Johannes Sixt schrieb: > René Scharfe schrieb: >> Johannes Sixt schrieb: >>> -GIT_CEILING_DIRECTORIES="foo:$TRASH_ROOT/sub" >>> +GIT_CEILING_DIRECTORIES="/foo:$TRASH_ROOT/sub" >>> test_fail second_of_two >>> >>> GIT_CEILING_DIRECTORIES="$TRASH_ROOT/sub:/bar" >>> test_fail first_of_two >>> >>> -GIT_CEILING_DIRECTORIES="foo:$TRASH_ROOT/sub:bar" >>> +GIT_CEILING_DIRECTORIES="/foo:$TRASH_ROOT/sub:/bar" >>> test_fail second_of_three >> >> I don't, which makes me uneasy -- the tests shouldn't depend on details >> in our setup. :-/ > > I updated my msysgit to the current master this moment, and I still need > these two changes. Do you use an older msysgit? With 31d5dfeb0 > (2008-06-08) Steffen changed path mangling so that a text without a > leading / or . is not converted. That's exactly what we see here. Oh, yes, good idea. I never updated the msys part of my installation. My "uneasiness" doesn't mean that I'm against your changes above, in any case. >>> That said, I'm in the process of preparing a series that includes your >>> patch and that does the proper cleanup and code moving that you again >>> didn't do :-/ But it turns out that this is non-trivial because of >>> bash's >>> (MSYS's) I-know-better-what-is-a-path-and-what-not behavior. It will >>> take >>> some time... > Johannes Sixt (4): > Make test-path-utils more robust against incorrect use > Move sanitary_path_copy to path.c (and rename to > normalize_path_copy) > Test and fix normalize_path_copy() > Remove unused normalize_absolute_path() Yeah, OK. I was aiming for a quick short fix, deferring cleanups to when I have more time. I wonder what patches one and three do, though. René -- To unsubscribe from this list: send the line "unsubscribe git" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
