Skip to main content

Exporting to PDF renders Hebrew text unselectable (exported as shapes/flattened)

  • July 20, 2026
  • 1 reply
  • 9 views

roei meiri

Hi Figma Team,

I am experiencing a critical issue when exporting frames containing Right-to-Left (RTL) languages, specifically Hebrew, to PDF.

While any English words within the text box remain perfectly selectable and searchable in the final PDF, all Hebrew characters become completely unselectable. It appears they are either being flattened into vector shapes or losing their text encoding during the export process.

This is a major blocker for localized designs, especially for resumes and documents that need to pass through automated screening tools (ATS) or simply require selectable text.

Could you please look into how the PDF engine encodes RTL/Hebrew text strings upon export?

Thank you!

1 reply

Jaycee Lewis
Figmate

Hello ​@roei meiri  👋 Thanks for the interesting question 🤩 I ran a quick test and I want to make sure I have fidelity with how you're testing.

A few follow-up questions for you:

  • When you say the Hebrew is "unselectable," which of these are you actually seeing: (a) the highlight refuses to cover the Hebrew at all, (b) the Hebrew highlights but the selection jumps in a weird order across the English/numbers, or (c) it highlights but a Ctrl/Cmd+F search for the Hebrew word returns no result?
  • Which font is applied to the Hebrew text? (And is it the same font for the English, or is Hebrew falling back to a different font?)
  • Which export path did you use — the right-panel Export with format set to PDF, or File → Export frames to PDF? And does it happen on both?
  • What are you opening the PDF in when you test selection (Chrome, Preview, Acrobat, etc.)?

My quick test:

  • I made a new Design file
  • Added a single text layer
  • Used this mixed-script string as my control:
Hello שלום world 2026
  • Exported via both PDF paths (right-panel Export set to PDF, and File → Export frames to PDF)
  • In the exported PDF, I then: drag-selected across the line, ran Cmd+F for the Hebrew word, and tried to copy-paste it into a plain text editor

What I found: the Hebrew highlights when I drag-select (so it's not flattened to vector shapes), but Cmd+F for the Hebrew word returned no results in any test. So the export carries visible glyphs but a broken character mapping for the Hebrew. This could explain why it can't be searched, copied, or read by tools like ATS, even though it looks like normal text.

Your case may differ (e.g. Hebrew that won't highlight at all, or a different font), so I want to make sure before moving forward. Thanks again! — Jaycee