Testing reports

[Derek Atkins]

"Colin Scott" <gnucash at double-bars.net> writes:

>> Another thing to keep in mind, Colin, is that the developer working 
>> on the report might not be the one who breaks it
>
> True, but so what?  *Whoever* broke it, it needs to be fixed.
> *Somebody* needs to find out why the report is broken, and until they
> do nobody will know who broke it (and thus who can fix it!)

True, but see below.  Sometimes the person who notices the breakage is
not the person who broke it, or the one who would even know how to fix
it.

>> Even worse, but most importantly, the test failure might not be
>> noticed right away
>
> Why on earth not?  There seems little point in running a test if you don't bother looking at its output!

Because honestly not every developer runs "make check" before every
commit.  For better or worse, that's the way it is.  You can rant and
rave about how wrong it is until you are blue in the face, but it wont
change the fact that it is what it is.

As a real-world example, I personally do run "make check" whenever I
build GnuCash locally.  Several months ago I noticed that it was broken.
I hadn't made any changes at all, someone had checked in code that broke
is sometime since the last time I pulled and built.  So, who is supposed
to fix that?  I just sent mail to gnucash-devel about it and it did
eventually get fixed, but I had nothing to do with the breakage.  Am I
expected to fix it only because I noticed it first?

> Colin

-derek

-- 
       Derek Atkins, SB '93 MIT EE, SM '95 MIT Media Laboratory
       Member, MIT Student Information Processing Board  (SIPB)
       URL: http://web.mit.edu/warlord/    PP-ASEL-IA     N1NWH
       warlord at MIT.EDU                        PGP key available