Skip to content

[Bug] Greek text is invisible in .txt files (UTF-8) on Windows (Java 8) #28

@greektv

Description

@greektv

When viewing a text file containing Greek characters encoded in UTF-8 (with BOM), the Greek characters are completely invisible in the Gophie text viewer. Punctuation and English characters in the same file render correctly:

Image Image

However, Greek characters render perfectly in the Gophie Menus/Directory listings:

Image

The issue is isolated strictly to the text file viewer.

To Reproduce

Launch Gophie on Windows (running under JRE 8).
Navigate to gopher://gopher.greektimes.ca
Enter the "Latest News Stories" folder.
Open any article (e.g., one with a Greek title).

Observation: The article body text is invisible where Greek characters should be. English text and punctuation appear fine.

Expected behavior
The text viewer should display the Greek characters, just as the Menu viewer does.

System Information
OS: Windows 10/11

Java Version: JRE 1.8.0_471 (x64)

Gophie Version: 1.4 (Latest Release)

Troubleshooting & Analysis Performed
I am the SysOp of the server in question (gopher.greektimes.ca). We have performed extensive testing to rule out server-side issues:

Server Encoding: The files are saved as UTF-8 with BOM (utf-8-sig in Python).

Cross-Client Verification: The exact same files render correctly in:

Gopherus (Console Client)
Gopher Browser for Windows

Image

Menu vs. Text: Since Gophie displays Greek correctly in the Menus, the application clearly supports UTF-8 decoding. The issue seems specific to the Font or Input Stream handling in the text file viewing component.

Encoding Test: We attempted to serve the file as Windows-1253 (Legacy Greek). Gophie rendered this as "Mojibake" (garbage characters), confirming that it does not auto-detect legacy code pages.

Hypothesis: It appears the text viewer component is either:

Using a default Monospaced font on Windows/Java 8 that lacks Greek glyphs (rendering zero-width or empty space).

Or, treating the InputStream as the system default encoding (CP1252) instead of detecting the UTF-8 BOM.

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions