Update:
- The bug with the validation of our file is with VeraPDF (see github
issue link)
- The file from the SO question wasn't generated by PDFBox (I was so
focused on the font that I forgot to look at /Info 🤦♂️, it was done by
OpenPDF which is an itext fork)
Tilman
On 13.07.2024 06:20, Tilman Hausherr wrote:
While working on https://stackoverflow.com/questions/78735443/ I
noticed that creating a PDF/A-*3*b file with a modified
CreatePDFA.java example and having an empty text as second parameter
produces a file that is not valid PDF/A-*3*b according to VeraPDF.
When there is a text then all is OK. I tried to remove the CIDSet
entry and then it works, but such a file would not be valid PDF/A-*1*b.
Thus either there is some PDFBox bug that I haven't understood re the
CIDSet content, or there is a bug in VeraPDF. Thus I have created this
ticket:
https://github.com/veraPDF/veraPDF-library/issues/1462
I believe that the file in SO 78735443 was not created by PDFBox.
Their subset font file has 938 glyphs, with 937 of them empty, our
subset font file doesn't look like that, it has only 1 glyph (the
"undefined" one).
Tilman
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscr...@pdfbox.apache.org
For additional commands, e-mail: dev-h...@pdfbox.apache.org