Troubleshoot user workstations in Unicode installations

In Unicode-mode environments, workstation configuration can affect how filenames and file content are displayed and processed. The following sections describe common issues and their resolutions.

"Cannot translate" error message

Symptoms

The "Cannot translate" error message is displayed when your workstation receives characters from the P4 Server that cannot be represented in the character set configured on the workstation.

For example, if P4CHARSET is set to shiftjis and your depot contains filenames that use characters from the Japanese EUC character set that do not have mappings in Shift-JIS, the error is displayed when you list the files using the p4 files command.

Cause

Your workstation is configured to use a character set that does not support all the characters being sent by the P4 Server. As a result, some characters cannot be translated and displayed correctly.

Resolution

To ensure correct character translation, do not use characters that cannot be mapped by the character set configured on the workstation.

Avoid using unmappable characters in:

• User specifications

• Client specifications

• Jobs

• File names

Strange display of file content

Symptoms

When you display the contents of a text file that contains extended characters, some characters might appear incorrectly. Typical symptoms include question marks (?) or boxes appearing in place of characters.

Cause

Your workstation might not have the font required to display the characters contained in the file.

Resolution

Install a font that supports the characters used in the file.

Files with non-ASCII names cannot be opened by some Windows applications

Symptoms

On a Windows workstation connected to a Unicode-mode P4 Server, a file whose name contains non-ASCII characters (for example, accented Latin characters or Chinese, Japanese, or Korean characters) might not be accessible from some applications after the file is synced, unshelved, or otherwise restored to the workspace.

The file is present in the workspace and can be accessed normally by Unicode-aware applications, but some older Windows applications, editors, utilities, or scripts might report errors such as No such file or directory or fail to locate the file at all.

This behavior can be confusing because the same application might have originally created the file before it was submitted to P4 Server.

Cause

When a Unicode-mode server writes a file with a non-ASCII name to a Windows workspace, the filename is created using Windows Unicode file APIs. Applications that support Unicode filenames can access the file normally.

Some older Windows applications use only legacy ANSI file APIs and do not support Unicode filenames correctly. These applications might be unable to locate or open files whose names contain non-ASCII characters, even though the files exist on disk.

This is a limitation of the application and the Windows API being used, not a P4 Server defect.

Resolution

Use applications and tools that support Unicode filenames when working with files whose names contain non-ASCII characters.

If you must use legacy applications that do not support Unicode filenames, avoid non-ASCII characters in filenames when creating files that will be accessed by those applications.