create_file' inputs at line '0' and column '0': 'The template language function 'base64ToBinary' expects its parameter to be a string. The provided value is of type 'Null'. Please see https://aka.ms/logicexpressions#base64ToBinary for usage details.'.

THANK YOU!!

[ZERAREZ]

Hi,

It's been a while but yes, I was able to resolve this just not sure which part I made an update. Below are the parts of the flow I made, hope it'll help:

I first set a condition to specify what file types I am looking for. I ad to do this to avoid creating files for all items included in the email. 

Then, the next step is to have the flow rename the items to be uploaded in SharePoint following a define format:

Get Attachment Name: 

first(split(variables('rename attachment'), '.'))

Get Attachment Extension:

last(split(variables('Rename Attachment'), '.'))

Compose:

concat(outputs('Get_Attachment_Name'), '_', formatDateTime(convertFromUtc(utcNow(), 'Singapore Standard Time'), 'MMddyyyy hh:mm'), '.', outputs('Get_Attachment_Extension'))

Then just create a file in SP:

I believe the update I did that fix the corrupted error was defining the file types I needed.

[traceyannspence]

I think I've *almost* figured this out 🙂
this is the gist of my flow that takes files from Onedrive and uploads to SP and then updates the properties with data from a spreadsheet. 
I think I need to add a "get file content" in before the create file step... maybe with some combination of using the binary formula

I'll keep you posted :).

[traceyannspence]

UPDATE lol... this was exactly my problem. It works perfectly now with get content added in .... don't even need the binary function.