Extracting Data from a PDF File

[CorrespondenceMan]

My organization uses third party software (OpenText - Exstream) to create PDF files.  Within one PDF file are multiple, multi-page documents for different recipients.  The first page of each document contains a white line of data (~30 characters) in the exact same location across all documents.  I need to be able to extract just this data from the first page of each document and output it to an external file (preferably a text file).  Any thoughts?  Thanks in advance.

[Thom Parker]

Were you planning on doing this extraction with Acrobat, or do you have some other PDF tool?

[CorrespondenceMan]

That is what I'm trying to determine.  What are my options?  Does Acrobat do what I'm describing?  My research suggests the Redact functionality might be relevant.  However, I don't know about the extracting to an external file requirement using Redact.

[Thom Parker]

In Acrobat, a a script or plug-in can scan page content. So yes, Acrobat can do this.

I written scripts for doing exactly this type of thing many times. 

In JavaScript the relevant fucntions are "doc.getPageNthWord()" and "doc.getPageNthWordQuad()".

Here's the reference entry:

https://help.adobe.com/en_US/acrobat/acrobat_dc_sdk/2015/HTMLHelp/#t=Acro12_MasterBook%2FJS_API_Acro...

[try67]

How many pages do your files can have?

[CorrespondenceMan]

Any one file can have hundreds of thousands of pages.  But the data only exists on the first page of each document.

[try67]

And how would the tool know where each "document" starts within the file?

[CorrespondenceMan]

I don't know as it depends on the robustness of the tool.  I can tell you the data resides consistently in the same location on the page and if it doesn't exist on a page, then nothing will be in that location.

[try67]

In that case I would not recommend doing it in Acrobat. It's just too much for a script to be able to handle.

I would do it using a stand-alone tool, which is much more robust and can process much larger files, much faster. If you're interested in hiring someone to develop such a tool for you feel free to contact me privately via [try6767 at gmail.com] to discuss it further.

[CorrespondenceMan]

Thank you.  I will add it to my list of possibilities which, as of the moment, is a list of one. LOL