Merge pull request #33 from w3c/jsonld-postprocessing

JSON-LD @context postprocessing

main.py

@@ -1,4 +1,5 @@
 import argparse
+import json
 import subprocess
 
 from linkml.generators.jsonschemagen import JsonSchemaGenerator
@@ -7,7 +8,12 @@
 from linkml.generators.jsonldcontextgen import ContextGenerator
 from linkml.generators.docgen import DocGenerator
 from linkml.generators.linkmlgen import LinkmlGenerator
+from linkml_runtime.utils.schemaview import SchemaView
+from linkml_runtime.linkml_model.meta import AnonymousSlotExpression
 from pathlib import Path
+from pyld import jsonld
+
+from linkml_runtime.linkml_model.meta import SlotDefinition
 
 RESOURCES_PATH = Path('resources')
 GENS_PATH = RESOURCES_PATH / 'gens'
@@ -25,28 +31,102 @@ def generate_docs():
     doc_generator = DocGenerator(YAML_SCHEMA_PATH, mergeimports=False)
     doc_generator.serialize(directory=str(DOCDIR))
 
+
+#camelCase conversion of class names in the generated context file
+    # updated_context = {}
+    # for cn, value in generated_context.items():
+    #     camel_case_key = cn[0].lower() + cn[1:]
+    #     if value != {}:
+    #         updated_context[camel_case_key] = value
+    # serialized_schema_json['@context'] = updated_context
+    # generated_context = updated_context
+def post_process_jsonld_context(schema_view: SchemaView, serialized_schema: str) -> str:
+    serialized_schema_json = json.loads(serialized_schema)
+    generated_context = serialized_schema_json.get('@context', {})
+    default_range = schema_view.schema.default_range if schema_view.schema.default_range else "string"
+    for slot in schema_view.all_slots().values():
+        slot_name = slot.name
+        context_entry = generated_context.get(slot_name, {})
+        # Update JSON-LD context for slots with multi-language support
+        is_langstring = slot.range == 'langString'
+        is_exactly_one_language = (
+            slot.range is None and
+            any(
+                expr.range == 'langString'
+                for expr in slot.exactly_one_of if isinstance(expr, AnonymousSlotExpression)
+            )
+        )
+        if is_langstring or is_exactly_one_language and isinstance(context_entry, dict):
+            serialized_schema_json['@context'][slot_name]['@container'] = '@language'
+            if '@type' in serialized_schema_json['@context'][slot_name].keys():
+                del serialized_schema_json['@context'][slot_name]['@type']
+        # Update JSON-LD context for slots with a specific encoded language
+        if slot.in_language and isinstance(context_entry, dict):
+            context_entry['@language'] = slot.in_language
+        # inlined and multivalued slot conditions are used to identify dictionaries
+        if slot.inlined and slot.multivalued and not slot.inlined_as_list and isinstance(context_entry, dict):
+            context_entry['@container'] = '@index'
+            if not hasattr(context_entry, '@type'):
+                context_entry['@type'] = '@id'
+            if slot.instantiates:
+                context_entry['@index'] = slot.instantiates
+        #exactly_one_of expressions
+        if hasattr(slot, 'exactly_one_of') and slot.exactly_one_of:
+            ranges = [opt['range'] for opt in slot.exactly_one_of if 'range' in opt]
+            if len(set(ranges)) == 1:
+                range_type = ranges[0]
+                context_entry["@type"] = f"xsd:{range_type}"
+            else:
+                print(f"Warning: Slot {slot_name} has different ranges")
+        elif slot.range == default_range:
+            context_entry["@type"] = f"xsd:{default_range}"
+        #Change property name with those that provide aliases
+        # if slot.aliases and isinstance(context_entry, dict):
