Re: backwards compatibility problem with .cvspass?

[Karl Fogel]

IMHO, if it's very hard to write a test for it, then the best thing to
do is check it in anyway, and somewhere list the absence of the test
case as a bug in itself (if you want).  Just because something's hard
to test automatically hopefully shouldn't mean it can't go into the
code. :-)

Thanks for the update, by the way, was wondering what was up.

-K

"Derek R. Price" <derek.price@openavenue.com> writes:
> > Okay.  I see your point.  Sorry.  I think I even have most of the code lying
> > around since that was how I designed it in the first place, so the change 
> > should
> > be in soon.  It takes a little longer to run, but there aren't many .cvspass
> > lookups per session, so it shouldn't be a big deal.
> 
> Just a quick note since it's been just over a month now since I said I'd fix 
> this.
> I've been done with the actual fix for about three weeks, but this feature is
> rather troublesome to write a sanity.sh test for.  I've been alternately 
> looking
> into using Automake's testing features to write a test external to sanity.sh 
> and
> getting into/trying to follow some of the current Automake and Autoconf issues
> because I wasn't sure things were working quite right.  :)  I can check in 
> the fix
> without tests if anybody wants it now.  Otherwise it'll probably still be 
> another
> two weeks.
> 
> Derek
> 
> --
> Derek Price                      CVS Solutions Architect ( http://CVSHome.org 
> )
> mailto:dprice@openavenue.com     OpenAvenue ( http://OpenAvenue.com )
> --
> 115. (A)bort, (R)etry, (I)nfluence with large hammer