I have been poking at an over optimization issue in segment stabbing.
In this case
If a Multipolygon has two outer rings and the left outer ring is inside
the ymin, ymax envelope of the right outer ring the left ring is dropped.
JTS 1.7 does not exhibit this behavior.
I have been making progress where the over optimzation is not.
I'll try to wrap it up by Monday.
C.
Frank Warmerdam wrote:
Folks,
Is there anything holding back promoting geos 3.0.0rc4 to 3.0.0 final?
Also, I'm getting emailed roughly daily about updates to a bug in the
geos
bug database from a spammer.
http://postgis.refractions.net/bugs/bug.php?op=show&bugid=118
Note the funky URL link.
Can something be done to secure this bug database against un-logged in
spammers?
Best regards,
<run>
<!--precisionModel scale="1.0" offsetx="0.0" offsety="0.0"/-->
<precisionModel type="FLOATING"/>
<case>
<desc>
One Outer ring is returned instead of two outer rings.
One of the Polygons is not in the resultant set
</desc>
<a>
MULTIPOLYGON (((
439 819,
747 185,
439 184,
439 819
)), ((
215 1191 0,
1620 1203,
1483 0,
215 1191
)))</a>
<b>
MULTIPOLYGON (((
439 819,
747 185,
439 184,
439 819
)), ((
215 1191 0,
1620 1203,
1483 0,
215 1191
)))
</b>
<test>
<op name="isValid" arg1="a">true</op>
</test>
<test>
<op name="isValid" arg1="b">true</op>
</test>
<test>
<desc>
One Outer ring is returned instead of two outer rings.
One of the Polygons is not in the resultant set
</desc>
<op name="union" arg1="a" arg2="b">
MULTIPOLYGON (((
439 819,
747 185,
439 184,
439 819
)), ((
215 1191 0,
1620 1203,
1483 0,
215 1191
)))
</op>
</test>
<test>
<desc>
One Outer ring is returned instead of two outer rings.
One of the Polygons is not in the resultant set
</desc>
<op name="buffer" arg1="a" arg3="0.0">
MULTIPOLYGON (((
439 819,
747 185,
439 184,
439 819
)), ((
215 1191 0,
1620 1203,
1483 0,
215 1191
)))</op>
</test>
<test>
<desc>
Works Correctly
</desc>
<op name="intersection" arg1="a" arg3="b">
MULTIPOLYGON (((
439 819,
747 24,
397 24,
439 819
)), ((
215 1191,
1620 1203,
1483 25,
215 1191
)))
</op>
</test>
</case>
</run>
_______________________________________________
geos-devel mailing list
geos-devel@geos.refractions.net
http://geos.refractions.net/mailman/listinfo/geos-devel
