Re: [geos-devel] [GEOS] #980: GEOSBuffer emits invalid buffering a collection

[GEOS]

#980: GEOSBuffer emits invalid buffering a collection
+---
 Reporter:  komzpa  |   Owner:  geos-devel@…
 Type:  defect  |  Status:  closed
 Priority:  major   |   Milestone:
Component:  Default | Version:  3.7.0
 Severity:  Unassigned  |  Resolution:  fixed
 Keywords:  |
+---
Changes (by strk):

 * status:  new => closed
 * resolution:   => fixed


Comment:

 The test is already included in GEOS testsuite, data from 979 passes in
 current GEOS so I'm closing this

-- 
Ticket URL: 
GEOS 
GEOS (Geometry Engine - Open Source) is a C++ port of the Java Topology Suite 
(JTS).
___
geos-devel mailing list
geos-devel@lists.osgeo.org
https://lists.osgeo.org/mailman/listinfo/geos-devel

Re: [geos-devel] [GEOS] #980: GEOSBuffer emits invalid buffering a collection

[GEOS]

#980: GEOSBuffer emits invalid buffering a collection
+---
 Reporter:  komzpa  |   Owner:  geos-devel@…
 Type:  defect  |  Status:  new
 Priority:  major   |   Milestone:
Component:  Default | Version:  3.7.0
 Severity:  Unassigned  |  Resolution:
 Keywords:  |
+---

Comment (by strk):

 I'd like to test this with current GEOS but there's in input in this
 ticket ?
 Martin where did you find the input ?
 Both exhibit1 and exhibit2 are self-intersecting multipolygons.
 In both cases Buffer(0) output a still-invalid geometry, while UnaryUnion
 fixes the invalidity.
 I don't think by contract Buffer(0) is to be guaranteed to output a valid
 geometry from an invalid one ?

-- 
Ticket URL: 
GEOS 
GEOS (Geometry Engine - Open Source) is a C++ port of the Java Topology Suite 
(JTS).
___
geos-devel mailing list
geos-devel@lists.osgeo.org
https://lists.osgeo.org/mailman/listinfo/geos-devel

Re: [geos-devel] [GEOS] #980: GEOSBuffer emits invalid buffering a collection

[GEOS]

#980: GEOSBuffer emits invalid buffering a collection
+---
 Reporter:  komzpa  |   Owner:  geos-devel@…
 Type:  defect  |  Status:  new
 Priority:  major   |   Milestone:
Component:  Default | Version:  3.7.0
 Severity:  Unassigned  |  Resolution:
 Keywords:  |
+---

Comment (by mdavis):

 Snap-rounding is highly applicable to buffer construction, and should
 solve problems like this.   It's probably too much impact to try and add
 it to the current (complex) buffer code, so that may have to wait for a
 rewrite using the upcoming OverlayNG framework. Although at that point
 UnaryUnion should work as well.

-- 
Ticket URL: 
GEOS 
GEOS (Geometry Engine - Open Source) is a C++ port of the Java Topology Suite 
(JTS).
___
geos-devel mailing list
geos-devel@lists.osgeo.org
https://lists.osgeo.org/mailman/listinfo/geos-devel

Re: [geos-devel] [GEOS] #980: GEOSBuffer emits invalid buffering a collection

[GEOS]

#980: GEOSBuffer emits invalid buffering a collection
+---
 Reporter:  komzpa  |   Owner:  geos-devel@…
 Type:  defect  |  Status:  new
 Priority:  major   |   Milestone:
Component:  Default | Version:  3.7.0
 Severity:  Unassigned  |  Resolution:
 Keywords:  |
+---

Comment (by mdavis):

 Confirmed that JTS `buffer(0)` exhibits the same issue.  It produces an
 invalid output on both geometries above (of which the first is a subset of
 the second).

 This is almost certainly due to a noding robustness failure, in an area
 with several very close high precision vertices:
 {{{
 A
 []  POLYGON(108)
 Ring-CW  Vert[0 11]  POINT ( -7885150.676182368 5282400.906109639 )
 Ring-CW  Vert[0 12]  POINT ( -7885150.676182364 5282400.90610964 )
 Ring-CW  Vert[0 13]  POINT ( -7885150.676182369 5282400.906109639 )
 }}}

 Some light precision reduction might solve this problem, in this case.

-- 
Ticket URL: 
GEOS 
GEOS (Geometry Engine - Open Source) is a C++ port of the Java Topology Suite 
(JTS).
___
geos-devel mailing list
geos-devel@lists.osgeo.org
https://lists.osgeo.org/mailman/listinfo/geos-devel

Re: [geos-devel] [GEOS] #980: GEOSBuffer emits invalid buffering a collection

[GEOS]

#980: GEOSBuffer emits invalid buffering a collection
+---
 Reporter:  komzpa  |   Owner:  geos-devel@…
 Type:  defect  |  Status:  new
 Priority:  major   |   Milestone:
Component:  Default | Version:  3.7.0
 Severity:  Unassigned  |  Resolution:
 Keywords:  |
+---

Comment (by dbaston):

 Can you please provide all arguments to GEOSBuffer in this ticket?

 What do exhibits 1 and 2 represent?

-- 
Ticket URL: 
GEOS 
GEOS (Geometry Engine - Open Source) is a C++ port of the Java Topology Suite 
(JTS).
___
geos-devel mailing list
geos-devel@lists.osgeo.org
https://lists.osgeo.org/mailman/listinfo/geos-devel