Skip to content

HTML entities rendered literally in PDF signature stamp (UTF-8 encoding issue) – LibreSign 12.2.0 #6961

@yharroch

Description

@yharroch

Describe the bug

Accented characters in the signature stamp/footer are rendered incorrectly in the signed PDF. Instead of displaying proper UTF-8 characters, HTML entities appear literally in the final document.

To reproduce

  1. Go to LibreSign settings.
  2. Configure the signature stamp template with accented characters (e.g. “Signé numériquement…”).
  3. Sign a document (email token or simple click).
  4. Open the generated signed PDF.
  5. Observe that HTML entities are rendered literally.

Expected behavior

Accented characters should be rendered correctly in UTF-8 format inside the signed PDF.

Screenshots

Image

Example :
Expected : Signé numériquement par USER
Actual result in signed PDF: Signé numériquement par USER

Environment information

  • LibreSign version: 12.2.0
  • Nextcloud version: 32.0.6
  • PHP version: 8.2
  • OS: Debian Bookworm
  • Embedded Java installed via occ libresign:install
  • Server locale and PHP are correctly configured in UTF-8.

Additional context

The issue seems to originate from the Java/PDF rendering layer.
It may be related to entity escaping in the Twig template or PDF generation pipeline.
No encoding issues are observed in Nextcloud itself.

Metadata

Metadata

Assignees

No one assigned

    Labels

    bugSomething isn't working

    Type

    No type

    Projects

    Status

    4. to release

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions