Add schema for an "object stream" dataset description

This is conceptually already quite close to what we need to get at --
for serializing datasets, starting from their Git repository
representation.

This is only a minimal sketch at this point, but it demos how the
recently found duality of semantic-defining and structure-defining
classes would look like in this contex.

There are a bunch of open question that need more consideration.
Something like:

- Should we make the SE-level classes define `exact_mappings` to their
  ontology counterparts, or use discovery like shown in
  #37

- what kind of value do we use for `meta_type`? CURIEs? or plain class
  names?

- ...

Makefile

@@ -24,6 +24,7 @@ build/context.jsonld: src/linkml/schemas/ontology.yaml
 build/linkml-docs: \
 	build/linkml-docs/ontology \
 	build/linkml-docs/data-access \
+	build/linkml-docs/datalad-dataset-components \
 	build/linkml-docs/datalad-dataset-version
 #   build/linkml-docs/git-provenance-schema
 build/linkml-docs/%: src/linkml/schemas/%.yaml src/extra-docs/%-schema
@@ -33,11 +34,11 @@ build/linkml-docs/%: src/linkml/schemas/%.yaml src/extra-docs/%-schema
 		--diagram-type er_diagram \
 		--metadata \
 		--format markdown \
-		--example-directory src/examples/$$(basename $@)-schema \
+		--example-directory src/examples/$* \
 		-d $$([ "$*" = "ontology" ] && echo $@ || echo $@-schema) \
 		$<
 	# try to inject any extra-docs (if any exist)
-	-cp -r src/extra-docs/$$(basename $@)-schema/*.md $@
+	-cp -r src/extra-docs/$*-schema/*.md $@
 
 build/mkdocs-site: build/linkml-docs src/extra-docs/*.md
 	# top-level content
@@ -49,6 +50,7 @@ check: check-models check-validation
 # add additional schemas to lint here
 check-models: \
 	check-model-data-access \
+	check-model-datalad-dataset-components \
 	check-model-datalad-dataset-version \
 	check-model-ontology
 #	check-model-git-provenance
@@ -80,6 +82,8 @@ check-model-%: src/linkml/schemas/%.yaml
 check-validation: \
 	convert-examples-data-access \
 	check-validation-data-access \
+	convert-examples-datalad-dataset-components \
+	check-validation-datalad-dataset-components \
 	convert-examples-datalad-dataset-version \
 	check-validation-datalad-dataset-version \
 	convert-examples-ontology
@@ -100,6 +104,7 @@ check-invalid-validation-%: tests/%-schema/validation src/linkml/schemas/%.yaml
 
 convert-examples: \
 	convert-examples-data-access \
+	convert-examples-datalad-dataset-components \
 	convert-examples-datalad-dataset-version \
 	convert-examples-ontology
 #	convert-examples-git-provenance

src/examples/datalad-dataset-components/ContainerSE-DataladDataset-minimal.json

@@ -0,0 +1,10 @@
+{
+  "components": [
+    {
+      "meta_id": "datalad:0b76362c-aa27-11ee-be29-b3b123281259",
+      "meta_type": "dlccs:DataladDatasetSE",
+      "uuid": "0b76362c-aa27-11ee-be29-b3b123281259"
+    }
+  ],
+  "@type": "ContainerSE"
+}

src/examples/datalad-dataset-components/ContainerSE-DataladDataset-minimal.rdf

@@ -0,0 +1,13 @@
+@prefix datalad: <https://concepts.datalad.org/namespace/dataset-uuid/> .
+@prefix dlccs: <https://concepts.datalad.org/schemas/datalad-dataset-components/> .
+@prefix dlco: <https://concepts.datalad.org/ontology/> .
+@prefix ns1: <https://concepts.datalad.org/schemas/datalad-dataset-components/:> .
+@prefix xsd: <http://www.w3.org/2001/XMLSchema#> .
+
+datalad:0b76362c-aa27-11ee-be29-b3b123281259 dlco:meta_type "dlccs:DataladDatasetSE"^^xsd:anyURI ;
+    dlco:uuid "0b76362c-aa27-11ee-be29-b3b123281259"^^<http://purl.obolibrary.org/obo/NCIT_C54100> .
+
+[] a dlccs:Container ;
+    ns1:components datalad:0b76362c-aa27-11ee-be29-b3b123281259 .
+
+

src/examples/datalad-dataset-components/ContainerSE-DataladDataset-minimal.yaml

@@ -0,0 +1,4 @@
+components:
+  - meta_id: datalad:0b76362c-aa27-11ee-be29-b3b123281259
+    meta_type: dlccs:DataladDatasetSE
+    uuid: 0b76362c-aa27-11ee-be29-b3b123281259

src/examples/datalad-dataset-components/ContainerSE-DataladDatasetVersion-linkage.json

