I've edited the first post in this thread to version 1.7.
I adopted
andresp's idea of using code page (CP) 65001 (UTF-8) to support file names that contain characters that are not in your active code page. However, the feature has not been adequately tested in my mind, so I only activate CP 65001 if you use the new /U option - and the original code page is restored upon completion.
I have a couple concerns:
1) I attempt to capture the current code page so I can reset the code page at the end. However I'm not sure that the code is robust enough to work with all languages - the CHCP output may change for different languages and break my code.
2) There are reports that CP 65001 breaks some legacy programs like FIND and MORE. So I replaced all three calls to FIND with FINDSTR /L. But I'm worried that I originally used FIND for a reason. FINDSTR is finicky/buggy, and I'm not 100% confident it will always work properly in this context.
Please report any problems with the new /U option here.
Dave Benham