Issues with the transformed things in SESAR iSB instance

[dannymandel]

Will use this issue to track the problems with Core JSON things in the SESAR iSB instance.

As a general rule, it looks like the data fields in the SESAR are camel-cased, while the Core JSON is now using lowercase separated by underscores.

e.g.

SESAR

    "curation":
    {

        "curationLocation": "Integrated Ocean Drilling Program (IODP)",
        "accessConstraints": "Not Provided"
    },

vs. Core JSON

    "curation":
    {
        "curation_location": "",
        "access_constraints":
        [
            ""
        ]
    },

[dannymandel]

The full list of constants is in this file:

isamples_inabox/isamples_metadata/metadata_constants.py

Line 4 in d6b8d31

[dannymandel]

I think the bulk of recent changes were in this PR: #311

e.g. sampling_site -> sample_location vs. samplingSite -> location

[dannymandel]

hasSpecimenCategory has been renamed to has_sample_object_type

[dannymandel]

A few issues explaining what changed, motivation, etc:

#307
#308
#310

[dannymandel]

It looks like the SESAR Controlled Vocabularies are all using labels instead of URIs, e.g.

    "hasMaterialCategory":
    [
        "Biogenic non-organic material",
        "Material",
        "Rock"
    ],

vs.

    "has_material_category":
    [
        {
            "identifier": "https://w3id.org/isample/vocabulary/material/1.0/anyanthropogenicmaterial"
        }
    ],

[dannymandel]

The mapping from labels to controlled vocabulary identifiers is non-trivial. Here's a script I wrote to convert things in Solr over to the identifiers:

isamples_inabox/scripts/migrations/solr/use_controlled_vocabulary_identifiers.py

Line 4 in d6b8d31

import click

[dannymandel]

The key renames should be encapsulated by this commit: isamplesorg/isamples_sesar@3591df9