+        #     generated_context[slot.aliases[0]] = generated_context.pop(slot_name)
+        #     context_entry = generated_context[slot.aliases[0]]
+        generated_context[slot_name] = context_entry
+    #The multivalued slots which do not already have a @container are assigned to a @set
+    for slot in schema_view.all_slots().values():
+        context_entry = generated_context.get(slot.name, {})
+        if slot.multivalued and isinstance(context_entry, dict) and '@container' not in context_entry:
+            context_entry['@container'] = '@set'
+            context_entry.pop('@type', None)
+        generated_context[slot.name] = context_entry
+    # for c in schema_view.all_classes().values():
+    #     class_name = c.name
+    #     context_entry = generated_context.get(class_name, {})
+    #     if c.aliases and isinstance(context_entry, dict):
+    #         generated_context[c.aliases[0]] = generated_context.pop(class_name)
+    #         context_entry = generated_context[c.aliases[0]]
+    #     generated_context[class_name] = context_entry
+    return json.dumps(serialized_schema_json, indent=3)
+
+
 def main(generate_docs_flag, serve_docs_flag):
     if not YAML_SCHEMA_PATH.exists():
         print(f"LinkML schema file does not exist: {YAML_SCHEMA_PATH}")
         return
+    linkml_schema_view = SchemaView(YAML_SCHEMA_PATH, merge_imports=True)
+    # TODO: add pre processing for LinkML if needed
     for generator in GENERATORS:
         output_dir = GENS_PATH / generator
         output_dir.mkdir(parents=True, exist_ok=True)
         if generator == 'jsonschema':
             # json_schema_generator = JsonSchemaGenerator(yaml_content, top_class="Thing")
-            json_schema_generator = JsonSchemaGenerator(YAML_SCHEMA_PATH, mergeimports=True)
+            json_schema_generator = JsonSchemaGenerator(linkml_schema_view.schema, mergeimports=True)
             (output_dir / 'jsonschema.json').write_text(json_schema_generator.serialize())
         elif generator == 'shacl':
-            shacl_generator = ShaclGenerator(YAML_SCHEMA_PATH, mergeimports=False, closed=True, suffix='Shape')
+            shacl_generator = ShaclGenerator(linkml_schema_view.schema, mergeimports=False, closed=True, suffix='Shape')
             (output_dir / 'shapes.shacl.ttl').write_text(shacl_generator.serialize())
         elif generator == 'owl':
-            owl_generator = OwlSchemaGenerator(YAML_SCHEMA_PATH,)
+            owl_generator = OwlSchemaGenerator(linkml_schema_view.schema,)
             (output_dir / 'ontology.owl.ttl').write_text(owl_generator.serialize())
         elif generator == 'jsonldcontext':
-            context_generator = ContextGenerator(YAML_SCHEMA_PATH, mergeimports=False)
-            (output_dir / 'context.jsonld').write_text(context_generator.serialize())
+            context_generator = ContextGenerator(linkml_schema_view.schema, mergeimports=True)
+            (output_dir / 'context.jsonld').write_text(post_process_jsonld_context(linkml_schema_view,
+                                                                                   context_generator.serialize()))
         elif generator == 'linkml':
-            linkml_generator = LinkmlGenerator(YAML_SCHEMA_PATH, mergeimports=True, format='yaml', output='linkml.yaml')
+            linkml_generator = LinkmlGenerator(linkml_schema_view.schema, mergeimports=True, format='yaml', output='linkml.yaml')
             (output_dir / 'linkml.yaml').write_text(linkml_generator.serialize())
         else:
             print(f"Unknown generator: {generator}")
@@ -69,3 +149,4 @@ def main(generate_docs_flag, serve_docs_flag):
                         help='Boolean for serving the generated documentation.')
     args = parser.parse_args()
     main(args.local_docs, args.serve_docs)
+

