[MetaCRS] Where's Denmark gone?

Norm Olsen norm.olsen at autodesk.com
Mon Jan 24 13:59:22 EST 2011


Hello Steve . . .

This all went down almost three years ago now, so my memory might not be that accurate.

While the source code which carries the SYS34 coefficients (there are two separate renditions, pre 1999 and post 1999 if memory serves) is freely available upon request, we were not able to get permission to distribute the coefficients or the code that uses them.  Thus, you have to obtain these yourself and play the games described in the source code module Hugues references.

The instructions provide the information necessary to implement the coefficients.  Note that the code which we distribute to use the coefficients is our rendition of the power series algorithm.  If your doing a lot of work using the Danish System 34, you might to consider porting the KMS code as well as the coefficients.  The KMS version of the code will probably run a lot faster.

The coordinate system compiler is smart enough to detect if you have done the code modifications referenced above and will not compile the SYS34 Danish definitions if it believes that the code to do the calculations properly is not present in the library.  Thus, you see the difference between the source code (.asc) and the compiled dictionaries (.CSD).  Once you complete the steps as described in CSsys34KMS.c, the compiler will compile all of the Danish systems.

This accounts for about 10 systems.  The remainder of the difference is (my guess) the 50 or so definitions in the TEST group which are not compiled by the compiler unless the /t (-t on Linux) command line option is present.  The definitions in the TEST group are essentially implementations of the systems in the Appendix of Snyder's Map Projections: A Working Manual, are used only for testing, and are typically left out of distribution dictionaries.

Norm
--
Norm Olsen
Software Engineer
Autodesk, Inc.
303-277-9967


-----Original Message-----
From: metacrs-bounces at lists.osgeo.org [mailto:metacrs-bounces at lists.osgeo.org] On Behalf Of Hugues Wisniewski
Sent: Friday, January 21, 2011 4:02 PM
To: Steve Stanton; metacrs at lists.osgeo.org
Subject: RE: [MetaCRS] Where's Denmark gone?

Hello Steve,

You'll most likely get a more accurate answer from Norm Olsen when he reads this

You're right these Danish systems are not in the open source repository of CsMap because of legal reasons Check out the content of the source file to understand how to get to the polynomial coefficients for Danish System 34 used in those systems:
http://svn.osgeo.org/metacrs/csmap/trunk/CsMapDev/Source/CSsys34KMS.c

You'll read that these coefficients are freely available from KMS but KMS never allowed them to be posted on osgeo.org

Hugues

-----Original Message-----
From: metacrs-bounces at lists.osgeo.org [mailto:metacrs-bounces at lists.osgeo.org] On Behalf Of Steve Stanton
Sent: Friday, January 21, 2011 2:38 PM
To: metacrs at lists.osgeo.org
Subject: [MetaCRS] Where's Denmark gone?

Are there any issues with the CSMap dictionary compiler? I ran it to convert the various *.ASC files, but when I loop through the resultant *.CSD files (with calls to CS_csrd), I seem to be missing a few Danish systems. Searching coordsys.asc for "CS_NAME: Denmark", I get this:

  Line 38672: CS_NAME: Denmark-S34-J
  Line 38685: CS_NAME: Denmark-S34-S
  Line 38698: CS_NAME: Denmark-S45-B
  Line 42340: CS_NAME: DenmarkS34-J
  Line 42353: CS_NAME: DenmarkS34-S
  Line 42366: CS_NAME: DenmarkS45-B
  Line 42379: CS_NAME: DenmarkS34-J99
  Line 42392: CS_NAME: DenmarkS34-S99
  Line 42405: CS_NAME: DenmarkS45-B99
  Line 42418: CS_NAME: DenmarkS45-B01
  Line 42450: CS_NAME: Denmark-KP2000-J
  Line 42464: CS_NAME: Denmark-KP2000-Z
  Line 42478: CS_NAME: Denmark-KP2000-B

But when I iterate using CS_csrd, I only get the 3 Denmark-KP systems. Could this be intentional (i.e. is there something that says the other systems should be ignored)?

Actually, it looks like there must be some other missing systems. When I count the number of "CS_NAME:" strings in coordsys.asc, I get 6338 occurrences. But a loop using CS_csrd only produces 6267 systems.

Steve

_______________________________________________
MetaCRS mailing list
MetaCRS at lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/metacrs


More information about the MetaCRS mailing list