(Adobe Acrobat DC) Converting TEXT layers to SHAPES during export...

[Under S.]

Using Acrobat only, I would like to load up a PDF file and re-save it with the text layers converted to shapes. Ideally, with absolutely nothing else being affected.

Is there an export option to help me do this?

Thanks!

[Dov Isaacs · Accepted solution]

There is no predefined concept in PDF of text layers. PDF files don't even have to have any layers and if they do, they are not bound by any rules (at least in terms of the PDF specification) as to whether they may or may not have text within them.

If you wish to convert text as realized with fonts into filled (and/or outlined) polygons, you can do this with Acrobat Preflight's Convert Fonts to Outlines profile. This won't affect anything other than text as realized with fonts, but it will make your PDF file uneditable with the Acrobat text editing tools, incapable of being searched for text, incapable of reasonably being exported to Word or similar formats, could seriously bloat the file size, and may significantly reduce quality of print and especially display, typically making text look overly bold.

Any good reason for wanting to do this? Other than for certain sign cutting equipment, there really isn't any rendering reasons that we know of that make this a good practice?!?!?

          - Dov

[Under S.]

https://forums.adobe.com/people/Dov+Isaacs  wrote

There is no predefined concept in PDF of text layers. PDF files don't even have to have any layers and if they do, they are not bound by any rules (at least in terms of the PDF specification) as to whether they may or may not have text within them.

In that case, I must be using the incorrect terminology. My apologies, if so.

After 15 years of using Adobe products, I know of 3 ways to display text in a PDF file :

1. As pixels (simply flattening the text with the images)
2. As text (vectors + you can copy it from a PDF file and paste it into notepad)
3. As vector shapes (vectors, but no copy/pasting)

Currently, I have PDF files with (what I like to call) regular text. Meaning, when I saved my PDF file in Photoshop, I did not manually convert the text layer to shape first. As a result, when I open the resulting PDF file in a reader and hover over the text, the cursor changes to the TEXT cursor (allowing me to select a sentence and copy it outside the document). Web crawlers (like Google's) can also peek inside the document's text content and gather info for ranking purposes, because it sees actual words there.

Had I converted the text layer to shape before exporting, there would be no way for the PDF file to know what words those shapes represent. Web crawlers won't see any words there. Placing my cursor on top of my text in this context does nothing, either. The cursor remains an arrow and none of the words are selectable. In this scenario, the text layers have no properties distinguishing them from other vector layers.

So my question is...

Is there a way, through Acrobat, to load up a PDF file with (what I like to call) regular text layers... and re-save it as (what I like to call) shapes, without affecting anything else? In other words, mimic what would've happened had I converted the text layer to shape before saving the PDF file during the original PS export.

Thanks.

[Dov Isaacs]

I provided the solution in my response above. Use the Preflight Convert Fonts to Outlines fixup. That will do exactly what Photoshop does when it converts text to shapes (as opposed to rasterizing them).

          - Dov

[Under S.]

https://forums.adobe.com/people/Dov+Isaacs  wrote

If you wish to convert text as realized with fonts into filled (and/or outlined) polygons, you can do this with Acrobat Preflight's Convert Fonts to Outlines profile. This won't affect anything other than text as realized with fonts, but it will make your PDF file uneditable with the Acrobat text editing tools, incapable of being searched for text, incapable of reasonably being exported to Word or similar formats, could seriously bloat the file size, and may significantly reduce quality of print and especially display, typically making text look overly bold.

If I upload a PDF file on my site's root folder, Google will eventually index it; especially if my site links to it in any way. It will sniff out whatever info it can gather from the file and use that as preview copy in its search result listings. That's why you can see preview text to PDF documents in Google results before opening the file.

But if the text is converted to outline/shapes (or polygons) then it can't do that. The only info that could be mined from the file is whatever's in the meta areas (title, description, etc.) because as far as crawlers are concerned, there's no text in there.

Regardless of why someone would want to protect the textual contents of their PDF files from outside sniffing (and want them only to use what's in the meta area) is there a way to do this WITHOUT converting all text layers to outline/shapes? Maybe a "Protect text" checkbox somewhere in the export dialogs that could 'shield' page content from a.i. while remaining visible to human eyes?

[Dov Isaacs]

I understand what you are trying to do but unfortunately, you can't have your cake and eat it too! 

There is nothing in the PDF file format that provides the facility to “somehow” protect the text from being accessed but allowing it to be rendered fully and properly other than password protecting the PDF, i.e., requiring a password to actually open up the PDF file.

In terms of Google, I was under the impression that there is an HTML tag you can use on your site to advise Google that you don't consent to their nosing around on your website and indexing anything or specific items. I don't know the details, but I am sure you might even be able to use Google to find that.

          - Dov

[Under S.]

I understand what you are trying to do but unfortunately, you can't have your cake and eat it too!

There is nothing in the PDF file format that provides the facility to “somehow” protect the text from being accessed but allowing it to be rendered fully and properly other than password protecting the PDF, i.e., requiring a password to actually open up the PDF file.

Exactly the confirmation I was looking for, thanks (you'd be surprised how many times I asked "cake & eat it too" questions that someone had a solution for, guess I'm bound to overreach every now and then).

Maybe a nifty feature to add in the future, then? As privacy becomes more of a concern, having a "switch" that could hide the PDF's text content *like* it was password-protected (so that it's really only visible to human eyes) could prove useful. After all, bots do more than simply index, they mine for personal data as well. Heck, you could even build in an exception for Google, if ranking is a concern.

Just throwing the idea out there in the universe.

PS: I just discovered that InDesign allows us to protect PDF text content from being COPIED into ram, without requiring a psw to open the document. So we're kinda halfway there already.

[Dov Isaacs]

This isn't a matter of adding a feature to Acrobat, but rather, adding a feature to PDF which is now an international standard. Adobe cannot do this unilaterally.

In terms of you PS, the protection mechanism you mention preventing “copying of PDF” doesn't prevent reading and more importantly, depends upon the reader obeying the specification. It is advisory and although Adobe PDF products obey these protection mechanisms, don't depend on other PDF products to do so; we know some that absolutely don't!

          - Dov

[atutton]

Thanks for the awesome info - I know you felt that this was a weird need, but we used it today to get artwork from a sign vendor into outlines and over to After Effects. There's applications for this and it definitely saves hours of liaison with the sign maker to try and find someone who can convert on there end, or locating fonts on my end, or butchering the design.