Re: [postgis-users] Re: Comparing Geometries with Different SRIDs

Charlie Savage Sun, 28 Sep 2008 22:26:38 -0700


ST_Equals(different srids) should return an error.  Its not that its unknown
its that we are too lazy to calculate it.  Its quite known if we had the
energy.


No - that is not true.  You cannot calculate it in some cases.

We should go with binary equal rather than geometric equality.  I think that
would satisfy Charlie too.  I guess I have my parents to blame too :)


I just want to be able to write this code:

IF (GEOM1 = GEOM2) THEN

And not have it blow up...

And related doing a UNION like described in an earlier email.

Last - I'm fine with what Paul proposes. That is that the = operator does not blow up, while ST_EQUALS does blow up. Having 2 versions of a function, one that raises an exception and one that does not, is not that uncommon...Alternatively, ST_EQUALS could take a 2nd parameter that specifies whether to blow up or not, with a default value of blow up....


Charlie


Thanks,
Regina

-----Original Message-----
From: [EMAIL PROTECTED]
[mailto:[EMAIL PROTECTED] On Behalf Of Brent
Wood
Sent: Sunday, September 28, 2008 6:40 PM
To: PostGIS Users Discussion
Subject: RE: [postgis-users] Re: Comparing Geometries with Different SRIDs

After a belated case of RTFM...

It seems UNION does an implict distinct on all rows, unless UNION ALL is
applied (How many Regina's do you want :-)

I guess it makes sense to drop duplicates unless otherwise specified.


stations=# select 1 union select 1;
 ?column?
----------
        1
(1 row)

stations=# select 1 union all select 1;
 ?column?
----------
        1
        1
(2 rows)

stations=# select 'Regina' union select 'REGINA';  ?column?
----------
 Regina
 REGINA
(2 rows)

stations=# select 'Regina' union select 'REGINA' union all select 'Regina';
?column?
----------
 Regina
 REGINA
 Regina
(3 rows)


Cheers

   Brent Wood




--- On Mon, 9/29/08, Paragon Corporation <[EMAIL PROTECTED]> wrote:

From: Paragon Corporation <[EMAIL PROTECTED]>

Subject: RE: [postgis-users] Re: Comparing Geometries with Different SRIDs To: "'PostGIS Users Discussion'" <[email protected]> Date: Monday, September 29, 2008, 11:13 AM Brent, Ah never mind. For some reason I was under the false assumption that in SQL Server with dictionary sort order that even though


REGINA = Regina

That I would get 2 Reginas when I union them. That is not the case. I just get one.

I'm not sure why I thought that or maybe it was like that a long time ago.

Anyrate - so I guess I am back to square one. By the way - I think we should be talking about ST_OrderingEquals not ST_Equals.

I'm not sure which if any of those PostGIS is using to do the check, but it should be ST_OrderingEquals since using ST_Equals introduces a whole lot more philosophical can of worms. Is space empty or is Empty space.


Thanks,
Regina

-----Original Message-----
From: Brent Wood [mailto:[EMAIL PROTECTED]
Sent: Sunday, September 28, 2008 3:19 PM
To: [EMAIL PROTECTED]

Subject: RE: [postgis-users] Re: Comparing Geometries with Different SRIDs


Hi Regina,

I think that is more an encoding/language issue than case in such an analogy.

To wax metaphysical, does a muslim worship a different god because they spell it allah? (Similarly Jews with Jehovah & French with dieu...)

So is st_equals() a cross-language thesaurus, or is it restricted to the specified SRID? :-)





Spotcha,

   Brent Wood


--- On Mon, 9/29/08, Paragon Corporation <[EMAIL PROTECTED]> wrote:

From: Paragon Corporation <[EMAIL PROTECTED]>
Subject: RE: [postgis-users] Re: Comparing Geometries

with Different

SRIDs
To: "'PostGIS Users Discussion'" <[email protected]>
Date: Monday, September 29, 2008, 8:04 AM Charlie,

On second thought not sure how DB II deals with

relational Union of

geometries.  Does anybody have that by chance to test?

