Stefan Fuhrmann <stefan.fuhrm...@wandisco.com> writes: > I had two basic options to fix the bug. > (1) Use svn_string_compare that is there for exactly that purpose. > (2) Fall back to low-level C-string ops and hope to get it right this time. > > I chose with option (1). We use the "sizeof(define) -1" construct > in at least 32 other places already. I simply went the extra mile > and documented it, which makes it look like something out of > the ordinary.
Indeed, this huge comment in r1658439 caught my attention, and I thought that it wasn't something we do a lot. Grepping gives more than a hundred hits on usages like this one, so I am fine with that. >> By the way, it looks like this changeset [1] did not really remove the >> XFail() marker from the new test, so I expect the buildbots to turn red. > > Yes, I noticed that and fixed it in r1658442. Thank you for fixing this and the problem itself. Regards, Evgeny Kotkov.
