On Wed, Apr 22, 2020 at 11:48 AM Brandon Casey <drafnel@xxxxxxxxx> wrote: > > I just looked in config.mak.uname, and I'm surprised to see > FREAD_READS_DIRECTORIES set for so many platforms. And it's set for > Linux and Darwin?!?!? Junio added it for Darwin > (8e178ec4d072da4cd8f4449e17aef3aff5b57f6a) and Nguyễn Thái Ngọc Duy > added it for Linux (e2d90fd1c33ae57e4a6da5729ae53876107b3463), but > also seemed to mistake the intention of FREAD_FREADS_DIRECTORIES as > being about the fopen(..., "r") of a directory rather than about an > fread() of a directory. > > I just wrote a test program and tested on Linux, Darwin, and Windows. > Linux and Darwin both succeed to fopen() a directory and fail to > fread() it, as expected. Windows fails to fopen() a directory. > > I notice this earlier commit mentions a failure of t1308 > (4e3ecbd43958b1400d6cb85fe5529beda1630e3a). I wonder if this is the > reason FREAD_READS_DIRECTORIES was added to so many platforms? Whoops, I got the order of e2d90fd1c33ae57e4a6da5729ae53876107b3463 and 4e3ecbd43958b1400d6cb85fe5529beda1630e3a wrong. Looks like the misunderstanding of FREAD_READS_DIRECTORIES in e2d90fd could have been the cause of all of this. That commit introduced the test t1308 and added FREAD_READS... to Linux, kFreeBSD, and FreeBSD, and the other additions followed shortly after. -Brandon
