>Tim, was your check for File#getName done manually or it's present in tests >somehow? If it's >present in tests we can check it on major platforms (I can >test on linux, win xp and maybe on >macosx) with different jdks.
It was a unit test that initially uncovered the problem -- all worked well on my windows dev environment but was failed when I made the patch in linux. I didn't add unit tests to show that File#getName failed. I can, though, if that would be of any use. I agree that we should keep this as a tiny bit of code and rely on the rest of commons io in core for 2.0. > Also, I'll take a look at unresolved jdk7 issues/patches today. +1, if you want to take over reviewing Yaniv's excellent work, please go for it. I'm mostly tied up in end of fiscal year administrivia. Do not wait for me on TIKA-1744 etc. Thank you!
