This is the mail archive of the cygwin-developers@cygwin.com mailing list for the Cygwin project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Re: stat matters


Corinna Vinschen wrote:
> 
> On Tue, May 27, 2003 at 07:48:43PM -0400, Pierre A. Humblet wrote:
> >    The bad news is that one old bug still isn't taken care of: "c:/." is
> >    a root dir without being a prefix of "c:/".
> 
> What are you refering to?  I don't see that it fails:
> 
> $ ls -ldsi C:/
>  327685    0 d---------   10 ???????? ????????        0 May 25 22:56 C:/
> $ ls -ldsi C:/.
>  327685    0 d---------   10 ???????? ????????        0 May 25 22:56 C:/.

That's probably because you are on NT/Win2000/XP, which doesn't use
stat_by_name. Here is the Win98 output (stock 1.3.22)

~: ls -ld c:/
drwxr-xr-x   14 pierre   all          1663 May 19 13:55 c://
~: ls -ld c:/.
ls: c:/.: No such file or directory

> > 3) GetCompressedFileSize does not always report the updated size between the
> >    moment where a write occurs and the moment where the buffers are flushed to
> >    disk. It can underreport the value of st_blocks, but it's not clear if this
> >    violates POSIX:
> >
> >    blkcnt_t  st_blocks  number of blocks allocated for this object
> >
> >    Allocated where? The blocks won't be on disk until flushed.
> 
> Do you suggest to call GetCompressedFileSize() only for files having
> the COMPRESSED or SPARSE_FILE attribute set?

Actually that's a good idea. I don't know if it's really an issue and I didn't 
know what to suggest, short of flushing the file, which wouldn't be that nice.

Pierre


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]