pyproject.toml

@@ -11,7 +11,7 @@ include = ["README.md", "src/thing_description_schema/schema", "project"]
 
 requires-python = ">=3.12"
 dependencies = [
-    "linkml-runtime>=1.8.1",
+    "linkml-runtime>=1.8.2",
     "linkml>=1.8.2",
     "mkdocs-mermaid2-plugin>=1.1.1",
     "mkdocs-material>=9.5.32",
@@ -23,4 +23,4 @@ requires = ["hatchling"]
 build-backend = "hatchling.build"
 
 [tool.hatch.build]
-only-include = ["main.py"]
+only-include = ["main.py"]

resources/schemas/hctl.yaml

@@ -4,6 +4,7 @@ title: hctl
 version: "1.1-11-June-2024"
 description: |-
   LinkML schema for modelling the TD Hypermedia Control information model, in particular links and forms.
+contributors: Mahda_Noura
 license: MIT
 see_also:
   - https://www.w3.org/TR/wot-thing-description11/
@@ -40,26 +41,32 @@ classes:
   Link:
     class_uri: hctl:Link
     description: >-
-      A link can be viewed as a statement of the form link context that has a relation type resource at link target, where the optional target attributes may further describe the resource.
+      A link can be viewed as a statement of the form link context that has a relation type resource at link target, 
+      where the optional target attributes may further describe the resource.
     attributes:
       type:
         description: Target attribute providing a hint indicating what the media type [IANA-MEDIA-TYPES] of the result of dereferencing the link should be.
         slot_uri: hctl:hintsAtMediaType
-      relation:
+      rel:
         description: >-
           A link relation type identifies the semantics of a link.
+        slot_uri: hctl:hasRelationType
       anchor:
         description: >-
-          By default, the context, or anchor, of a link conveyed in the Link header field is the URL of the representation it is associated with, as defined in RFC7231, Section 3.1.4.1, and is serialized as a URI.
-        range: uriorcurie
+          By default, the context, or anchor, of a link conveyed in the Link header field is the URL of the 
+          representation it is associated with, as defined in RFC7231, Section 3.1.4.1, and is serialized as a URI.
+        slot_uri: hctl:hasAnchor
+        range: uri
       sizes:
         description: >-
-          Target attribute that specifies one or more sizes for the referenced icon. Only applicable for relation type 'icon'. The value pattern follows {Height}x{Width} (e.g., \"16x16\", \"16x16 32x32\").
+          Target attribute that specifies one or more sizes for the referenced icon. Only applicable for relation type 
+          'icon'. The value pattern follows {Height}x{Width} (e.g., \"16x16\", \"16x16 32x32\").
+        slot_uri: hctl:hasSizes
       hreflang:
         description: >-
-          The hreflang attribute specifies the language of a linked document. The value of this must be a valid language tag [[BCP47]].
-        multivalued: true
-        pattern: "^(((([A-Za-z]{2,3}(-([A-Za-z]{3}(-[A-Za-z]{3}){0,2}))?)|[A-Za-z]{4}|[A-Za-z]{5,8})(-([A-Za-z]{4}))?(-([A-Za-z]{2}|[0-9]{3}))?(-([A-Za-z0-9]{5,8}|[0-9][A-Za-z0-9]{3}))*(-([0-9A-WY-Za-wy-z](-[A-Za-z0-9]{2,8})+))*(-(x(-[A-Za-z0-9]{1,8})+))?)|(x(-[A-Za-z0-9]{1,8})+)|((en-GB-oed|i-ami|i-bnn|i-default|i-enochian|i-hak|i-klingon|i-lux|i-mingo|i-navajo|i-pwn|i-tao|i-tay|i-tsu|sgn-BE-FR|sgn-BE-NL|sgn-CH-DE)|(art-lojban|cel-gaulish|no-bok|no-nyn|zh-guoyu|zh-hakka|zh-min|zh-min-nan|zh-xiang)))$"
+          The hreflang attribute specifies the language of a linked document. The value of this must be a valid 
+          language tag [[BCP47]].
+        slot_uri: hctl:hasHreflang
     slots:
       - href
   Form:
@@ -81,20 +88,13 @@ classes:
           Content codings are primarily used to allow a representation to be compressed or otherwise usefully transformed 
           without losing the identity of its underlying media type and without loss of information. Examples of content
           coding include \"gzip\", \"deflate\", etc.
-      securityDefinitions:
-        description: >-
-          Set of security definition names, chosen from those defined in securityDefinitions. These must all be satisfied for access to resources.
-        slot_uri: td:hasSecurityConfiguration
-        exactly_one_of:
-          - range: string
-          - range: string
-            multivalued: true
+        slot_uri: hctl:forContentCoding
       scopes:
         description: >-
-          TODO Check, was not in hctl ontology, if not could be source of discrepancy.
           Set of authorization scope identifiers provided as an array. These are provided in tokens returned by an 
           authorization server and associated with forms in order to identify what resources a client may access and how. 
           The values associated with a form SHOULD be chosen from those defined in an OAuth2SecurityScheme active on that form.
+        slot_uri: hctl:scopes
         exactly_one_of:
           - range: string
           - range: string
@@ -104,8 +104,8 @@ classes:
           This optional term can be used if, e.g., the output communication metadata differ from input metadata (e.g., output contentType differ from the
           input contentType). The response name contains metadata that is only valid for the response messages.
         slot_uri: hctl:returns
-        aliases:
-          - returns
+        exact_mappings:
+          - hctl:returns
         range: ExpectedResponse
       additionalResponse:
         description: >-
@@ -115,8 +115,8 @@ classes:
         slot_uri: hctl:additionalReturns
         multivalued: true
         range: AdditionalExpectedResponse
-        aliases:
-          - additional returns
+        exact_mappings:
+          - hctl:additionalReturns
       subprotocol:
         slot_uri: hctl:forSubProtocol
         description: >-
@@ -134,6 +134,7 @@ classes:
             multivalued: true
     slots:
       - href
+      - security
   ExpectedResponse:
     class_uri: hctl:ExpectedResponse
     description: >-
@@ -174,10 +175,10 @@ slots:
     description: >-
       Target IRI of a link or submission target of a Form.
     slot_uri: hctl:hasTarget
-    aliases:
-      - target
     required: true
-    range: uriorcurie
+    range: uri
+    exact_mappings:
+      - hctl:target
 
 enums:
   SubProtocolTypes: