[RFC] depend on testtools for testing?

[Ian Clatworthy]

Martin Pool wrote:
> 2009/10/29 Robert Collins <robertc at robertcollins.net>:
>> Are there folk who object to adding a hard dependency on testtools for
>> 'bzr selftest', given the above? Please speak up - I don't want to do
>> anything that makes developing or using bzr harder [but I also want to
>> reduce duplicated code :)]
> 
> As I said to Robert yesterday, to me, given the technical environment,
> having the code factored out is approximately in balance with the
> hassle a new dependency creates.
> 
> (I feel like this is going to be bait for someone to argue that it is
> technically possible to have an external dependency without problems.
> Sure.  But given the state of the art for Python, and for cross-OS
> packaging, it does in practice cause problems.

I'll bite. :-)

Integrated is easier than external dependencies. OTOH, both have
problems and it comes down to a question of which set of problems do you
want to be focussing energy on.

Strategically, I'd prefer to see the bzr core getting smaller and for us
to rely on carefully selected, specialist tools and libraries. I'd
rather be using, for example, Sphinx for docs and some actively
maintained 3rd party tool for testing than maintaining our own smart doc
tools, test libraries, etc. There's initial pain but the payoff is good
IMO if you choose wisely.

Tactically though, timing and ease-of-development is everything. For
example, I'd like assurances than the developers making this change
would test and document the setup for developers on Windows as part of
the switch. We need more people running the test suite there, not less.

Ian C.