Eric Sunshine <sunshine@xxxxxxxxxxxxxx> writes: > This reminded me very much of [1] which exhibited the same failure > mode and was due to the same limitation(s) of the OS. > > [1]: https://lore.kernel.org/git/20211209051115.52629-3-sunshine@xxxxxxxxxxxxxx/ Ahhhh. That one gives the official excuse to apply the band-aid. You quoted from their documentation Do not end a file or directory name with a space or a period. Although the underlying file system may support such names, the Windows shell and user interface does not. As this test _is_, unlike the cited patch that was not about a directory with a funny name, about parsing a patch and applying it to a path with a directory with a funny name, I am tempted to keep the test with the filesystem, instead of replacing it with the one using the "--cached" that Peff suggested. I am _also_ tempted to add that "--cached" thing (instead of replacing), though. Thanks
