Andrea Aime a écrit :
> Can't set a value to the parameter "Latitude of natural origin".
> Can't set a value to the parameter "Scale factor at natural origin".

May be missing aliases. I will look at that we I will have a chance.


> No transform for classification "American Polyconic".
> No transform for classification "Bonne (South Orientated)".
> No transform for classification "Cassini-Soldner".
> No transform for classification "Guam Projection".
> No transform for classification "Hyperbolic Cassini-Soldner".
> No transform for classification "Laborde Madagascar".
> No transform for classification "Lambert Azimuthal Equal Area (Spherical)".
> No transform for classification "Lambert Azimuthal Equal Area".
> No transform for classification "Lambert Conic Conformal (West Orientated)".
> No transform for classification "Lambert Conic Near-Conformal".
> No transform for classification "Lambert Cylindrical Equal Area (Spherical)".
> No transform for classification "Modified Azimuthal Equidistant".
> No transform for classification "Transverse Mercator Zoned Grid System".
> No transform for classification "Tunisia Mining Grid".

They are unimplemented map projections, except "Lambert Azimuthal Equal Area"
which is available only on trunk at this time. We need volunter time for
implementing the remaining ones. We can't expect those projections to be
implemented for Geotools 2.3.


> Type "North along 0 deg" is unknow in this context.
> Type "North along 105 deg East" is unknow in this context.
> Type "North along 105 deg West" is unknow in this context.
> Type "North along 120 deg East" is unknow in this context.
> Type "North along 120 deg West" is unknow in this context.
> Type "North along 130 deg West" is unknow in this context.
> Type "North along 135 deg East" is unknow in this context.
> Type "North along 135 deg West" is unknow in this context.
> Type "North along 15 deg East" is unknow in this context.
> Type "North along 15 deg West" is unknow in this context.
> Type "North along 160 deg East" is unknow in this context.
> Type "North along 165 deg East" is unknow in this context.
> Type "North along 165 deg West" is unknow in this context.
> Type "North along 180 deg" is unknow in this context.
> Type "North along 45 deg East" is unknow in this context.
> Type "North along 45 deg West" is unknow in this context.
> Type "North along 60 deg East" is unknow in this context.
> Type "North along 60 deg West" is unknow in this context.
> Type "North along 75 deg East" is unknow in this context.
> Type "North along 75 deg West" is unknow in this context.
> Type "North along 90 deg East" is unknow in this context.
> Type "South along 10 deg West" is unknow in this context.
> Type "South along 100 deg East" is unknow in this context.
> Type "South along 180 deg" is unknow in this context.
> Type "South along 45 deg East" is unknow in this context.
> Type "South along 50 deg East" is unknow in this context.
> Type "South along 60 deg West" is unknow in this context.
> Type "South along 90 deg East" is unknow in this context.
> Type "South along 90 deg West" is unknow in this context.
> Type "east-south-east" is unknow in this context.
> Type "north north east" is unknow in this context.
> Type "north-east" is unknow in this context.
> Type "north-west" is unknow in this context.

The message is a little bit misleading. On trunk, I'm changing it for "Unknown
axis direction". This is related to http://jira.codehaus.org/browse/GEOT-936 and
I started to work on it yesterday. I don't plan to port it to the 2.3 branch
however.

This issue appears in map projections used for North or South poles. Instead of
"East" and "North" axis directions, map projection has for example the following
axis:

  * South along 180 deg
  * South along 90 deg East

The referencing module was not prepared to handle this kind of axis directions,
so supporting them will need some though and will not be fixed in 2.3.


> Unit conversion from "DMS" to "°" is non-linear.

"DMS" stands for "Degrees-Minutes-Seconds": this is sexagesimal degrees, as
opposed to decimal degrees represented by the "°" symbol.

This issue is similar to the "axis order" issue, and we need to look at what is
the common practice in the field. For some CRS in the EPSG database, (latitude,
longitude) axis units are sexagesimal degrees instead of decimal degrees. For
example 18°30' is represented by 18.3 rather than 18.5.

It is very hard to perform any computation on sexagesimal degrees, and they
should never have existed in the EPSG database in my opinion. But the EPSG
database is like that. In current implementation, Geotools can not do much with
such CRS since conversions from "DMS" units to degrees or radians is non-linear.

Geotools provides a hint for forcing units to decimal degrees:
Hints.FORCE_STANDARD_AXIS_UNITS. The idea is similar to
Hints.FORCE_LONGITUDE_FIRST_AXIS_ORDER. By default this hint is not set.

I suspect that many GIS implementation ignore sexagesimal units and use decimal
units instead, in the same way than they ignored the (latitude, longitude) axis
order. However I would need a confirmation of that before changing the default,
if we change it.

I'm not sure that we should change the default since many CRS using sexagesimal
degrees in the EPSG database have been deprecated and replaced by an other CRS
using decimal degrees. So I assume that if a user ask for a deprecated CRS, he
know what he is doing. Do we receive bug reports from users about the "Unit
conversion from DMS to ° is non-linear" exception?


> Unsupported unit: 9115
> Unsupported unit: 9116
> Unsupported unit: 9117
> Unsupported unit: 9118
> Unsupported unit: 9119
> Unsupported unit: 9120

They are different flavour of DMS units. Not all of them are supported, and I do
not plan to support them shortly (it will be easier after we switch to the
JScience units framework).

        Martin


-------------------------------------------------------------------------
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

Reply via email to