Re: Automake 1.4l released

[Charles Wilson]

Tim Van Holder wrote:
> 
> > But I'm not talking about FAT or VFAT.  I'm talking about NTFS, a whole
> > different beast.  IMNSHO, the linux and dos behaviors you describe ARE
> > the mistake, and cygwin's behavior (on NTFS) is correct -- why should
> > you be allowed to change the file metadata (timestamp, etc) if you don't
> > have write access to the file?
> 
> Don't know - but since the majority of systems allow it, I assume there
> is a good reason.  For example, it makes perfect sense to change, say, the
> archive bit for a read-only file on DOS (to mark it for backup).
> Similarly, I see nothing wrong with chown'ing or chmod'ing a read-only
> file.  So why would the timestamp be handled differently from other file
> metadata?  In all the cases, the file itself is not written to, which
> satisfies the read-only constraint implied by the file's attributes.
> So IMHO, cygwin should recognize the "unusual" behaviour of NTFS, and
> perhaps internally do 'chmod +w; touch; chmod -w' when changing the
> timestamp of a read-only file.

Omygod.  You have NO idea how much overhead the necessary checks would
add -- it would slow down file access on cygwin to a crawl.  I still
have not heard a cogent explanation about WHY "chmod -R go-w" is
undesirable -- for any filesystem.  Why is 'chmod -R a-w' used in the
distdir: target?  Is it just "that's the way it's always been done?"

What am I missing?

--Chuck