[mythtv] development process

Paul Harrison mythtv at dsl.pipex.com
Tue Oct 3 09:56:02 UTC 2006


D. Hugh Redelmeier wrote:
> Last week, I came across a previously observed bug in mythreplex,
> learned how to build Myth, isolated the bug, added the information to
> TRAC, and proposed a fix in the email list.
>
> My suggested fix has been ignored without comment.  Another fix has
> been adopted, so that is OK.  Except that I think that my fix is
> cleaner.
>
> One trouble with my fix is that it was more extensive: I rewrote a
> routine instead of changing the callers to work around its bug.  My
> rewritten routine is simpler than the original.
>
> How should I have proceeded to get my fix adopted?
>
> Without a test suite, it seems very likely that fixes could introduce
> new bugs.  Has any thought been given to adding a test suite?
>
> PS: in my brief experience with Myth source, I've seen a number of
> small things that could be improved.  Not worth destablizing the
> programs, but stylistic improvements that I should have long-erm
> benefits in readability and maintainability.  Is there a consensus of
> "if it ain't broke don't fix it", or, alternatively, "lets polish this
> thing until it shines"?
>   
I think if Michael is playing Miss Marple I must be the villain in all 
this :-)
 
Apologies I didn't mean to ignore your comments, I only saw your patch 
after I committed my fix. Unlike many of the other devs I'm not a 
professional programmer so my aim in fixing the the problem was to fix 
it in as simple a way as possible without breaking anything else in the 
process even more so because I didn't know how the code was supposed to 
be working. If Geoff or anyone wants to apply you patch that's fine with 
me. I'm a "if it ain't broke don't fix it" man by the way ;-)

Paul H.
 



More information about the mythtv-dev mailing list