We've got some smart guys like Paul, Larry, and Stefan. I'm sure we can deal with the refactoring when the time comes.
Landon On Dec 13, 2007 2:38 PM, Martin Davis <[EMAIL PROTECTED]> wrote: > I'm actually pretty hopeful that switching to Geometry interfaces won't > be that big of a change - at least for code which is using > GeometryFactory to create geometry, rather than the raw constructors. > Code which simply uses Geometry objects and calls methods on them > shouldn't need to be changed. > > In this case, it's only GeometryFactory that will have to change (since > it will turn into an interface, rather than being a concrete class). > > Unfortunately, JUMP is probably a bit cavalier about creating new > GeometryFactory instances (rather than having a single central place > where GeometryFactory is created and managed). So there might be a few > places where code that looks like: > > GeometryFactory gf = new GeometryFactory(); > > needs to change to > > GeometryFactory gf = new DefaultGeometryFactory(); > > The problem with keeping the original Geometry classes (from my point of > view) is that you then need new names for all the interfaces. I > hesitate to introduce this kind of ugliness, since it will be lived with > for a long, long time. Short-term pain for long-term gain (admittedly > the pain is not mine! - but the gain is everyone's as well) > > > Sunburned Surveyor wrote: > > Thanks for the information Martin. I wonder what the appearance of the > > Geometry interface will mean for OpenJUMP. It sounds like it might > > require a major refactoring effort. > > > > Please forgive my Jaav ignorance, but might there be a way to > > introduce the Geometry interface without breaking backward > > compatability? What if you keep the existing Geometry class but make > > it implement the interface? Would that break existing code? > > > > I know I'm a little ignorant when it comes to the bigger picture with > > JTS, but I am curious about it and interested in its interaction with > > OpenJUMP. > > > > Landon > > > > On Dec 13, 2007 1:33 PM, Martin Davis <[EMAIL PROTECTED]> wrote: > > > >> No, Geometry interfaces have not yet made their appearance. When that > >> happens it willl probably break old code, so I've been holding off doing > >> that. I plan on making the release that switches to Geometry interfaces > >> be 2.0, so it's very clear that there's a breakage. > >> > >> AFAIK JTS 1.9 should be backwards compatible (unless people are doing > >> crazy things like calling code way down inside the Buffer package - as > >> just happened in another project). If there are differences, they > >> should be very minor and easy to fix. > >> > >> > >> Sunburned Surveyor wrote: > >> > >>> Martin, > >>> > >>> I look forward to the release of JTS 1.9. Has Geometry been refactored > >>> into an Interface in this release? I thought I remember reading that > >>> change would be included in the 1.9 release...but I might be wrong. > >>> > >>> Do you think the changes in 1.9 will break OpenJUMP? I know we are > >>> already one version behind. Maybe we should talk on the JPP list about > >>> what we will need to do to migrate to 1.9 in OpenJUMP after the bugs > >>> are worked out. > >>> > >>> The Sunburned Surveyor > >>> > >>> On Dec 12, 2007 4:22 PM, Martin Davis <[EMAIL PROTECTED]> wrote: > >>> > >>> > >>>> Possibly of interest to some - here's a link to a blog entry describing > >>>> some new functionality which will be appearing in JTS 1.9 > >>>> > >>>> http://lin-ear-th-inking.blogspot.com/2007/12/new-buffer-styles-in-jts-19.html > >>>> > >>>> JTS 1.9 is in late-stage testing, and is on track to be released by the > >>>> end of the year. > >>>> > >>>> -- > >>>> Martin Davis > >>>> Senior Technical Architect > >>>> Refractions Research, Inc. > >>>> (250) 383-3022 > >>>> > >>>> _______________________________________________ > >>>> jts-devel mailing list > >>>> jts-devel@lists.jump-project.org > >>>> http://lists.refractions.net/mailman/listinfo/jts-devel > >>>> > >>>> > >>>> > >>> _______________________________________________ > >>> jts-devel mailing list > >>> jts-devel@lists.jump-project.org > >>> http://lists.refractions.net/mailman/listinfo/jts-devel > >>> > >>> > >>> > >> -- > >> > >> Martin Davis > >> Senior Technical Architect > >> Refractions Research, Inc. > >> (250) 383-3022 > >> > >> _______________________________________________ > >> jts-devel mailing list > >> jts-devel@lists.jump-project.org > >> http://lists.refractions.net/mailman/listinfo/jts-devel > >> > >> > > _______________________________________________ > > jts-devel mailing list > > jts-devel@lists.jump-project.org > > http://lists.refractions.net/mailman/listinfo/jts-devel > > > > > > -- > Martin Davis > Senior Technical Architect > Refractions Research, Inc. > (250) 383-3022 > > _______________________________________________ > jts-devel mailing list > jts-devel@lists.jump-project.org > http://lists.refractions.net/mailman/listinfo/jts-devel > _______________________________________________ jts-devel mailing list jts-devel@lists.jump-project.org http://lists.refractions.net/mailman/listinfo/jts-devel
