Closed eliotjordan closed 7 years ago
I agree, Eliot. I thought that we could just add more values to this. I know we have done this in our own local instance of GeoBlacklight. We recently added ESRI Geodatabase (https://geo.nyu.edu/?f%5Bdc_format_s%5D%5B%5D=ESRI+Geodatabase&f%5Bdct_provenance_s%5D%5B%5D=NYU) for instance. But I like the suggestion of using the GDAL and OGR file format list as a benchmark for the controlled vocabulary.
That's what we're doing in GeoConcerns. Finding a comprehensive list of geospatial file formats is hard. GDAL and OGR are the most comprehensive and widely used tools for working with geodata. While not perfect or totally consistent, I'd argue that the library's list of formats is a de facto standard.
+1 .. We currently use the locales to translate to a preferred label. Most likely we need to figure out something else there to support this. https://github.com/geoblacklight/geoblacklight/blob/master/config/locales/geoblacklight.en.yml#L16-L21
+1000 We are dealing with a ton of formats for the CIC project (ArcInfo Coverages, ASCII Grids, ERDAS img, File GDBs, KML) etc. and I've been putting off dealing with the problem.
@krdyke It would be great to start a list of the formats that you need. That way we can look at starting to support these.
@eliotjordan do we have the extracted GDAL/OGR file format list somewhere?
@drh-stanford I don't think we do. We just referred to the gdal manual pages. This seems like a great RDF property and value set to talk about with the Hydra URI Management Working Group.
I guess the issue is whether to use their code
value or human-readable one. If it's going to be a facet value the latter is probably preferable.
See http://gdal.org/formats_list.html http://gdal.org/ogr_formats.html
See https://github.com/geoblacklight/geoblacklight/pull/477 -- this PR removes the enum
to enforce a controlled vocab during validation.
To me, this is the same discussion as https://github.com/geoblacklight/geoblacklight/issues/429 - Going to close this and move the discussion over there
I'm working on creating geoblacklight documents from hydra/geo_concerns works and came across a limitation with the current geoblacklight schema. The
dc_format
field is currently limited to "Shapefile", "GeoTIFF", and "ArcGRID", but GeoConcerns supports a far larger set of formats. To fix this, I propose that we expand the list of formats:FileGDB
, or alternatively, by its format nameESRI FileGDB
.