The plan for eliminating lp_posix_pathnames()

[Uri Simchoni]

I for one could benefit from a review of the posix vs NT path names 
issue and how it affects smbd and VFS modules.

Uri.

On 02/12/2016 02:45 AM, Jeremy Allison wrote:
> Just in case you wondered what I was up to with
> this :-).
>
> Currently my plan is to move to using
> const struct smb_filename * instead of const char *
> on more of the VFS interfaces (such as
> get_nt_acl_fn(), mkdir, rmdir, rename etc.)
>
> Once this is plumbed through the VFS I
> think the best idea is to add a bool posix_pathnames
> to the struct smb_filename, which gets set
> from the toplevel parse of the incoming
> client SMB1/SMB2 request.
>
> At that point all the functions on the
> other side of the VFS that do lp_posix_pathnames()
> can be changed to look into smb_fname->posix_pathname
> instead !
>
> The nice thing is I can get the patches that
> make these VFS changes reviewed slowly, one
> at a time without having to do a flag day/big
> change.
>
> I'm currently in the process of doing this
> to SMB_VFS_GET_NT_ACL() as a test case (and
> it seems to be going quite well).
>
> Sounds like a plan ?
>
> Jeremy.