Re: Policy about not-yet-regression tests?

[Graham Percival]

On Sat, Apr 09, 2011 at 01:40:36AM +0200, David Kastrup wrote:
> Graham Percival <address@hidden> writes:
> 
> >> I am talking about bugs without a fix.  Sometimes they disappear as a
> >> sideeffect of fixing a different bug, or restructuring code.
> >
> > I think the issue tracker is the best place for that.  It's true
> > that sometimes issues get fixed magically, and occasionally we
> > look through the regtest to identify those.
> 
> Uh, what?  If bugs must not be added to the regtest, how can looking
> through the regtest help with identifying bugs that have been fixed?

Sorry, I didn't mean to suggest that bugs must not be added to a
regtest.  Rather, I mean to suggest that it is false that every
patch must be added to a regtest.

For example, look at Mike's recent
b4b6904a4e8c3468a33960de2f9fa96645e54542

Do we seriously need a regtest for this particular bugfix?  I
don't think so.


In an attempt to forestall a long and probably fruitless policy
discussion, I'm going to say this: do whatever seems reasonable.

- if you have a bugfix and want to add a regtest, go ahead.
- if you have a bugfix, but no regtest, and somebody thinks that
  the bugfix *should* have a regtest, they'll let you know.
- if you have a bug, but no bugfix, then don't add a regtest.
  Add it to the tracker instead.
- when somebody fixes that bug, if they don't add a regtest,
  you can let them know.  If you're particularly concerned
  that you might not be paying attention when their 48-hour
  countdown is happening, then by all means add "a bugfix for
  this should include a regtest which is the following code"
  to the issue.

Cheers,
- Graham