lat error from unfamiliar keyword "pseudo_standard_parallel_1"

[thare]

There is what appears to be a new keywrod listed as "pseudo_standard_parallel_1" for the Equirectangular projection. I think this has caused a problem in GDAL? There is the original "latitude_of_origin" which appears to be correct. Now the "pseudo..." keyword is the same value as the "latitude_of_origin" and this keyword seems to be the cause for the lat range to off by exactly the same amount (in the example off by 10 degrees).

-Trent

---

Correct center lat,lon range of image:
Latitude (centered): 10.6 Longitude (East): 156.4

Gdal 1.6.0dev reports the center as:
Center      (-1374994.500,  630301.250) (156d26'27.39"E, 20d38'7.68"N)

File: http://hirise-pds.lpl.arizona.edu/download/PDS/RDR/PSP/ORB_008100_008199/PSP_008199_1910/PSP_008199_1910_COLOR.JP2 (355MB)

Using GDAL 1.6.0dev built inside FWTools 2.2.1.
-------------

> gdalinfo PSP_008199_1910_COLOR.jp2

...

PROJCS["Equirectangular MARS",
    GEOGCS["GCS_MARS",
        DATUM["unknown",
            SPHEROID["unnamed",3395582.0270805,0]],
        PRIMEM["Greenwich",0],
        UNIT["degree",0.0174532925199433]],
    PROJECTION["Equirectangular"],
    PARAMETER["latitude_of_origin",10],
    PARAMETER["central_meridian",180],
    PARAMETER["false_easting",0],
    PARAMETER["false_northing",0],
    PARAMETER["pseudo_standard_parallel_1",10],
    UNIT["metre",1,
        AUTHORITY["EPSG","9001"]]]
Origin = (-1376386.875000000000000,637256.375000000000000)
Pixel Size = (0.250000000000000,-0.250000000000000)

Corner Coordinates:
Upper Left  (-1376386.875,  637256.375) (156d25'1.51"E, 20d45'10.17"N)  //Note the lat range is off by 10 degrees
Lower Left  (-1376386.875,  623346.125) (156d25'1.51"E, 20d31'5.19"N)
Upper Right (-1373602.125,  637256.375) (156d27'53.28"E, 20d45'10.17"N)
Lower Right (-1373602.125,  623346.125) (156d27'53.28"E, 20d31'5.19"N)
Center      (-1374994.500,  630301.250) (156d26'27.39"E, 20d38'7.68"N)

...

[warmerdam]

Trent,

Would there happen to be any smaller files demonstrating this problem?

It appears SetEquidistantRectangular2() was introduced in r13482 for ticket #2134. I'm adding dgrichard to the cc.