I suppose we could say this is the same deal as

difference in casing.

SELECT 'Regina'
UNION
SELECT 'REGINA'
UNION
SELECT 'Regina'

Even if both geometries are the same which SRID would

you choose?  So

we should not be doing a geometry compare, but a

binary compare.

SELECT ST_AsEWKB('SRID=4269;POINT(1 3)') =
ST_AsEWKB('SRID=4326;POINT(1 3)')


Thanks,
Regina

-----Original Message-----
From: [EMAIL PROTECTED]
[mailto:[EMAIL PROTECTED]

On Behalf Of

Paragon Corporation
Sent: Sunday, September 28, 2008 2:48 PM
To: 'PostGIS Users Discussion'
Subject: RE: [postgis-users] Re: Comparing Geometries

with Different

SRIDs

I feel your pain Charlie, but I'm on Reid's

side.

Given that some spatial
databases do support cross compare between geometries

of different

SRID (e.g. DB II), not sure about Oracle or SQL Server

2008, I would

prefer an answer that is in compliance or throws an

exception when it

can't be sure.
Its annoying from a debugging perspective  to deal

with things that

fail silently.  Returning false is a fail silently in

my book.

If we deal with this as a special case - why

wouldn't we deal with

everything else the same with every other relational

compare we do?

Its just wrong, but granted doing the wrong thing

would be

conveniently useful in many cases.

Thanks,
Regina

-----Original Message-----
From: [EMAIL PROTECTED]
[mailto:[EMAIL PROTECTED]

On Behalf Of

Charlie Savage
Sent: Sunday, September 28, 2008 2:10 PM
To: PostGIS Users Discussion
Subject: Re: [postgis-users] Re: Comparing Geometries

with Different

SRIDs



Reid Priedhorsky wrote:

Charlie Savage wrote:

The 2 point could be spatially equal

given

different SRIDS and

coordinates if they were projected to a

common

SRID. So should

geometry operators silently Call

st_transform

to make the righthand

match the lefthand before comparing? This

would be quite the

expensive operation.

No. You can't automatically transform

between

different SRID values

Think of the case of one geometry with an

SRID

value of 4326 and one

with an SRID value of -1 (no coordinate

system).

Exactly.

So different SRID values, then the geometries

are

not equal.

No -- as Stanley said, the geometries could be in

fact

equal, but

expressed in different SRS. So if ST_Equals()

returned

False, it would

be wrong.

Maybe.  But returning "Operation permitted"

is even worse because it

means you can't do natural things like this

(without extra annoying

SRID checking
code) in plpgsql:

IF (geom1 == geom2) ... END IF;

Or the example with the union earlier posted

It is up to the user to transform geometries

to

the same SRID before

calling ST_EQUALS.

Exactly. ;)

So, I still vote that st_equals will not blow up when

comparing two

geometries with different SRID values, it just will

return false.

Charlie


_______________________________________________
postgis-users mailing list
[email protected]

http://postgis.refractions.net/mailman/listinfo/postgis-users



_______________________________________________
postgis-users mailing list
[email protected]

http://postgis.refractions.net/mailman/listinfo/postgis-users



_______________________________________________
postgis-users mailing list
[email protected]
http://postgis.refractions.net/mailman/listinfo/postgis-users

_______________________________________________
postgis-users mailing list
[email protected]
http://postgis.refractions.net/mailman/listinfo/postgis-users


_______________________________________________
postgis-users mailing list
[email protected]
http://postgis.refractions.net/mailman/listinfo/postgis-users


--
Charlie Savage
http://cfis.savagexi.com

smime.p7s
Description: S/MIME Cryptographic Signature

_______________________________________________
postgis-users mailing list
[email protected]
http://postgis.refractions.net/mailman/listinfo/postgis-users

Re: [postgis-users] Re: Comparing Geometries with Different SRIDs

Reply via email to