On Wed, Oct 06, 2004 at 12:00:15PM -0400, Pierre A. Humblet wrote: > >Christopher Faylor wrote: >> >> On Wed, Oct 06, 2004 at 11:30:17AM -0400, Pierre A. Humblet wrote: >> > >> >Christopher Faylor wrote: >> >> >> >> On Wed, Oct 06, 2004 at 03:12:45PM +0200, Bas van Gompel wrote: >> >> >Another (hopefully trivial) patch, to help in trouble-shooting. >> >> >> >> Wasn't there another problem where "foo\/bar" type of entries were >> >> showing up? Could you add a check for that, too? >> > >> >I while ago I have modified Cygwin to accept this kind of syntax. >> >Is there a remaining problem in the current release? >> >Otherwise I don't see the need to alarm the user. >> >> It's just a warning. This really shouldn't be in the mount table >> and it really should be corrected. > >I don't think it's checking the mount table,
Ok, sigh, yes, I "mispoke". s/mount table/registry/. >it's checking the registry. The entry will be cleaned up by the time >it gets to the mount table. What would be useful is a check that >::add_item will accept the registry entry, i.e. won't return EINVAL or >perhaps "path too long". The relevant part of add_item is pasted >below. It shows when EINVAL is returned. cygcheck is not a cygwin program. It has it's own implementation of getmntent which reads the registry directly, and, it displays the \/ paths when you do a 'cygcheck -s'. It can check for these types of problems and it should warn about them. cgf