Google Summer of Code - Case Insensitive Filenames

[L. Rahyen]

On Tuesday March 25 2008 07:26:50 Kai Blin wrote:
> On Tuesday 25 March 2008 01:45:11 Cesar Izurieta wrote:
> > >  Perhaps this shouldn't be an extra tool but more like a setting what
> > > to do in a case like that.
> >
> > That sounds good. Maybe doing something like what windows does to long
> > file names when looking those files in a console, for example
> > "C:\Program Files" gets converted to "C:\Progra~1", we could apply the
> > same logic, have the first file "A.txt" and the second "a~1.txt". What
> > do you think?
>
> Well, that's probably an option. But I wouldn't even mind a mode that
> simply deletes one of A.txt or a.txt on mounting.

	Deleting files like that (just because of case insensitive match) on mounting 
is very, very bad idea. And I don't see any reason for such destructive 
actions. Better idea to pick up one and use it (for example, choice may be 
either pseudo-random or by alphabetical order where big letters comes first - 
latter is more preferable in my opinion because it let easily guess what 
file/directory will be used).
	However it is understandable and intuitive that if I have *already* mounted 
case-insensitive file system then I cannot create both A.txt and a.txt there. 
But this is safe because I can get a warning that I'm going to overwrite 
existing file.