In the months leading up to my taking the plunge full force into OSX world, I read alot of stories by people who had "Switched" before Apple made that a marketing ploy. Invariably people would complain about the case-insensitive nature of HFS+. Flame wars would ensue, insults thrown, international incidents nearly caused.
Finally, one guy, a former BeOS guru who had switched, wrote something about "Who cares, how many times do you ACTUALLY need two files in the same directory, one named foo.doc and the other Foo.doc?"
I'll tell you when. In the PATH.
I am currently installing mod_perl on my iBook just to play. The config kept dying, and I couldn't grasp why. And then, it hit me.
The HEAD script that comes with LWP was earlier in my path than the stock /usr/bin/head. The config calls "head -1 somefile", and HFS+ helpfully served up LWP's HEAD, which dies complaining that -1 is an invalid switch.
Go figure.
There are entirely legitimate reasons to have case distinguish. For example, at the day job, we have files in the document tree that are accessible only internally or by the web server. These are all in directories that start with capital letters. Apache knows to reject requests for any file or directory matching/^[A-Z]. I suspect that wouldn't work under HFS, either.
Worst of all, Windows doesn't differentiate. Who wants to be like that?
Re:Case-insensitivity
ct on 2002-07-23T16:53:46
I seem to remember something in an apache changelog a while back specifically aimed at OSX/Darwin.Seems that they had to change the logic in the auth bits, because people were limiting access to Locations like
/foo, but apache was ignoring it for requests to /Foo or even /FoO. Apache on windows has logic to handle that, but Darwin, being Unix, apparently was ignoring that.
Re:HEAD
ct on 2002-07-24T23:45:48
I certainly will answer no in the future. It just never occurs to you. I haven't done anything of the sort on a case insensitive fs in so long I never dreamed it would happen.
