add stamp/free text annotation to existing document inc. save, extract embedded font warning comment

[hello_earth]

Hello, thank you in advance for offering a place for the users to ask questions.
My problem is a bit technical - I am modifying a PDF file, adding some stamp (or FreeText - tried both) annotations with embedded subset fonts. I am using incremental save, but this doesn't appear to be related. I have tried multiple fonts too. No matter what I do, the text I draw in the appearance stream of the custom annotation, using a subset font, causes Acrobat Reader DC to show a warning "Cannot extract the embedded font 'XXXXXX+FontName' ...".

IMPORTANT: This warning appears only when I open the Comments pane - when it is not selected to be open by default when "a document with comments is open", a user might not any warnings at all.

I have tried various combinations of options but cannot come up with what is actually needed. I have checked and saw this problem appears in at least 2021.001.20149 and 2020.012.20043 versions. Specifically in the last version, it appears inconsistently - probably because of some caching going on.

Should I use a different annotation type? Is it an expected behavior, because stamps are supposed to be pictures, or some other reason?

I am attaching two examples, one with free text annotation, and another with rubber stamp annotation and greek text. The documents were generated by Microsoft Word 2016 -> Export To PDF/XPS Document. The appending of the annotations and the incremental save were done using Apache PDFBOX v2.0.23.

Acrobat DC Preflight searching for possible Font problems shows no problems. The documents are usable, it is just there is this nagging error box that appears and indicates that something is going wrong. I just wanted to make sure that I am indeed doing something wrong - or this is something expected.

Thank you - and if it is not the right place for such types of questions, I will accept that.

[try67]

I think you'll have to manually embed the font in the file to prevent this error message from happening, or use a Type1 font, or one's that already embedded in it. But since you're using PDFBox to do it I would strongly recommend posting your question to that library's mailing-list: [users at pdfbox.apache.org]

[hello_earth]

try67, thank you for your answer and sorry it took me so long to reply. I was just trying out various things - which ended up in the following 3 files. All three were blank initially with some minimal text - and a single annotation with embedded font with Greek characters:
one file is created with PDFBOX
one file is created with OpenPDF (iText fork)
one file is created with Acrobat from a blank HTML page, with a Text annotation (Comment pane -> Add text box) added to it later using Acrobat Reader DC.

All three when open with my copy of Acrobat Reader DC show ALRIGHT, the text is rendered correctly, but the error "Cannot extract the embedded font" appears when you open the Comment pane (sometimes it opens automatically when you open the document - i know it's somewhere in the Preferences - so some users will have it open automatically and will see the error message box).

These files (at least the ones I created with PDFBOX and Acrobat) are passing the Preflight / PDF-A checks as per their fonts at least. I also tried setting the RC value on the annotation - it works well, but again the text doesn't render at all in Chrome / Firefox default PDF viewers. The 3 files I mentioned above contain AP (appearance stream) on the annotations, and are rendered correctly in all viewers, but they produce this annoying message box in Acrobat Reader DC, which will most surely make the users think that there's something wrong with the PDF.

I am just sharing my pain - I don't expect something to be fixed - I don't even understand what I am breaking here. It doesn't seem that the incremental save is at fault - the files created with Acrobat/OpenPDF are not saved incrementally. It is also not an improper subsetting or broken initial font issue - the original files are alright, and I think I wouldn't get this error in all 3 of these files produced in different ways. It seems that just the fact that I am embedding fonts into annotations' appearance streams triggers this error in Acrobat Reader DC specifically - and that was the reason I wrote on this forum. 

Thank you for reading and looking into this. I am having a question open on stackoverflow about this too: https://stackoverflow.com/questions/67174320/pdfbox-embedding-subset-font-for-annotations-part-2

[hello_earth]

Please, can somebody clarify on this - is this something that might get fixed in newer versions of Reader DC? How would one explain a document modified with Acrobat Reader DC itself producing the same kind of error? (see example from my reply from 30/04/21) I am stuck here because annotations were a design decision on my side, and I need to go with something else if annotations and embedded fonts do not go well (something I didn't exactly saw coming from the PDF specification)

There is a related thread from 2019 that got bumped in march 2020, https://community.adobe.com/t5/acrobat-reader/19-012-20036-introduced-cannot-extract-the-embedded-fo...

They mention the problem appearing only on interactive PDFs - which matches my own problem, as we are using annotations - but it is not clear what is going wrong and what should be fixed. The fonts are rendered alright both during viewing and printing of the PDF - they are also rendered alright with other viewers. It is just that specifically Acrobat Reader DC gives out this error message. Given that the majority of users out there have Acrobat Reader DC installed as the default PDF viewer application, I expect that most of them will start seeing this error message and start panicking.

Thank you very much!