[gdal-dev] Re: patch for ISIS2 driver

[Trent M Hare]

Ludovic,
        This is excellent news and again any improvement to the original 
code is welcome. I actually am in the process of how to map the projection 
information for a potential ISIS3 writer. I also have a Python PDS label 
creation routine that I was thinking about pushing into library properly 
(instead of just using a Python script). It would also make sense to merge 
the PDS and ISIS2 code set however there are subtitle differences in both 
formats that we must be careful of.

Before we can incorporate your changes into GDAL, I would like to see 
projection export support implemented also. The main reason for these 
readers was to help propagate the projection, so I would want to see this 
working both directions. I will see where it stands and if I can help. I'm 
not sure how to implement "create_copy" so we might need help from the 
GDAL team.

Let me check out your patch and test some images. If all looks good we ask 
the GDAL team to also help incorporate your patch. Also there is another 
PDS fix that needs to be incorporated. Maybe you might have time to see if 
it would be a quick fix since you have been working wit the code recently 
(please see issue below). If you don't have time please let me know. I am 
also curious how you managed to support ISIS2 side planes? This capability 
was removed from ISIS3 due to lack of use and complexity. 

Thanks again for your willingness to share your improvements. Just out of 
curiosity - what facility are you working with?

I will be in touch shortly.

Regards,
Trent

----------
Comments
i have a patch for gdal 1.7.3 to improve isis2 driver.
* writing functionality for QUBE object  - again I would like to see an 
attempt at supporting projections also - I will help.
* grab null value   - What does this mean as support for a single NULL 
value was there? Does this address the fact that all ISIS2, PDS, and ISIS3 
can support multiple NULL values? 
* improve the type reading - great
* can read qube with back plane, side plane and bottom plane - great again 
sI will need to see how you supported side/bottom planes. Do you make a 
new file? 
* can read a file with label detached or attached. - I thought this was 
already implemented? Maybe there were some fixes though. 

----------

New PDS bug which I could use help with -- I have recently seen a new 
label from the LOLA team that allows one label to point to two different 
image types. This breaks the simple GDAL parser. 

In these LOLA labels there is a new section which wraps the file 
description.
OBJECT                    = UNCOMPRESSED_FILE
...
END_OBJECT                = UNCOMPRESSED_FILE
from: http://imbrium.mit.edu/DATA/LOLA_GDR/CYLINDRICAL/IMG/LDEC_16.LBL

and then for a compressed file. I know the PDS reader doesn't support JP2s 
but it would be nice (or at least report an appropriate warning):
OBJECT                    = COMPRESSED_FILE
  FILE_NAME                 = "LDEC_16.JP2"
  RECORD_TYPE             = UNDEFINED
  FILE_RECORDS            = UNK
  ENCODING_TYPE           = "JP2"
  INTERCHANGE_FORMAT      = BINARY
  UNCOMPRESSED_FILE_NAME  = "LDEC_16.IMG"
  REQUIRED_STORAGE_BYTES  = "14831085"
  ^DESCRIPTION            = "JP2INFO.TXT"
END_OBJECT                = COMPRESSED_FILE
http://imbrium.mit.edu/DATA/LOLA_GDR/CYLINDRICAL/JP2/LDEC_16_JP2.LBL

Unfortunately I have seen these two sections in the same label! I'm not 
sure how to handle that. I guess just pick up the first one that works. 
And again, it will only be the "UNCOMPRESSED_FILE" section for now.






From:
Ludovic Mercier <ludovic.mercier at gmail.com>
To:
gdal-dev at lists.osgeo.org
Cc:
thare at usgs.gov, rsoricone at usgs.gov
Date:
12/06/2010 09:02 AM
Subject:
patch for ISIS2 driver



Hi,

i have a patch for gdal 1.7.3 to improve isis2 driver.
The patch had
* writing functionality for QUBE object
* grab null value
* improve the type reading
* can read qube with back plane, side plane and bottom plane
* can read a file with label detached or attached.

and i have modify nasakeyword handler to reduce sensibility of parser to 
little error on comment. (Like if we have char after end comment.)

but i have rewrite/move a lot of lines... I use this patch for a 
research project from one year ago. I work on hyperspectral data with 
OTB library.

I think is it possible to easily merge pdsdataset.cpp and isis2dataset.cpp
pdsdataset work on IMAGE tag and isis2 on QUBE tag but it is the same 
workflow for all object (QUBE, IMAGE or SPECTRAL_QUBE)

Warning for writing no geographical information are writing now, and 
create_copy is not implemented.

Ludovic.

[attachment "gdal-1.7.3-lpg-613.patch" deleted by Trent M 
Hare/GD/USGS/DOI] 

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.osgeo.org/pipermail/gdal-dev/attachments/20101206/bdeee151/attachment-0001.html