OpenGeoMetadata/edu.virginia

Rights statements

Opened this issue · 6 comments

kimdurante commented

Hi Wayne,
I noticed that many of these records contain statements in the abstract and the legal constraints of the ISO metadata which state that the layers are restricted to UVa affiliates only. But they are all classified as public in geoBlacklight.

waynegraham commented

Hi Kim,

Thanks for noticing. I believe that was all pulled from a template in GeoNetwork (and improperly applied to all of the records). All of the metadata is going through review by metadata specialists to correct this (and a lot of other issues with this particular data).

Wayne

kimdurante commented

ok, Wayne. I know remediating old metadata is a lot of fun!

I have been working on an XSLT that transforms ISO19139 to GeoBlacklight. I might send it your way for review if you wouldn't mind taking a look?

waynegraham commented

That would be quite useful. I did a quick-and-dirty one (https://github.com/scholarslab/geoblacklight-converter/blob/master/xslt/geonetwork2solr.xsl) just to start looking at the data in the system, but there are (major) issues.

Let me know when you have something and we'll take a look.

kimdurante commented

Excellent, thanks!
I'll work with our 2 files and send you something we can test.

kimdurante commented

Hi Wayne,

Have a look at this:
https://github.com/kimdurante/metadataTransformations/blob/master/iso2geoBL.xsl

There are some elements, such as uuid and dc_identifier that I'm not sure where you are storing these values in ISO. Usually, this is a URI to the data.

You are correct that there is no way in ISO to determine geometry type. @drh-stanford has determined a way to read this info directly from the data using GDAL.

There's other issues in transforming certain metadata, such as rights - I've tried to set up some conditions for those. But I'm sure I'm missing a lot there.

Anyway, if this looks good, maybe we could move this to a new thread and build it out into something useful.

waynegraham commented

This looks really good!

I did also wrote a quick-and-dirty script to get the geometry types by querying the WFS for the layer (https://github.com/scholarslab/geoblacklight-converter/blob/master/types.rb) which gets around downloading the items to inspect in GDAL.

I had a thought on determining the dc_format_s (at least for our setup) that if the WFS returns an exception (e.g. Unknown namespace [McGregor]) it is a Raster, otherwise a Shapefile. A bunch of testing would need to be done to see if that is in fact true.

Anyway, want to move this discussion over to your metadataTransformations repo?