Eric Blake wrote: > According to Jim Meyering on 10/1/2009 1:59 AM: >>> So I'm guessing that the bug was related to actual directories, where rename >>> ("dir/","new") failed (with what errno?) while rename("dir","new") >>> succeeded; >> >> That is definitely the case in question, as seen in the >> original tests/mv/trailing-slash. I don't recall which errno. > > I think my current patch series will still accommodate that ancient bug, > but I'm no longer worried if it doesn't (the new test-rename will catch > it, and we can deal with any user reports if they come in). > > Here's the current state of the series, finally ready for review. If we > check it in as-is, then coreutils will have everything it needs to ensure > consistent behavior of 'mv -T a b/' on every platform it already supports > except for cygwin 1.5 (which has never been a show-stopper for coreutils
I've skimmed through all of that. Considering I spent only 20 minutes, I can't have done it justice, but didn't spot any problems, either. > releases in the past). Or you can wait a bit longer for the next few > things I want to fix: > > test-rename fails on mingw (but coreutils already fails to support mingw) > > renameat is available on Solaris 9 and 10, and needs the same fixes as > rename (but coreutils isn't using renameat yet) Please push that, and I'll make a coreutils snapshot soon. The beta release will probably wait at least a day or two more, regardless, so maybe that will give you enough time.