PDF generated by OO Document Server has CID/Identity-H fonts — text becomes unreadable after Adobe Acrobat page insert/replace operations
Environment:
- OnlyOffice Document Server (Kubernetes deployment)
- Converting .docx files to PDF via the conversion API
- Custom corporate font
Problem:
PDFs generated by OnlyOffice Document Server use CID fonts with Identity-H encoding for the corporate font. When we perform page insert/replace/delete operations in Adobe Acrobat Pro on these PDFs, the text becomes unreadable — characters are garbled (e.g. S#. FNBNDICL 1 instead of SCHEDULE 1). The layout is preserved but character mapping is corrupted.
Adobe Document Properties → Fonts shows: corporate font (Embedded Subset) — Type: Type 1 (CID), Encoding: Identity-H
What we’ve tried:
- Ghostscript post-processing with -dSubsetFonts=false -dEmbedAllFonts=true — fonts become “Anonymous” because GID mapping in the PDF doesn’t match standard TTF encoding
- Ghostscript with -sFONTPATH pointing to the actual TTF files — same result
Questions:
- Is there a configuration in local.json or the conversion API to make Onlyoffice use standard Type 1 encoding instead of CID/Identity-H when embedding fonts in PDF output?
- Is there a setting to fully embed fonts (not subset) during DOCX→PDF conversion?
- Is there a known workaround for making OO-generated PDFs compatible with Adobe Acrobat page manipulation?
@Alexandre would you plz respond to this issue.
Hello @kashyaprajan056
Please keep in mind that this forum is a community support platform where we try to help everyone. Due to the large number of requests, responses may sometimes be delayed, but we still do our best to assist all users (Get dedicated support).
If you have a commercial license (which I assume may be the case, since Kubernetes deployment requires a commercial edition), it would be better to use our dedicated support channel via Zendesk:
https://helpdesk.onlyoffice.com/hc/en-us
Since colleagues of mine there will be able to assist you more promptly.
Now, about the issue itself, could you please provide the following additional details:
- Your exact Document Server version, and the installation guide you followed.
- Please demonstrate the issue step by step. If it’s possible, record a videofile showing the workflow and the resulting issue.
- If you are using the Conversion API to convert docx to PDF, please provide the exact conversion request you are sending. You can replace the Document Server url and any sensitive values with gibberish.
- If the issue reproduces only with this specific corporate font, could you provide the font itself for internal testing?
I also tried to identify your license but was not able to find it. If you have a commercial license, you can send the license details to me via private message.