On Jan 21, 2008, at 6:00 PM, Theodore Tso wrote:
On Mon, Jan 21, 2008 at 05:46:27PM -0500, Kevin Ballard wrote:I find it amusing that you keep arguing against having git treat filenames as unicode when, if you had actually taken my advice and read my previousemail talking about "ideal" vs "practical"...If by "ideal" you mean a world where 100% of all computers were designed by Steve Jobs, you might have a point.
NO NO NO NO NO. READ MY EMAIL. STOP MAKING ASSUMPTIONS ABOUT WHAT I'M TALKING ABOUT.
The most frustrating thing about this thread is everybody keeps arguing about what they *assume* I'm talking about without actually bothering to read what I'm saying.
In other words, I was trying to illustrate thatHFS+ isn't wrong, it's just different, and the difference is causing theproblem.And if you want to interoperate with the rest of the world, where at least count over 92% of computers are NOT running HFS+, then "ThinkingDifferent" is indeed causing the problem, yes. And whose fault is that?
And if you want to interoperate with the rest of the world, where at least count over 92% of computers are running Windows, then using another OS is stupid, right? Right? I mean, if everyone else is doing it, we should too, shouldn't we?
The whole point of interoperability is that when we communicate, we have to do so in a uniform and predictable way. If we can't, the next best thing is to have protocol translators; but in order to do that, we must avoid lossy transformations, such as HFS+'s pseudo-normalization. (Why, by the way, will not result in a "normal" form for any glyph which can be encoded with and without a combining character if said glyph was introduced into Unicode after 1988. So you can't even call it a "normalization" algorithm, but just a pseudo-normalization transformation which is lossy and which DESTROYS filename information in an irrecoverable way.)
Sure it's normalization, it's just not using one of the standard forms. But the form is well-defined.
And yes, protocol translators are a good idea. That's why I thought the original suggestion of using a table to map index filenames <-> HFS + filenames sounded like it could work. The only time that should fail is if the index contains multiple filenames that HFS+ will treat as a single filename. Is there a problem with this approach?
-Kevin Ballard -- Kevin Ballard http://kevin.sb.org kevin@xxxxxx http://www.tildesoft.com
<<attachment: smime.p7s>>
