(no subject)

I downloaded the latest binary version of darcs (1.0.0rc3) and used it 
to get the latest from the darcs darcs repository, which I put in 
~/dev/darcs.
Do developers usually "install" the darcs they are currently working 
on?  It seems more reasonable to install the stable one, and run the 
dev one manually after make to test it out.
I noticed the test suite is a bunch of shell scripts with no exec 
priviledges.  Is there a way to run the whole suite without recording 
new code to the repository (which triggers the test run)?  I'll 
probably just chmod the tests to run them individually as well.   If I 
add a new test, is it picked up by the test suite runner automatically 
or do I have to add it to a list somewhere?
I was thinking of making a wiki page to organize all of these 
acceptance tests.   I'd also like to find a Haskell code coverage tool 
to do some analysis, but a quick Google search didn't turn up much.
BTW, the online manual still says 1.0.0rc2.

Ryan Lowe
ryan@xxxxxxxxxxxx