From: Matthew Boeh <mboeh@...>
Date: 2008-07-19T10:04:37+09:00
Subject: [ruby-core:17849] Re: FAILURE of  "expand_path"


--32u276st3Jlj2kUU
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
Content-Transfer-Encoding: quoted-printable

On Sat, Jul 19, 2008 at 08:58:42AM +0900, C.E. Thornton wrote:
> Matthew,
>=20
>   Well something is inconsistent  somewhere.
>=20
>   However, the reason this came up was I
> wrote a class to perform a tree scan.  It recursively
> stepped through to directory tree.  It failed when
> it ran across a directory that had a number of directories
> prefixed with a  tilde '~'.=20
>=20
> When you have a dir containing this
>  dir1:
>    ~dira  ~dirb  ~dirc
>=20
>  Then you list the 'files' in a dir and
> then recursively call expand_path on each
> dir you find and agin list the files--
> this failure stops the scan.

Why are you calling File.expand_path on these? The two uses of expand_path =
I'm=20
familiar with are to expand ~ and to resolve .. and ., neither of which see=
m=20
to apply in this case.

> I want to make it consistent with the way the
> Bash Shell works.
>=20
> mkdir ~dir1   produces "./~dir" NOT "/home/dir"

Only if there's no user named ~dir1.=20

> Am I wrong here?

No, but your patch removes the ~user -> /home/user behavior. A better fix t=
o=20
match the Bourne shell behavior would be to get rid of the "user %s doesn't=
=20
exist" exception entirely and leave the string unchanged.

--=20
Matthew Boeh

--32u276st3Jlj2kUU
Content-Type: application/pgp-signature; name="signature.asc"
Content-Description: Digital signature
Content-Disposition: inline

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.1 (GNU/Linux)

iD8DBQFIgT7Yfi+i1R7d/ZwRAnRnAJ9bsdHBWbIE6y7CrX27kwPV460J6QCdFjHC
LzpzZSVwq+058sKWI1gKDBc=
=A3ZS
-----END PGP SIGNATURE-----

--32u276st3Jlj2kUU--