|
Re: esri .hdr open problem; minor hdf4 channels issue: msg#00162gis.gdal.devel
On Fri, Sep 26, 2003 at 10:41:59AM +0400, Vladimir Slepnev wrote: > Also, HDF4 datasets with 3D SDS sometimes open blatantly wrong: e.g. a > 6000x3x6000 SDS opened as 6000 bands, each of size 3x6000. This one > seems to be pretty easy to correct, here are the changes we've made to > hdf4dataset.cpp: ... > case 3: > // BIP > if((poDS->aiDimSizes[0] < poDS->aiDimSizes[1]) && > (poDS->aiDimSizes[0] < poDS->aiDimSizes[2]) ) > { > poDS->iBandDim = 0; > poDS->iYDim = 1; > poDS->iXDim = 2; > } > else > { > // BIL > if((poDS->aiDimSizes[1] <= poDS->aiDimSizes[0]) && > (poDS->aiDimSizes[1] <= poDS->aiDimSizes[2]) ) > { > poDS->iYDim = 0; > poDS->iBandDim = 1; > poDS->iXDim = 2; > } > else > { // BSQ > poDS->iYDim = 0; > poDS->iXDim = 1; > poDS->iBandDim = 2; > > } > } Vladimir, I'm not sure it is a good solution... But well, I shall commit proposed logic into the driver. There is much of unnatural intellect in the HDF4 driver already, not a problem if we add a bit more :-) Regards, Andrey -- Andrey V. Kiselev Home phone: +7 812 5274898 ICQ# 26871517 |
|
| <Prev in Thread] | Current Thread | [Next in Thread> |
|---|---|---|
| Previous by Date: | Re: porting of gdal to (optional) libtool build finally completed: 00162, Andrey Kiselev |
|---|---|
| Next by Date: | Memory leak fixes in ogrfeaturesytle.cpp: 00162, Steve Brooks |
| Previous by Thread: | Re: esri .hdr open problem; minor hdf4 channels issuei: 00162, Frank Warmerdam |
| Next by Thread: | tiger 108th glitch?: 00162, Heitzso |
| Indexes: | [Date] [Thread] [Top] [All Lists] |
| News | FAQ | advertise |