One thing I was wondering as I hunted down the FeatGeomFactoryImpl in
search of a CoordinateReferenceSystem is *why* DimensionModel would ever
of been made.
Well now I have an answer.
Those working on this library *never* had a real
CoordinateReferenceSystem object to work with. I know because I just
tried and I am in a maven dependency conflict.
unsupported/geometry/pom.xml include the following:
> <dependency>
> <groupId>org.opengis</groupId>
> <artifactId>geoapi</artifactId>
> <version>2.1-SNAPSHOT</version>
> </dependency>
> <dependency>
> <groupId>javax.units</groupId>
> <artifactId>jsr108</artifactId>
> <version>0.01</version>
> </dependency>
> <dependency>
> <groupId>junit</groupId>
> <artifactId>junit</artifactId>
> <version>3.8.1</version>
> <scope>test</scope>
> </dependency>
So far all pretty normal - trick is the following:
- That is a dependency on geoapi (the real Java 5 interfaces not our
usual geoapi-nogenerics stuff
Attempting to test the code with a real CoordinateReferenceSystem
involves the use of:
> <dependency>
> <groupId>org.geotools</groupId>
> <artifactId>gt2-referencing</artifactId>
> <version>${project.version}</version>
> <scope>test</scope>
> </dependency>
This results in a conflict:
1. geoapi-nogenerics is brought into the picture because gt2-referencing
needs it
2. geoapi was already brought into the picture because we need to
implement it
So no can go ... I have looked at this every which way and cannot figure
out a way for maven to let us do what we need. I am going to have to
stub a MockCoordinateReferenceSystem in order to get this thing tested.
Cheers,
Jody
PS. Sanjay did you ask for help on this stuff last fall?
> I am working on that now (the official build is Java 1.4 based) - I can
> commit code that compiles (but I have test failures).
>
> For interest I found that the unsupported/geometry implementation is
> built around some flaws ....
>
> FeatGeomFactoryImpl - is a "container" that holds the following
> - CoordianteReferenceSystem
> - PrecisionModel - *new for geoapi* (implementation already existed)
> - CoordinateFactoryImpl
> - PrimitiveFactoryImpl
> - PositionFactoryImpl - *new for geoapi* isolate position creation code
> in one spot
> - ComplexFactoryImpl
> - CollectionFactory - own invention to isolate create list code
> - dimensionModel - 2, 2.5, or 3 dimension (used as a strategy object by
> some calculations)
>
> It also looks like three singletons are available and used:
> - singleton2D,singleton205D, singleton3D
>
> I have two problems with this:
> - dimension model should not be an alternative to checking the
> coordinate reference system (should it?)
> - all instances contain back pointers to FeatGeomFactoryImpl - and often
> perform casts to get there (not safe)
> - singletons do not mix with the idea of changing your coordinate
> reference system
>
> So right now this implementation appears to not support coordinate
> reference systems (I hope I am reading the code wrong...debugger is
> warmed up now).
>
> Cheers,
> Jody
-------------------------------------------------------------------------
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT & business topics through brief surveys-and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV
_______________________________________________
Geotools-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/geotools-devel
