Of course it's implemented. Please Revert. If you want to add a note at the code where it ought to be implemented, please do, not in the header...
The whole point is that _any_ platform can choose not to implement a given field. Win32 and OS2 already populate that field, and Unix will if you are searching a directory. OS2 doesn't have inodes. Rather than 'invent' a random number, time/path signature hash or other bogus entry, it simply never sets the APR_FINFO_INODE bit before it returns. If you asked for that bit, you get APR_INCOMPLETE. It's part of the spec. If the spec isn't clear, feel free to document this more completely :-) Bill ----- Original Message ----- From: <[EMAIL PROTECTED]> To: <[EMAIL PROTECTED]> Sent: Wednesday, May 23, 2001 10:08 AM Subject: cvs commit: apr/include apr_file_info.h > sussman 01/05/23 08:08:49 > > Modified: include apr_file_info.h > Log: > (apr_finfo_t): note that this field doesn't work. > > Revision Changes Path > 1.17 +1 -1 apr/include/apr_file_info.h > > Index: apr_file_info.h > =================================================================== > RCS file: /home/cvs/apr/include/apr_file_info.h,v > retrieving revision 1.16 > retrieving revision 1.17 > diff -u -r1.16 -r1.17 > --- apr_file_info.h 2001/04/10 19:22:13 1.16 > +++ apr_file_info.h 2001/05/23 15:08:44 1.17 > @@ -189,7 +189,7 @@ > apr_time_t mtime; > /** The time the file was last changed */ > apr_time_t ctime; > - /** The full pathname of the file */ > + /** The full pathname of the file. NOT YET IMPLEMENTED. */ > const char *fname; > /** The file's name (no path) in filesystem case */ > const char *name; > > > >
