Re: PSPP-BUG: match_nodes() in covariance-matrix.c

[Jason Stover]

On Sat, Jan 17, 2009 at 11:21:01PM -0800, Ben Pfaff wrote:
> Hi Jason.  It looks to me that some of the logic in match_nodes()
> in covariance-matrix.c is reversed.  In particular,
> compare_values_short() returns nonzero if values differ, which it
> seems to me should make match_nodes() return nonzero also, but in
> fact it makes match_nodes() return 0.  So I think that each
> "compare_values_short(...)" should really be
> "!compare_values_short(...)".
> 
> Is my analysis correct?  (I don't understand anything above the
> level of this single function, so I could be way off.)

You're right. I just committed a fix.

-Jason