[gdal-dev] GDAL/OGR 1.9.1 Release Candidate

Even Rouault even.rouault at mines-paris.org
Tue May 15 16:23:32 EDT 2012


Le mardi 15 mai 2012 16:35:17, Frank Warmerdam a écrit :
> Folks,
> 
> I have prepared a GDAL/OGR 1.9.1 Release Candidate at:
> 
>    http://download.osgeo.org/gdal/gdal-1.9.1RC1.tar.gz
>    http://download.osgeo.org/gdal/gdal191RC1.zip
> 
> Please test and confirm there are no regressions since 1.9.0.  I'll send
> a separate message motioning to declare this 1.9.1 final once I've done a
> bit of testing myself.

./swig/include/perl/gdal_perl.i hasn't been updated, so the Perl bindings 
don't compile natively.

I although see that tiff_srs_epsg_2066 (gcore/tiff_srs.py) fails. Apparently it 
seems related to http://trac.osgeo.org/gdal/changeset/24390 (reverting it 
makes the test succeeds)

In GDAL 1.9.1, the SRS that we get back after the round-trip to geotiff is now 
:

PROJCS["Mount Dillon / Tobago Grid",
    GEOGCS["Mount Dillon",
        DATUM["Mount_Dillon",
            SPHEROID["Clarke 1858",6378293.645208759,294.2606763692569,
                AUTHORITY["EPSG","7007"]],
            AUTHORITY["EPSG","6157"]],
        PRIMEM["Greenwich",0],
        UNIT["degree",0.0174532925199433],
        AUTHORITY["EPSG","4157"]],
    PROJECTION["Cassini_Soldner"],
    PARAMETER["latitude_of_origin",11.25217861111111],
    PARAMETER["central_meridian",-60.68600888888889],
    PARAMETER["false_easting",37718.66159325],
    PARAMETER["false_northing",36209.91512952],
    UNIT["unknown",0.201166195164],
    AUTHORITY["EPSG","2066"]]

whereas in 1.9.0, it was 

PROJCS["Mount Dillon / Tobago Grid",
    GEOGCS["Mount Dillon",
        DATUM["Mount_Dillon",
            SPHEROID["Clarke 1858",6378293.645208759,294.2606763692569,
                AUTHORITY["EPSG","7007"]],
            AUTHORITY["EPSG","6157"]],
        PRIMEM["Greenwich",0],
        UNIT["degree",0.0174532925199433],
        AUTHORITY["EPSG","4157"]],
    PROJECTION["Cassini_Soldner"],
    PARAMETER["latitude_of_origin",11.25217861111111],
    PARAMETER["central_meridian",-60.68600888888889],
    PARAMETER["false_easting",187500],
    PARAMETER["false_northing",180000],
    UNIT["Clarke's link",0.201166195164],
    AUTHORITY["EPSG","2066"]]

Note the difference : the unit name is "unknown" instead of "Clarke's link". It 
doesn't seem very critical however. But I'm not sure why the test succeeds in 
trunk and not in 1.9 branch (tiff_srs.py is identical in trunk and 1.9). 
Perhaps another change in trunk counterbalances that.

For the record, on trunk, listgeo on "gdal_translate foo.tif 2066.tif -a_srs 
EPSG:2066" gives :
   Keyed_Information:
      GTModelTypeGeoKey (Short,1): ModelTypeProjected
      GTRasterTypeGeoKey (Short,1): RasterPixelIsArea
      GTCitationGeoKey (Ascii,27): "Mount Dillon / Tobago Grid"
      GeogCitationGeoKey (Ascii,13): "Mount Dillon"
      GeogAngularUnitsGeoKey (Short,1): Angular_Degree
      ProjectedCSTypeGeoKey (Short,1): Unknown-2066
      ProjLinearUnitsGeoKey (Short,1): Unknown-9039
      End_Of_Keys.
   End_Of_Geotiff.

On 1.9.1 :

   Keyed_Information:
      GTModelTypeGeoKey (Short,1): ModelTypeProjected
      GTRasterTypeGeoKey (Short,1): RasterPixelIsArea
      GTCitationGeoKey (Ascii,27): "Mount Dillon / Tobago Grid"
      GeogCitationGeoKey (Ascii,13): "Mount Dillon"
      GeogAngularUnitsGeoKey (Short,1): Angular_Degree
      ProjectedCSTypeGeoKey (Short,1): Unknown-2066
      PCSCitationGeoKey (Ascii,23): "LUnits = Clarke's link"
      ProjLinearUnitsGeoKey (Short,1): User-Defined
      ProjLinearUnitSizeGeoKey (Double,1): 0.201166195164   
      End_Of_Keys.
   End_Of_Geotiff.


> 
> Best regards,


More information about the gdal-dev mailing list