[BUG]Garbled characters in Japanese environment

[mushi-chan]

Configuration

• Keka version: [1.3.8 (5320)]
• macOS version: [Ventura 13.0.1]

Describe the bug

In Japanese environment, files zipped with keka by Mac are garbled on Windows side. The file name is in Japanese.
Windows on the receiving side is also Japanese environment.
Has there been any other reports of similar errors?
We are not sure if it is a problem with our machine or with the application, but we have verified this on multiple Windows systems, and the garbled characters occur on all of them.

[gingerbeardman]

Here is one I made using Keka to create a ZIP (and 7Z which is not attached here)

• It extracts fine using Windows Explorer or other apps (see below)
• 岡本博-keka.zip

If I make it using standard macOS compress feature, results differ on Windows:

• if I expand it using Windows Explorer results are garbled
• if I use something like NanaZip or PeaZip to exrtact on Windows, everything is fine.
• 岡本博-mac.zip

If you're still seeing problems:

• Can you please provide a sample file?
• And let us know how you are extracting on Windows

Keka version: 1.3.8 (5320)
macOS version: Sonoma 14.4.1 (23E224)

[DaddyF]

This appears to be the same issue I raised earlier: ‘Cannot create zip files for Japanese Shift-JIS’ (#1302). Since Windows uses Shift-JIS and macOS uses UTF-8, Japanese characters are not interpreted correctly. Tools like NanaZip, PeaZip, or 7-Zip may handle UTF-8 and convert characters to Shift-JIS, resolving the issue.

As far as I know, Keka has no plans to implement Shift-JIS support for zip files. Therefore, the solution lies on the Windows side, where apps should support UTF-8, which is becoming the global standard.