11
Replies
++EDITED REPLY , fixed typos
Hi pdc@TD ,
If I am not mistaken, all PDF producing software use some sort of core JavaScript engine to edit PDF objects in a PDF document and also to perform arithmetic operations (among other built-in features).
But if you were specifically referring on how to find out the presence of any JavaScript scripts while a PDF document is viewed in Adobe Acrobat Pro DC (requires paid-subscription), you may go to Tools => JavaScript => All JavaScript.
This method will show the user all of the scripts that are currently used in a PDF file .
Now, based on your primary inquiry, I will assume that in both of your questions you are referring to how to check many documents in bulk and determine if all of those files are both tagged and also contain JavaScript objects.
Is that correct?
If yes, you may employ the Action Wizard tool and customize an new action in combination of other readily available Acrobat built-in tools or with a custom JavaScript script.
However, based on your requirement criteria I wouldn't recommend doing it from the Acrobat Pro program due to possible crashes.
In addition, the Action Wizard will try to open up each document one at a time as it performs the custom actions to allow user interaction as each document is processed on real time (which makes the whole experience tedious and not efficient).
You're better off if you manually do a batch script.
Be aware that in the case of checking for tagged PDFs with a batch script, such script will only check for a text string that may indicate the presence of a parent tree mapping object. In this case, if the batch script is successful it will indicate if such PDFs have a document structure defined (nothing more).
Moreover, it is also worth noting that the output results of a batch script (like the one I am sharing below) shouldn't be confused with Accessibility checks nor it is meant to validate PDFs to see if they conform to a required standard; there is so much more involved with accessible PDFs (such as PDF/UA compliance standards, or detecting problems with embedded fonts, for example).
Anyway, if you open a PDF that you know is tagged using a file text editor (such as Microsoft's Notepad (if on MS Windows), or Notepad++, and TextEdit (if on a macOS)), you would want to empasize in your batch script to look for the string of the "StructTreeRoot" element (or property).
So, just for testing purposes, if you open a PDF with a text editor and perform a search for the "StructTreeRoot" string pattern, then chances are that the PDF document is tagged; and that is due to the presence of the StructTreeRoot catalog in those files. See screenshot:
You may test with another file using the same method above to search for the "JavaScript" string pattern; if the PDF has JavaScript scripts on it, the search will highlight such text string indicating that one or more JavaScript object(s) are in use with that file.
With these observations in mind, now you can employ a batch script like the example below:
@ECHO OFF
cd C:\Users\userAccount\Desktop
mkdir C:\Users\userAccount\Desktop\Tagged_PDFs
for /f %%a in ('findstr /M "StructTreeRoot" *.pdf') do XCOPY /Y %%a C:\Users\userAccount\Desktop\Tagged_PDFs >NUL
cd C:\Users\userAccount\Desktop\Tagged_PDFs
mkdir C:\Users\userAccount\Desktop\Tagged_PDFs\PDFs_with_JS
for /f %%b in ('findstr /M "JavaScript" *.pdf') do XCOPY /Y %%b C:\Users\userAccount\Desktop\Tagged_PDFs\PDFs_with_JS >NUL
START C:\Users\userAccount\Desktop\Tagged_PDFs\PDFs_with_JS
This batch script is meant to be executed from any directory on a Microsoft Windows computer. And it will expect to look for files in the User account's Desktop folder (i.e. C:\Users\yourUserAccount\Desktop directory) .
As you execute this batch script, the first action will create a new subfolder: "Tagged_PDF's" and then it will look for every PDF in the Desktop's parent directory for tagged PDFs that match the "StructTreeRoot" string pattern.
The last portion of the first script will copy the list of matched files that were found in C:\Users\yourUserAccount\Desktop parent directory to the new "Tagged_PDF's" subfolder that was created earlier.
Soon after the first script finalizes it will execute the second script, which will switch directory from C:\Users\yourUserAccount\Desktop to the Tagged_PDF's subfolder.
