Five build fixes a day

[Barry Warsaw]

On Sep 13, 2011, at 07:10 AM, Martin Pitt wrote:

>Fixing an FTBFS is not all that simple:

So true.  One of the ones I worked on yesterday took way longer than I
expected.  The ftbfs was caused by test failures due to upstream
incompatibility with Python 2.7 on amd64.  I eventually decided to punt by
disabling the tests in that case, though I didn't forward it to Debian because
the patch was pretty gross.  Since the upstream looks inactive, the
alternative was to spend a lot more time drilling down into the package to
essentially do the upstream's work.  It just didn't seem like a good use of my
time, tbh, especially because I have no idea how important that package is to
our users.

I'm not sure what the moral of the story is, except that one difficult package
can kill your hopes of fixing five ftbfs per day.

-Barry