On Wed, 2005-11-23 at 15:02 -0600, Joshua Varner wrote:
> >
> Do we need something like this:
>
> svn_error_t *
> svn_opt_parse_path_resolved (svn_opt_revision_t *rev,
> const char **truepath,
> const char *path /* UTF-8! */,
> apr_pool_t *pool)
> {
> svn_boolean_t is_special;
> svn_node_kind_t kind;
>
> SVN_ERR (svn_opt_parse_path (rev, truepath, path, pool));
>
> SVN_ERR (svn_io_check_special_path (*truepath, &kind, &is_special, pool));
>
> if ( is_special == TRUE )
> {
> svn_string_t *resolved_path;
>
> // resolve the symlink
> SVN_ERR (svn_io_read_link (&resolved_path, *truepath, pool));
>
> *truepath = svn_path_canonicalize (resolved_path->data, pool);
> }
>
> return SVN_NO_ERROR;
> }
>
> To use with the commands that can't operate on symlinks, give this is
> a temporary workaround, but it can fix the problem at the client level,
> until something more permanent is put together.
>
This is more like a work-around. I already tried this for #2305 and some
of the regression tests failed. My patch for #2305 really solves the issue.
-AT
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org
Received on Thu Nov 24 08:54:59 2005