On 05/18/17 18:52, Stefan Dösinger wrote:
Am 2017-05-18 um 17:46 schrieb Stefan Dösinger:
Am 2017-05-18 um 14:35 schrieb Piotr Caban:
Could you please check if there really is such a length limit in wchar version of the function?
I'm working on it. What is the deal with wchar_t vs WCHAR in the implementation vs the tests?
Also I am not sure the existing test establishes the -3 for the A version. I'll extend it if my reading is right.
Yes, the restriction exists. I guess it is to make sure concatenating "\.." is safe, or something like that.
I had to create a bunch of (too) long directories on my Linux computer and share it with Samba on Windows to test this. Windows apparently refuses to create a directory with a name longer than 240 characters. I guess this is why we don't have an actual test for this. msvcp doesn't make a difference between a directory that is too long and one that doesn't exist.
I guess we could try to create long directories by creating short ones in a a subdirectory, and then moving this subdirectory into a longer subdirectory. This smells like the entrance to a dragon lair housing 260 dragons and I'd rather not venture inside.
I don't think it's a test that should go into wine. It was just something worth checking. Thanks for doing that.
Piotr