ENVI metadata in machine parseable form

[verosk]

Dear GDAL people,

I'm trying to use GDAL/Python to read and process ENVI data produced by our airborne sensor. It already saved me lot of my time.

I realized that there is no easy(portable) way how to extract metadata of ENVI-saved data, e.g. wavelength, band width, names and other values. But these properties are sometimes crucial for processing of remote sensed imagery.

I suggest these metadata should be stored in Metadata structure of GDALObject in the similar way to current way of saving metadata ("Band_XX"), but in machine parseable format.

I'm able to prepare the patch for C++ code, but I'm not sure if there is a chance of accepting this type of patch. So I'm politely asking, may I prepare patch for GDAL, or should I solve this problem in separate way outside of the GDAL library?

Thank you for your responses.

--Veros

[warmerdam]

Věroš,

I'm sure we would be interested in such a submission. Please be clear that such a code submission will be assumed to be under the MIT/X open source license used for the rest of GDAL. It would be helpful if you also provide a smallish data sample with the metadata. I also think it would be more appropriate to attach the metadata to the band's themselves if it is band specific rather than attaching it to the dataset with band specific names.

[verosk]

Patch to envidataset.cpp - reading wavelengths to metadata

[verosk]

Attached patch to envidataset.cpp for reading wavelengths to bands object metadata.

Test data are available on ​http://truhla.lep.brno.cas.cz/~veros/gdal_3682_testdata.tar.gz

[verosk]

I would like to politely ask if there is any progress in processing of this ticket. The patch is already attached for several weeks.

If there is some problem with the patch, I can prepare a new better version.

Thank you for your response.

--Veros

[warmerdam]

Veros,

I will try to ensure the patch is reviewed and hopefully applied before the 1.8 release. Is there any particular reason you removed Chaitanya as a cc: on this ticket?

[verosk]

Thank you for your reply. I wasn't sure whether the patch is lost.

Removal of chaitanya in cc: is my error. Sorry for that, I'm still not very familiar with Trac.

--Veros

[robintw]

Has there been any progress on this? The last discussion seemed to be around two years ago, and it was suggested that it would be implemented by 1.8.0.

I am currently running 1.9.2, and the test provided by Veros doesn't seem to work.

Does this mean that this enhancement has not yet been implemented? If so, is it likely to be implemented fairly soon? Like Veros, I need to be able to get access to the Wavelengths list in the ENVI header as part of my processing of remotely-sensed imagery.

[Even Rouault]

trunk r25537 "ENVI: report wavelength and wavelength_units as band metadata (#3682)"