I just did an SVN checkout of the 2.7 source, ran the following in order:
configure
make
make test

When I run make test it gets to the lines:

test_curses
test_curses skipped -- Use of the `curses' resource not enabled

but I have ncurses installed and enabled.  In fact I was using the
python curses library on my machine when I was running the latest Fedora
YUM install of Python.  

So I am not sure if anyone knows what might be responsible for this message.

You have to pass the -ucurses options to regrtest.py. See regrtest.py -h.

Benjamin,

I ran the regrtest.py file with the -ucurses option and didn't get the error.

However, I don't see how it is possible to send the -ucurses option
through via the make file.

Steve Owens

>
> Benjamin Peterson <benjamin@python.org> added the comment:
>
> You have to pass the -ucurses options to regrtest.py. See regrtest.py -h.
>
> ----------
> nosy: +benjamin.peterson
> resolution:  -> invalid
> status: open -> closed
>
> _______________________________________
> Python tracker <report@bugs.python.org>
> <http://bugs.python.org/issue5356>
> _______________________________________
>

On Mon, Feb 23, 2009 at 8:02 PM, Steve Owens <report@bugs.python.org> wrote:
>
> Steve Owens <steve@integrityintegrators.net> added the comment:
>
> Benjamin,
>
> I ran the regrtest.py file with the -ucurses option and didn't get the error.
>
> However, I don't see how it is possible to send the -ucurses option
> through via the make file.

To enable all the options, you can use "make testall". Otherwise you
can give custom switches to regrtest.py with the TESTOPTS variable.
"make test TESTOPTS=-ucurses"

I know that this issue is not valid, but I believe that it ought to be
searchable since I am sure I am not the last person to ask this question
and I don't think the knowledge is currently documented any where.

On Mon, Feb 23, 2009 at 8:36 PM, Steve Owens <report@bugs.python.org> wrote:
>
> Steve Owens <steve@integrityintegrators.net> added the comment:
>
> I know that this issue is not valid, but I believe that it ought to be
> searchable since I am sure I am not the last person to ask this question
> and I don't think the knowledge is currently documented any where.

I added a note to the README in r69924. I hope that does the trick.