Sorry, I can't quite tell what's what in the blob of text below. If it's not trailing spaces, it could be either unprintable characters or newline characters in either the name or the path. They usually come from Zone OCR reading multiple lines and regular expression in Pattern Matching being less restrictive than needed and including them in the value.

They're not easy to find, you would have to open the document's XML file in a text editor that renders them, like Notepad++. The files would have extension .bxd and be located in subfolders in <ProgramData>\Laserfiche\Quick Fields\Files\<Session ID>\bxd.

For ex, if you see something like in the screenshot below, the folder name has a newline in the middle so it would be invalid:

A proper one would be on a single line:

If you open the application errors pane (F12) and copy the details of the "Invalid object name" and paste them here, it might give us a better idea whether it is the folder or document name.

so that is the issue. This is what I'm getting. 

How do we resolve the issue?

With QF closed, take out the line break between the \and "AM Conservation...", save the file and reopen the session. That will allow the existing docs to be sent.

Where is that value coming from?

Just like you explained, its a token coming from a regular expression. This customer deals with a shifting issue on her AP paper work and a small OCR zone box is not working. I instead went with a larger zone and made an expression to capture the vendor name. The zone always gets the top of the paperwork which the last line of text is a phone number. The regular expression is set to capture everything after that. The vendor name is lower then the phone number so it must be grabbing the line too. This is what it looks like.

I'm trying to get the regular expression but quick fields wont let me edit the pattern matching configurations or anything for that matter. What is preventing me from editing the session?

ok, I restarted Quick Fields and got it. 

Is there anything i can do to the pattern to not capture the line break?

I think if you change it to 733-5423.*([^\n]+)\[\d+\] it will do what you want. The capture group now specifies anything but a line break ("\n") and allows for extra lines between the phone number and the company name.

Perfect, that's what I needed. Thank you. Ill let you know how it goes. 

When I tested yours in the token editor, it wasn't working. I put the ( in front of .* and the testing seamed to work but its also looks like its adding spaces. I reran the document with your and my new expression with no luck. Yours returned no value, mine got the vendor name but its the same problem again with blank spaces. 

ok, I feel like I'm going crazy here. I copied the text out of the Zone OCR so that I can test the actually text the the box is retrieving. I past that into the token editor so I can test my expression with quick fields. This is what it get. 

Looks great right? Also I have been testing in Regex101.com here is what i get from them.

ok, now here is what happens when I run the session on my document.

Vendor Name3 is my token... I'm completely  stumped. Please help!

Hello again, your expression seamed to work great in regex101 and in the token editor testing. When I run the sessions everything looks great too. The issue it still there though. whats really weird is that all the fields looks great but this is what I get from the Notepad++

as you can see in folder path, there is no line break but there is also no vendor name, even though that field is being populated. What does that mean?

here is another one i just did.

Which one? The first or the second one above. It might be easier to just open a support case and attach sample images at this point...