[sorry for late answer] Ok to add normalize() and equalsExact() to the C-API. No problem with buffer.xml, it's not even run by 'make check' (see xmltester/Makefile.am for a list of actually run tests)
--strk; On Wed, Jul 12, 2006 at 10:07:23PM -0600, Charlie Savage wrote: > Hi everyone, > > Strk pointed me at the XML test suite as a good source of test data. So > using Ruby, I now load the xml test cases and can run them through the > SWIG bindings which use the c api. > > To do this, however, requires adding 2 methods to the C API. > > The first method is Geometry::normalize which converts a geometry to its > cannonincal form. Without this method, the results from performing > operations (like buffer) will be in some undetermined order and thus the > test fails since the actual result does not match the expected result. > Since this seems to be a generally useful method I'd like to add in to > the C API. > > The second method is one I mentioned the other day, equalsExact. The > XMLTestRunner uses this method to verify that the expected and actual > results match. Unfortunately, using equals does not work since it > returns false in some cases where equalsExact returns true (you can set > a tolerance with equalsExact). Once again, this seems like a generally > useful method so I'd also like to add in to the C API. > > On my local box I have added both of these methods to the C API, and the > SWIG bindings now generate the same results as the C++ API (i've only > done some of the tests so far). > > Note that a number of the XML test cases fail using either API. For > example, in buffer.xml roughly half the tests fail. Is this a problem? > > Charlie > > > > > > _______________________________________________ > geos-devel mailing list > geos-devel@geos.refractions.net > http://geos.refractions.net/mailman/listinfo/geos-devel -- /"\ ASCII Ribbon Campaign \ / Respect for low technology. X Keep e-mail messages readable by any computer system. / \ Keep it ASCII. _______________________________________________ geos-devel mailing list geos-devel@geos.refractions.net http://geos.refractions.net/mailman/listinfo/geos-devel
