Gentlemen, a call to arms!

chromatic chromatic at wgz.org
Thu Oct 12 02:14:45 BST 2006


On Wednesday 11 October 2006 17:57, Matt S Trout wrote:

> As I said, this isn't about "right" at all.
>
> This is about not screwing people's installability without warning.

Okay then.  Warning: I may fix bugs in my code.  If you have bugs in your 
code, my bugfixes may expose your bugs.

> I've spent weeks making sure Catalyst installs cleanly everywhere I can
> get access to. I intend to spend more weeks doing so.
>
> If you don't care about that, please feel free to stick to development
> and I'll happily volunteer to take over the release management so the
> people trying to make a living off perl development and deployment
> aren't disadvantaged, because I'm one of them and I have better things
> to do than rewrite tests because a bugfix blows away the installability
> of perfectly functional modules.

How can I refuse such a sweet request, with all of its "Hey, thank you for 
writing this useful code and documenting this useful code and fixing bugs in 
it and making it available for free -- we use it a lot and it makes our lives 
much easier" and a wonderful, clear bug reporting in the appropriate place 
with a minimal test case?

To be perfectly honest, I can't think of a way to use 
Test::MockObject::Extends *sanely* to trip this error without introducing 
serious bugs in a test suite.  If you can provide me a minimal test case 
(please to the RT queue), I'll work out a different approach.

If you want to bundle an old version of Test::MockObject, knock yourself out.  
I hear Module::Install is good at that.

If you want to fork or write your own, you have that right.

I just can't even muster up a *yawn* for the "You crazy amateur, your reckless 
regard for fixing bugs instead of maintaining complete and unchanging 
stability broke the work of Real Serious Developers who can't be bothered to 
file a proper bug report about it, instead complaining on a completely 
unrelated mailing list" approach, so please do let me know if you choose one 
of the other approaches that deserves my attention.

-- c


More information about the london.pm mailing list