Woops.  Typos.  Should have been:

class BurndownTests < Test::Unit::TestCase

  @@db = nil
  @@done_with_db = false

  def setup
    @@db = DBI.connect(.......) if not @@db
  end

  def teardown
    @@db.disconnect if @@done_with_db
  end

  def test_zzzzzz_should_be_the_last_test_run_due_to_zzzzzz
    @@done_with_db = true
  end

  # continue on with tests that use @@db ....

Jeff

On 9/5/07, Jeff - Burly Systems <jeff.burly / gmail.com> wrote:
> Somewhat lame, but this might do what you're looking for:
>
> class BurndownTests < Test::Unit::TestCase
>
>   @@db = nil
>   @@done_with_db = false
>
>   def setup
>     db =DBI.connect(.......) if not @@db
>   end
>
>   def teardown
>     db.disconnect if @@done_with_db
>   end
>
>   def test_zzzzzz_should_be_the_last_test_run_due_to_zzzzzz
>     @@done_with_db = true
>   end
>
>   # continue on with tests that use @@db ....
>
> Jeff
>
> On 8/31/07, Max Russell <thedossone / gmail.com> wrote:
> > I am trying to take a test-driven approach to a script I am writing.
> >
> > The script needs to connect to SQLServer using DBI.
> > Currently it looks like this:
> >
> >
> > class BurndownTests < Test::Unit::TestCase
> >
> >   def setup
> >     db = DBI.connect("DBI:ODBC:driver={SQL
> > Server};Server=INPSESVER1;Database=RMT;Trusted_Connection=yes;")
> >   end
> >
> >   def teardown
> >     db.disconnect
> >   end
> >
> >   def test_connectrmtrack
> >     pass
> >   end
> >
> >   def test_export_schedule
> >     pass
> >   end
> >
> >   def test_format_datecols
> >     pass
> >   end
> >
> >   def test_count_resolutions
> >     pass
> >   end
> >
> >   def test_dumptofile
> >     pass
> >   end
> >
> > end
> >
> > The issue with this is that using setup like this, from what I
> > understand is used for each new test method I add. As I want to connect,
> > test the connection and then run tests which have a reliance on the
> > connection being established, how can I do a onetime setup and then test
> > this?
> > --
> > Posted via http://www.ruby-forum.com/.
> >
> >
>
>