Mail Archives: cygwin/2003/01/22/12:35:00
On Wed, 22 Jan 2003, Randall R Schulz wrote:
> At 08:44 2003-01-22, Igor Pechtchanski wrote:
> >On Wed, 22 Jan 2003, Gael Mulat wrote:
> >
> > > OK. If I summurize all that has been said, this problem is
> > > well-known but has no solution in a near future.
> > >
> > > But a workaround would be very useful for people (me and Brian
> > > Kelly, for instance) who use plenty of rm -rf in cross-platform shell
> > > scripts.
> > >
> > > Thanks to Shankar Unni, I have mine: I didn't noticed that only rm
> > > -rf had the trouble, and not rm -r.
> > >
> > > So my workaround will be to replace all the '/bin/rm -rf dir' by
> > > 'chmod -R +w dir; /bin/rm -r dir'. The semantic is not exactly the same
> > > (especially on write-protected directories), but that will allow my
> > > scripts to work well...
> > >
> > > Thanks everybody.
> > > Gael.
> >
> >A suggestion:
> >
> >'/bin/find dir -depth -exec /bin/chmod +w {} \; -exec /bin/rm -r {} \;'
> >might work, and will only parse the tree once... It will invoke chmod and
> >rm multiple times, though, but that's a tradeoff.
> > Igor
>
>
> Igor,
>
> I think we've drifted off the problem here. This isn't a issue of file
> modes interfering with file removal, its an interaction with window's
> "locking" of files that are in-use (open).
>
> And as the discussion from last April showed, not all in-use states
> interfered with removal via Cygwin unlink(). While we were having
> trouble getting my original problem reproduced, I decided to write a
> Java program that opened a file and kept it open. Something about the
> way Java opened files triggered the problem where not all other uses did.
>
> Perhaps it was just a Cygwin (non-interfering) vs. non-Cygwin
> (interfering) use of the file. I don't know. I don't know what was the
> exact diagnosis of the problem at the level of the Cygwin unlink code.
>
> Chris January might know.
>
> Randall Schulz
Randall,
Yes, I realize that. I wasn't suggesting a fix for the problem, I was
suggesting a workaround that would let people do what they want without
having to traverse (apparently huge) trees twice. I agree that the
problem exists and needs to be fixed (at some point, by someone who
understands what's going on).
Igor
--
http://cs.nyu.edu/~pechtcha/
|\ _,,,---,,_ pechtcha AT cs DOT nyu DOT edu
ZZZzz /,`.-'`' -. ;-;;,_ igor AT watson DOT ibm DOT com
|,4- ) )-,_. ,\ ( `'-' Igor Pechtchanski
'---''(_/--' `-'\_) fL a.k.a JaguaR-R-R-r-r-r-.-.-. Meow!
Oh, boy, virtual memory! Now I'm gonna make myself a really *big* RAMdisk!
-- /usr/games/fortune
--
Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
Bug reporting: http://cygwin.com/bugs.html
Documentation: http://cygwin.com/docs.html
FAQ: http://cygwin.com/faq/
- Raw text -