minc_modify_header can't delete top-level variables

[gdevenyi]

Not sure if this is a bug or a missing feature:

With a minc header such as:

<CLIP>
zspace = -39.3054472522805 ;

 acquisition = 0 ;

 dicom_0x0008 = 0 ;

 dicom_0x0009 = 0 ;

 dicom_0x0010 = 0 ;

 dicom_0x0018 = 0 ;

 dicom_0x0019 = 0 ;

 dicom_0x0020 = 0 ;

 dicom_0x0021 = 0 ;

 dicom_0x0023 = 0 ;

 dicom_0x0025 = 0 ;

 dicom_0x0027 = 0 ;

 dicom_0x0028 = 0 ;

 dicom_0x0029 = 0 ;

 dicom_0x0040 = 0 ;

 dicom_0x0043 = 0 ;

 dicom_0x0054 = 0 ;

 dicom_groups = 0 ;

 dicominfo = 0 ;

 patient = 0 ;

 study = 0 ;
}

I'd like to be able to delete these top-level dicom variables.

minc_modify_header -delete dicom_0x0018 17.mnc
-delete option requires argument <var>:<attr>

minc_modify_header -delete dicom_0x0018: 17.mnc
<no output, but variable still exists>

Is it possible to do this?

[rdvincent]

Not sure off the top of my head, but I assume it should work.

[andrewjanke]

If it's possible, I'm not aware of how. I once spent a large amount of time trying to get this to happen and in the end gave up, dumped the header to cdf, parsed it and then back to MINC to ensure things were gone when anonymising data. See mincanon and mincclean here:

https://github.com/BIC-MNI/minc-widgets/tree/master/mincanon

a

On 21 April 2016 at 07:29, Robert D Vincent notifications@github.com wrote:

Not sure off the top of my head, but I assume it should work.

— You are receiving this because you are subscribed to this thread. Reply to this email directly or view it on GitHub https://github.com/BIC-MNI/minc-tools/issues/34#issuecomment-212614235

[rdvincent]

there's some ability to delete things in HDF5. Possibly not in NetCDF. In neither case would I assume it would make the file smaller, or automatically overwrite the values (e.g. erase sensitive information).

[gdevenyi]

Well, mincanon and mincclean are exactly what I was trying to put togeather. So I guess my work is done :)

@andrewjanke expect an incoming patche for mincanon which uses a temp directory. I found that high speed read/writes using minc_modify_headers triggers a nasty bug in my NFSv4 systems