Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!samsung!uakari.primate.wisc.edu!sdd.hp.com!usc!snorkelwacker.mit.edu!bloom-beacon!eru!hagbard!sunic!mcsun!cernvax!chx400!ugun2b!ugsc2a!fisher
From: fisher@sc2a.unige.ch
Newsgroups: comp.os.msdos.misc
Subject: Re: xcopy /s fails
Message-ID: <1990Nov22.164420.313@sc2a.unige.ch>
Date: 22 Nov 90 14:44:20 GMT
References: <90324.035212TOMIII@MTUS5.BITNET>
Organization: University of Geneva, Switzerland
Lines: 24

In article <90324.035212TOMIII@MTUS5.BITNET>, TOMIII@MTUS5.BITNET (Thomas Dwyer III) writes:
> Suppose you have the following directory tree:
> 
>    \---+---DIR1---+---DIR2
>        |          |
>        |          +---DIR3
>        |
>        +---DIR4
> 
> 
> 
> DIR2 and DIR3 have their R/O bits set.  Why does XCOPY fail to recurse with
> the command line "XCOPY \DIR1 \DIR4 /S" ?

You're right, XCOPY seems to have a wrong file attribute pattern when looking
for sub-directories.  (Probably looks for "attrib == dir_attrib" instead of
"attrib & dir_attrib != 0", which means that subdirectories must have exactly
the "normal" attribute pattern...)

But I'm curious about your file system:  What's the use of setting the
read-only flag of a directory under MS-DOS?  The flag isn't even enforced
at DOS level, so who cares?

Markus Fischer, Dpt. of Anthropology, Geneva CH.