pdfbox COSName should be written with same charset as it was read. PDFBOX-4728

If read is done using Windows-1252 and write using UTF-8 then PDF containing Windows-1252 encoded XObject dictionry names will be broken after doin load and save

PDDocument document = PDDocument.load(sourcePath);
document.save(targetPath); // This will write broken PDF file

Dec 03 '19 11:12 oikku

This should also go to 2.0 branch because this bug affects 2.0.17 version.

Dec 04 '19 06:12 oikku

As far as I know the pdfbox maintainers do not monitor github pull request.

I would advice you to add a comment with a link to this pull request to the issue (or send an email to the pdfbox developer mailing list).

Dec 23 '19 10:12 PascalSchumacher

@oikku the issue you are mentioning has been closed and resolved in pdfbox 2.0.9. Please open a new issue in https://issues.apache.org/jira/projects/PDFBOX describing your issue. Feel free to mention the pull request. It is also advisable to attach a sample PDF and a unit test to reproduce the issue.

Sorry for any inconvenience this may cause but this is only a mirror so the pdfbox maintainers are not notified of this issue

Dec 23 '19 19:12 msahyoun