Mor1arty fixes Mike's Chinese filename problem
A tiny upload patch that quietly matters for any firm handling non-English documents.
When you upload a file to Mike, the server reads the filename using an old text-encoding convention that mangles anything that isn't plain English letters. Chinese characters, accented names, Cyrillic, Arabic - all turn to garbage by the time the file is stored. Mor1arty, who appears to be running Mike against Chinese-language documents, has applied the standard fix: re-read the filename in the modern encoding the rest of the world uses, so the original name survives intact.
It's a ten-line change, low-risk, and a no-op for English-only users. But it points at a real gap in the upstream project - any deployment touching documents named in a non-Latin script has been silently losing that information. For a legal tool, where the filename is often the only human-readable handle on an exhibit, that's not a cosmetic issue.
Spotted something wrong? Or know the PR text has fresher detail than the writeup above?