@@ -0,0 +1,15 @@
+{
+  "components": [
+    {
+      "meta_id": "datalad:0b76362c-aa27-11ee-be29-b3b123281259",
+      "meta_type": "dlccs:DataladDatasetSE",
+      "uuid": "0b76362c-aa27-11ee-be29-b3b123281259"
+    },
+    {
+      "meta_id": "gitsha:558275f650574389dcbbf7cd8ab5046482473fc8",
+      "meta_type": "dlccs:DataladDatasetVersionSE",
+      "is_version_of": "datalad:0b76362c-aa27-11ee-be29-b3b123281259"
+    }
+  ],
+  "@type": "ContainerSE"
+}

src/examples/datalad-dataset-components/ContainerSE-DataladDatasetVersion-linkage.rdf

@@ -0,0 +1,19 @@
+@prefix DCAT: <http://www.w3.org/ns/dcat#> .
+@prefix datalad: <https://concepts.datalad.org/namespace/dataset-uuid/> .
+@prefix dlccs: <https://concepts.datalad.org/schemas/datalad-dataset-components/> .
+@prefix dlco: <https://concepts.datalad.org/ontology/> .
+@prefix gitsha: <https://concepts.datalad.org/namespace/gitsha/> .
+@prefix ns1: <https://concepts.datalad.org/schemas/datalad-dataset-components/:> .
+@prefix xsd: <http://www.w3.org/2001/XMLSchema#> .
+
+gitsha:558275f650574389dcbbf7cd8ab5046482473fc8 DCAT:isVersionOf datalad:0b76362c-aa27-11ee-be29-b3b123281259 ;
+    dlco:meta_type "dlccs:DataladDatasetVersionSE"^^xsd:anyURI .
+
+datalad:0b76362c-aa27-11ee-be29-b3b123281259 dlco:meta_type "dlccs:DataladDatasetSE"^^xsd:anyURI ;
+    dlco:uuid "0b76362c-aa27-11ee-be29-b3b123281259"^^<http://purl.obolibrary.org/obo/NCIT_C54100> .
+
+[] a dlccs:Container ;
+    ns1:components datalad:0b76362c-aa27-11ee-be29-b3b123281259,
+        gitsha:558275f650574389dcbbf7cd8ab5046482473fc8 .
+
+

src/examples/datalad-dataset-components/ContainerSE-DataladDatasetVersion-linkage.yaml

@@ -0,0 +1,7 @@
+components:
+  - meta_id: datalad:0b76362c-aa27-11ee-be29-b3b123281259
+    meta_type: dlccs:DataladDatasetSE
+    uuid: 0b76362c-aa27-11ee-be29-b3b123281259
+  - meta_id: gitsha:558275f650574389dcbbf7cd8ab5046482473fc8
+    meta_type: dlccs:DataladDatasetVersionSE
+    is_version_of: datalad:0b76362c-aa27-11ee-be29-b3b123281259

src/examples/datalad-dataset-components/DataladDatasetSE-minimal.json

@@ -0,0 +1,6 @@
+{
+  "meta_id": "datalad:0b76362c-aa27-11ee-be29-b3b123281259",
+  "meta_type": "dlccs:DataladDatasetSE",
+  "uuid": "0b76362c-aa27-11ee-be29-b3b123281259",
+  "@type": "DataladDatasetSE"
+}

src/examples/datalad-dataset-components/DataladDatasetSE-minimal.rdf

@@ -0,0 +1,8 @@
+@prefix datalad: <https://concepts.datalad.org/namespace/dataset-uuid/> .
+@prefix dlco: <https://concepts.datalad.org/ontology/> .
+@prefix xsd: <http://www.w3.org/2001/XMLSchema#> .
+
+datalad:0b76362c-aa27-11ee-be29-b3b123281259 dlco:meta_type "dlccs:DataladDatasetSE"^^xsd:anyURI ;
+    dlco:uuid "0b76362c-aa27-11ee-be29-b3b123281259"^^<http://purl.obolibrary.org/obo/NCIT_C54100> .
+
+

src/examples/datalad-dataset-components/DataladDatasetSE-minimal.yaml

@@ -0,0 +1,3 @@
+meta_id: datalad:0b76362c-aa27-11ee-be29-b3b123281259
+meta_type: dlccs:DataladDatasetSE
+uuid: 0b76362c-aa27-11ee-be29-b3b123281259

src/extra-docs/index.md

@@ -1,4 +1,5 @@
 - [DataLad concepts ontology](ontology/)
+- [DataLad dataset components schema](datalad-dataset-components-schema/)
 - [DataLad dataset version schema](datalad-dataset-version-schema/)
 - [Data access schema](data-access-schema/)
 - [Git provenance schema](git-provenance-schema/)

src/linkml/ontology/datalad.yaml

@@ -39,3 +39,5 @@ classes:
           distinguishing feature between a DataLad dataset and
           a plain Git or git-annex repository branch.
         required: true
+    close_mappings:
+      - dlco:Dataset

