MathGroup Archive 2010

[Date Index] [Thread Index] [Author Index]

Search the Archive

Re: NotebookDirectory[] and UNC paths

  • To: mathgroup at
  • Subject: [mg113508] Re: NotebookDirectory[] and UNC paths
  • From: John Fultz <jfultz at>
  • Date: Sun, 31 Oct 2010 02:10:30 -0500 (EST)

I think you're probably running version 7.0.0.  Version 7.0.1 has this bug
fixed.  You should talk to Wolfram Research Tech Support or Customer Serviced
about acquiring this.


John Fultz
jfultz at
User Interface Group
Wolfram Research, Inc.

On Sat, 30 Oct 2010 04:37:03 -0400 (EDT), Ulf-Dietrich Braumann wrote:
> Hi,
> the following describes a marginal dysfunction of
> NotebookDirectory[] concering a missing leading $PathnameSeparator in UNC
> paths under v7, which was not there in v6:
> Once logged in a Windows 2003 terminal server, one can navigate to
> directories mounted on a local client Windows XP PC. Those folders can be
> accessed using a special UNC (uniform naming convention) path, e.g.
> \\tsclient\V\notebooks (written in Mathematica as string
> "\\\\tsclient\\V\\notebooks").
> Issuing the following from a previously stored notebook residing there,
> under v7 I get this failure:
> In[1]:== ParentDirectory[NotebookDirectory[]]
> During evaluation of In[1]:== ParentDirectory::npdir:
> \tsclient\V\notebooks\ has no parent directory. >>
> Out[1]== ParentDirectory["\\tsclient\\V\\notebooks\\"]
> Due to a wrongly returned UNC path, ParentDirectory[] fails. While using
> v6 it works:
> In[1]:== ParentDirectory[NotebookDirectory[]]
> Out[1]== "\\\\tsclient\\V"
> Obviously, in v7 the command NotebookDirectory[] returns only one
> $PathnameSeparator ("\\") instead of required two. Maybe, this point (I
> expect the problem exists for any UNC path notations but not only for
> terminal clients) can be fixed in the coming v8.
> Regards - Ulf-Dietrich

  • Prev by Date: Re: How to convert Inset graphics to absolute coords?
  • Next by Date: Adding labels to a GraphicsGrid object
  • Previous by thread: NotebookDirectory[] and UNC paths
  • Next by thread: Re: Determining the root of the characteristic equation for time