In there, it will create a new subfolder named "PDFs_wtih_JS", and it will perform another string search on the files that were identified with a tagged structure and copied to the Tagged_PDF's folder ; this time it will look for the "JavaScript" string pattern and copy the PDFs files that match that string pattern from the Tagged_PDF's folder to the new "PDFs_wtih_JS" subfolder.
The last line of the batch script just opens up a new File Explorer window of the "PDFs_wtih_JS" folder, in which only the files that are both tagged and that have a JavaScript code in them will be listed.
Although this solution is NOT entirely realted to Adobe Acrobat, it will allow you to run through thousands of PDFs and sort them in a folder to indicate which files contain both tag properties and JavaScript objects .
ADDITIONAL NOTES:
Batch scripts that process too many files may hog other background services or programs that may be opened simultaneously as the script executes and parse those files.
I would suggest to test first with no more than 250 PDF files at a time and see how it performs.
Thanks for the detailed answer! I have tried to look for keywords in PDFs using Notepad++, but I wasn't sure what to look for. Are "StructTreeRoot" and "JavaScript" 100% dependable?
Surely there's a totally reliable way of doing this. I've heard of a product called SiteImprove which can tell if a PDF is tagged. How does it know? And when a screen reader like JAWS reads a PDF, it reads the tag tree. How does it do it? There's an API that can be exploited for such things. All you need is a language like C to access it.
Adobe is still part of the ISO's specification committees. Two full-time Adobe engineers are on the committees for all types of PDF and are active contributors, sometimes chairing working groups.
The PDF Association is designated by the ISO as the entity to write and develop all PDF standards, and the association has a fair amount of reference material on their website. Here are some materials you might find helpful:
- Free access to the ISO 32000 reference standard for PDFs. This is the core standard that every type of PDF must follow. Download form https://pdfa.org/resource/iso-32000-pdf/
- The companion standard for PDF/UA-1 (ISO 14289) is available (for a fee) from the PDF.org: https://pdfa.org/resource/iso-14289-pdfua/
- The PDF.org also has a free syntax guide to accessible tagged PDF. This is used by most of us in the accessibility industry to ensure the tags and encoding meet accessibility requirements. You might find this helpful for your work in accessible PDF. Download from https://pdfa.org/resource/tagged-pdf-best-practice-guide-syntax/
... tagged is not equal to accessible.
This is correct. Not only must a PDF be tagged, all real content must be tagged — with the correct tag, as well. Structure and syntax are critical for accessible PDFs just as with any markup language...HTML, XML, JATS, etc.
Looking for only the metadata info that indicates the PDF is tagged is not sufficient. A PDF with only <P> tags is not accessible by any means of the imagination. And a PDF with an incorrect reading order is totally useless to those using assistive technologies.
There are many 3rd party tools and services that will examine PDFs for accessibility, and some claim to remediate the errors. But none of them can determine the status with 100% accuracy; this process still requires a human interpretation to determine what the correct, most appropriate tag and structure is for the content. AI doesn't do that well enough for most documents.
But these tools will do a much better job of evaluating PDFs than you'll have with your own script. For one, all of these vendors know the PDF and PDF/UA standards and most are on the standards committees. And they didn't develop their software tools in a day, either. Their tools have been used by the industry for 10-15 years.
Dust off your wallet as none of these tools/services are cheap:
- SiteImprove https://www.siteimprove.com/ mainly checks websites, but now includes PDFs on websites.
- CommonLook https://commonlook.com/accessibility-software/
- Allyant, which recently acquired CommonLook and other PDF-remediation tools https://allyant.com/
- Equidox https://equidox.co/
- PAC and Axes4 https://www.axes4.com/en The PAC checker is free, but the full software to correct the PDFs isn't. Axes4's checking service is at https://www.axes4.com/en/products-services/axescheck
- Crawford Technology https://crawfordtech.com/
AdChoices