src/linkml/schemas/datalad-dataset-components.yaml

@@ -0,0 +1,98 @@
+id: https://concepts.datalad.org/schemas/datalad-dataset-components
+name: datalad-dataset-components
+title: Schema for any DataLad dataset component
+description: |
+  This specification defines a schema for describing a DataLad dataset
+  via a flat list of linked metadata objects. Each metadata object
+  has a simple structure with minimal property nesting.
+
+  Each object is of type `ComponentSE`, a class that has two required slots:
+  `meta_id` and `meta_type`. The former must be a unique object identifier,
+  and the latter must identify a subclass of `ComponentSE` that the object
+  represents. Both value must be given as a CURIE.
+
+  Nesting of objects is avoided, unless a DataLad dataset contains insufficient
+  information to guarantee the availability of a unique identifier for a given
+  entity.
+
+  Subclasses of `ComponentSE` represent all recognized dataset components.
+  These classes, within this schema, only define the nature of the employed
+  identifier, and the linkage to other dataset components. All other
+  properties are defined by a corresponding ontology class that is applied
+  as a "mixin".
+
+  An additional `ContainerSE` class is provided. It has a single slot
+  `components`, and can be used to represent a list of components for formats
+  and use cases that require a root class/instance.
+
+prefixes:
+  annex: https://concepts.datalad.org/namespace/annex-uuid/
+  datalad: https://concepts.datalad.org/namespace/dataset-uuid/
+  DCAT: http://www.w3.org/ns/dcat#
+  dcterms: http://purl.org/dc/terms/
+  dlco: https://concepts.datalad.org/ontology/
+  dlccs: https://concepts.datalad.org/schemas/datalad-dataset-components/
+  gitsha: https://concepts.datalad.org/namespace/gitsha/
+  linkml: https://w3id.org/linkml/
+  prov: http://www.w3.org/ns/prov#
+  spdx: http://spdx.org/rdf/terms#
+
+imports:
+  - ../ontology/meta_utils
+  - ../ontology/datalad
+  - ../ontology/git-annex
+
+classes:
+  ContainerSE:
+    class_uri: dlccs:Container
+    description: >-
+      A container for dataset component objects.
+    tree_root: true
+    attributes:
+      components:
+        description: >-
+          Component list.
+        multivalued: true
+        inlined_as_list: true
+        range: ComponentSE
+
+  ComponentSE:
+    class_uri: dlccs:Component
+    description: >-
+      Base class for any recognized dataset component type. This class
+      should never be used directly, only its subclasses.
+    slots:
+      - meta_id
+      - meta_type
+    slot_usage:
+      meta_type:
+        required: true
+
+  DataladDatasetVersionSE:
+    class_uri: dlccs:DataladDatasetVersionSE
+    is_a: ComponentSE
+    description: >-
+      TODO
+    mixins:
+      - DataladDatasetVersion
+    slot_usage:
+      is_version_of:
+        inlined: false
+        range: DataladDatasetSE
+      meta_id:
+        equals_expression: "gitsha:{gitsha}"
+
+  DataladDatasetSE:
+    class_uri: dlccs:DataladDatasetSE
+    is_a: ComponentSE
+    description: >-
+      Schema element for a `DataladDataset`.
+    see_also: 
+      - dlco:DataladDataset
+    comments:
+      - The required identifier format is `datalad:<dataset-uuid>`.
+    mixins:
+      - DataladDataset
+    slot_usage:
+      meta_id:
+        equals_expression: "datalad:{uuid}"

src/linkml/schemas/ontology.yaml

@@ -1,4 +1,4 @@
-id: https://concepts.datalad.org/schemas/ontology
+id: https://concepts.datalad.org/ontology
 name: datalad-concepts-ontology
 title: DataLad Concepts Ontology (DLCO)
 description: |-

tests/datalad-dataset-components-schema/validation/Container-DataladDatasetSE-minimal.yaml

@@ -0,0 +1,5 @@
+components:
+  - meta_id: datalad:0b76362c-aa27-11ee-be29-b3b123281259
+    meta_type: dlccs:DataladDataset
+    #meta_type: DataladDatasetSE
+    uuid: 0b76362c-aa27-11ee-be29-b3b123281259

tests/datalad-dataset-components-schema/validation/ContainerSE.valid.cfg.yaml

@@ -0,0 +1,10 @@
+schema: src/linkml/schemas/datalad-dataset-components.yaml
+target_class: ContainerSE
+data_sources:
+  - src/examples/datalad-dataset-components/ContainerSE-DataladDataset-minimal.yaml
+  - src/examples/datalad-dataset-components/ContainerSE-DataladDatasetVersion-linkage.yaml
+plugins:
+  JsonschemaValidationPlugin:
+    closed: true
+    include_range_class_descendants: false
+  RecommendedSlotsPlugin: