Le Friday 27 March 2009 20:05:11 Lucena, Ivan, vous avez écrit :
> Even,
>
> I see. You are right. With the absence of ComputeRasterStatistics() I got
> confused and I made the wrong
>
> assumption. BTW, why there is no wrapper for ComputeRasterStatistics()? That
would work then:
> >>> bd1.SetStatistics(bd2.ComputeRasterStatistics())
Because you don't use the latest and greatest GDAL ;-)
Extract from swig/include/Band.i from SVN head:
/* Interface method added for GDAL 1.7.0 */
%apply (double *OUTPUT){double *min, double *max, double *mean, double
*stddev};
%apply (IF_ERROR_RETURN_NONE) { (CPLErr) };
CPLErr ComputeStatistics( bool approx_ok, double *min = NULL, double *max =
NULL, double *mean = NULL, double *stddev = NULL,
GDALProgressFunc callback = NULL, void* callback_data=NULL){
return GDALComputeRasterStatistics( self, approx_ok, min, max, mean,
stddev, callback, callback_data );
}
%clear (CPLErr);
>
> Not a big deal. The work around is easy.
>
> Thank you very much,
>
> Ivan
>
> > -------Original Message-------
> > From: Even Rouault <even.roua...@mines-paris.org>
> > Subject: Re: [gdal-dev] python-gdal ComputStatistics return just two
> > values Sent: Mar 27 '09 18:45
> >
> > Hi Ivan,
> >
> > I don't think there's a bug.
> >
> > The C API for ComputeBandStats is :
> >
> > CPLErr CPL_DLL CPL_STDCALL
> > GDALComputeBandStats( GDALRasterBandH hBand, int nSampleStep,
> > double *pdfMean, double *pdfStdDev,
> > GDALProgressFunc pfnProgress,
> > void *pProgressData );
> >
> > So, the python API is consistant with it.
> >
> > So it is expected that "bd1.SetStatistics(bd2.ComputeBandStats())" will
> > fail as it is provided with 2 values instead of the expected 4.
> >
> > Best regards,
> > Even
> >
> > Le Friday 27 March 2009 19:02:14 Lucena, Ivan, vous avez écrit :
> > > Hi there,
> > >
> > > I believe that ComputeBandStats() should return 4 values min, max,
> > > mean and stddev) [1] but the Python API is returning just mean and
> > > stddev.
> > >
> > > python:
> > > >>> bd.ComputeBandStats()
> > >
> > > (9.0071713001936118, 11.350094331448236)
> > >
> > > >>> bd.ComputeRasterMinMax()
> > >
> > > (-1.7899999618530273, 31.149999618530273)
> > >
> > > gdalinfo:
> > >
> > > Band 1 Block=360x1 Type=Float32, ColorInterp=Gray
> > > Min=-1.790 Max=31.150
> > > Minimum=-1.790, Maximum=31.150, Mean=9.007, StdDev=11.350
> > > Unit Type: unspecified
> > > Metadata:
> > > STATISTICS_MINIMUM=-1.789999961853
> > > STATISTICS_MAXIMUM=31.14999961853
> > > STATISTICS_MEAN=9.0071713001936
> > > STATISTICS_STDDEV=11.350094331448
> > >
> > > Is that a bug?
> > >
> > > See what happens in that case:
> > > >>> bd1.SetStatistics(bd2.ComputeBandStats())
> > >
> > > Traceback (most recent call last):
> > > File "<interactive input>", line 1, in <module>
> > > File "E:\Python25\Lib\site-packages\osgeo\gdal.py", line 729, in
> > > SetStatistics return _gdal.Band_SetStatistics(*args)
> > > TypeError: Band_SetStatistics() takes exactly 5 arguments (2 given)
> > >
> > > My best regards,
> > >
> > > Ivan
> > >
> > > [1] http://www.gdal.org/gdal_8h.html#a93b6b6ea6e71017ce25524e1a9ef1e3
> > >
> > > _______________________________________________
> > > gdal-dev mailing list
> > > gdal-dev@lists.osgeo.org
> > > http://lists.osgeo.org/mailman/listinfo/gdal-dev
_______________________________________________
gdal-dev mailing list
gdal-dev@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/gdal-dev
