Paul, Any update to this bug? It sounded like you were working on a patch...
Stephen
On Mon, Aug 11, 2008 at 3:53 PM, Paul Ramsey <[EMAIL PROTECTED]>wrote:
> In retrospect, the problem was (surprise!) between my ears. The unit
> test harness runs in the context of a directory one lower than your
In retrospect, the problem was (surprise!) between my ears. The unit
test harness runs in the context of a directory one lower than your
test case directory. When I copied the test data down into the
expected location, it ran, and failed.
So, consider your test case confirmed on OS/X. It be broken
FWIW, this test case works in the current JTS codebase, but apparently
not in JTS 1.9. So this might narrow down where to look. There has
been at least one change to the buffering code since 1.9, to do with
improving how offset curve generation is handled.
Stephen Wong wrote:
Hi Paul,
Than
Hi Paul,
Thanks for the quick response. I just ran the test again (Windows Vista,
32-bit) and it doesn't get stuck on reading the WKB, so maybe it has to do
with cross-platform compatibility. I took the WKB and converted it to hex,
so hopefully, it will work this time. I've attached it to the tick
Stephen,
When I integrate your code into my OS/X build and run the unit tests,
I don't get as far as buffering...
---> group: geos::operation::BufferOp, test: test<1>
problem: unexpected exception
exception typeid: N4geos2io14ParseExceptionE
message: "ParseException: Unknown WKB ty
Hello,
I have found that when using BufferOp on a specific polygon, GEOS crashes
with an assertion failure. This failure appears to be within
RightmostEdgeFinder, and I have documented it on a ticket (
http://trac.osgeo.org/geos/ticket/188). On that ticket, I have also provided
a .zip file that re
