Re: [HACKERS] Floating point comparison inconsistencies of thegeometric types - Mailing list pgsql-hackers

> Backing up a bit here, have we lost track of the problem that we're
> trying to solve?  Tom gave his opinion here:
>
> https://www.postgresql.org/message-id/3895.1464791274@sss.pgh.pa.us
>
> But I don't see that the latest patch I can find does anything to fix
> that.

This is what he wrote:

> As I understand it, the key problem is that tests like "is point on line"
> would basically never succeed except in the most trivial cases, because of
> roundoff error.  That's not very nice, and it might cascade to larger
> problems like object-containment tests failing unexpectedly.  We would
> need to go through all the geometric operations and figure out where that
> kind of gotcha is significant and what we can do about it.  Seems like a
> fair amount of work :-(.  If somebody's willing to do that kind of
> investigation, then sure, but I don't think just blindly removing these
> macros is going to lead to anything good.

I re-implemented "is point on line" operator on my patch so that it
would, at least, work on the most trivial cases.  This is not very
nice, but it is predictable.  I have tried to prevent it cascade to
larger problems like object-containment tests failing unexpectedly.  I
have gone through all of the geometric operations and re-implemented
the ones with similar problems, too.  It is no work at all compared to
discussions we have to have :-).

My initial idea was to keep the fuzzy behaviour for some operators
like "is point on line", but the more I get into it the less value I
see in doing so.  The same family operators like "is point on line" is
currently badly broken:

> postgres=# select '(1,0)'::point ## '{1,2,0}'::line;
> ?column?
> ----------
> (2,2)
> (1 row)

This makes me wonder if anybody is ever using those operators.  In the
end, I don't care about those operators.  They are unlikely to be
supported by indexes.  I can simplify my patch to leave them as they
are.

> Now maybe that's not the problem that Emre is trying to solve,
> but then it is not very clear to me what problem he IS trying to
> solve.  And I think Kyotaro Horiguchi is trying to solve yet another
> problem which is again different.  So IMHO the first task here is to
> agree on a clear statement of what we'd like to fix, and then, given a
> patch, we can judge whether it's fixed.

I listed the problems I am trying to solve in here:

https://www.postgresql.org/message-id/CAE2gYzzNufOZqh4HO3Od8urzamNSeX-OXJxfNkRcU3ex9RD8jw%40mail.gmail.com

> Maybe I'm being dumb here and it's clear to you guys what the issues
> under discussion are.  If so, apologies for that, but the thread has
> gotten too theoretical for me and I can't figure out what the
> top-level concern is any more.  I believe we all agree these macros
> are bad, but there seems to be no agreement that I can discern on what
> would be better or why.

We couldn't agree on how we should treat on floating point numbers.  I
think we should threat them just like the "float" datatype.