Merge pull request #42 from biolink/add_resources_to_automation

add unsecret to the automated run, move weekly URL check that inevita…

.github/workflows/weekly-generate-sankey.yaml

@@ -0,0 +1,42 @@
+name: Check Infores Catalog Links and Regenerate Artifacts
+on:
+  schedule:
+    - cron: '0 8 * * 1'  # at 08:00 UTC every Monday
+  workflow_dispatch:
+    types: trigger-url-check
+
+
+jobs:
+  check-external-links:
+    runs-on: ubuntu-latest
+    steps:
+      - name: Checkout
+        uses: actions/[email protected]
+        with:
+          fetch-depth: 0
+
+      - name: Set up Python 3.11
+        uses: actions/setup-python@v4
+        with:
+          python-version: '3.11'
+
+      - name: Install dependencies
+        run: |
+          pip install --upgrade pip
+          pip install poetry
+          poetry install 
+
+      - name: Run URL validation and regenerate sankey diagram
+        run: |
+          make sankey
+          if [ $? -eq 0 ]; then
+            git config --local user.email "[email protected]"
+            git config --local user.name "GitHub Action"
+            git add project/*
+            git add src/*
+            git add infores_catalog.yaml
+            git commit -m "Regenerate data_flow diagram and validate URLs in infores_catalog.yaml"
+            git push
+          else
+            echo "'make' validation and sankey generation failed to finish successfully"
+          fi

project/information_resource_registry.py

@@ -1,5 +1,5 @@
 # Auto generated from information_resource_registry.yaml by pythongen.py version: 0.0.1
-# Generation date: 2024-10-28T22:09:38
+# Generation date: 2024-10-28T15:22:54
 # Schema: Information-Resource-Registry-Schema
 #
 # id: https://w3id.org/biolink/information_resource_registry.yaml

project/jsonld/information_resource_registry.context.jsonld

@@ -1,7 +1,7 @@
 {
    "comments": {
       "description": "Auto generated by LinkML jsonld context generator",
-      "generation_date": "2024-10-28T22:09:38",
+      "generation_date": "2024-10-28T15:22:54",
       "source": "information_resource_registry.yaml"
    },
    "@context": {

project/jsonld/information_resource_registry.jsonld

@@ -647,9 +647,9 @@
   ],
   "metamodel_version": "1.7.0",
   "source_file": "information_resource_registry.yaml",
-  "source_file_date": "2024-10-28T22:08:39",
+  "source_file_date": "2024-09-18T16:41:40",
   "source_file_size": 11239,
-  "generation_date": "2024-10-28T22:09:38",
+  "generation_date": "2024-10-28T15:22:54",
   "@type": "SchemaDefinition",
   "@context": [
     "project/jsonld/information_resource_registry.context.jsonld",

project/owl/information_resource.owl.ttl

@@ -64,71 +64,71 @@ biolink:information_resource_registry.yaml.owl.ttl a owl:Ontology ;
 infores:InformationResource a owl:Class ;
     rdfs:label "InformationResource" ;
     rdfs:subClassOf [ a owl:Restriction ;
-            owl:allValuesFrom infores:InformationResourceStatusEnum ;
-            owl:onProperty infores:status ],
+            owl:allValuesFrom infores:AgentTypeEnum ;
+            owl:onProperty infores:agent_type ],
         [ a owl:Restriction ;
-            owl:allValuesFrom xsd:string ;
-            owl:onProperty infores:name ],
+            owl:minCardinality 0 ;
+            owl:onProperty infores:synonym ],
         [ a owl:Restriction ;
             owl:maxCardinality 1 ;
-            owl:onProperty infores:knowledge_level ],
+            owl:onProperty infores:status ],
         [ a owl:Restriction ;
             owl:minCardinality 0 ;
             owl:onProperty infores:status ],
         [ a owl:Restriction ;
             owl:maxCardinality 1 ;
-            owl:onProperty infores:name ],
+            owl:onProperty infores:agent_type ],
         [ a owl:Restriction ;
-            owl:minCardinality 0 ;
-            owl:onProperty infores:description ],
+            owl:minCardinality 1 ;
+            owl:onProperty infores:id ],
         [ a owl:Restriction ;
             owl:minCardinality 0 ;
-            owl:onProperty infores:knowledge_level ],
+            owl:onProperty infores:description ],
         [ a owl:Restriction ;
-            owl:allValuesFrom infores:KnowledgeLevelEnum ;
+            owl:maxCardinality 1 ;
             owl:onProperty infores:knowledge_level ],
         [ a owl:Restriction ;
-            owl:allValuesFrom infores:AgentTypeEnum ;
-            owl:onProperty infores:agent_type ],
+            owl:allValuesFrom infores:InformationResourceStatusEnum ;
+            owl:onProperty infores:status ],
         [ a owl:Restriction ;
             owl:maxCardinality 1 ;
-            owl:onProperty infores:agent_type ],
-        [ a owl:Restriction ;
-            owl:minCardinality 1 ;
-            owl:onProperty infores:id ],
-        [ a owl:Restriction ;
-            owl:minCardinality 0 ;
-            owl:onProperty infores:xref ],
+            owl:onProperty infores:description ],
         [ a owl:Restriction ;
-            owl:minCardinality 0 ;
-            owl:onProperty infores:synonym ],
+            owl:allValuesFrom xsd:string ;
+            owl:onProperty infores:name ],
         [ a owl:Restriction ;
             owl:maxCardinality 1 ;
-            owl:onProperty infores:id ],
+            owl:onProperty infores:name ],
         [ a owl:Restriction ;
             owl:allValuesFrom xsd:string ;
             owl:onProperty infores:id ],
         [ a owl:Restriction ;
-            owl:allValuesFrom xsd:string ;
-            owl:onProperty infores:synonym ],
+            owl:maxCardinality 1 ;
+            owl:onProperty infores:id ],
         [ a owl:Restriction ;
-            owl:minCardinality 0 ;
-            owl:onProperty infores:agent_type ],
+            owl:allValuesFrom infores:KnowledgeLevelEnum ;
+            owl:onProperty infores:knowledge_level ],
         [ a owl:Restriction ;
             owl:allValuesFrom xsd:string ;
             owl:onProperty infores:xref ],
         [ a owl:Restriction ;
             owl:minCardinality 0 ;
             owl:onProperty infores:name ],
         [ a owl:Restriction ;
-            owl:maxCardinality 1 ;
-            owl:onProperty infores:status ],
-        [ a owl:Restriction ;
-            owl:maxCardinality 1 ;
+            owl:allValuesFrom xsd:string ;
             owl:onProperty infores:description ],
+        [ a owl:Restriction ;
+            owl:minCardinality 0 ;
+            owl:onProperty infores:knowledge_level ],
         [ a owl:Restriction ;
             owl:allValuesFrom xsd:string ;
-            owl:onProperty infores:description ],
+            owl:onProperty infores:synonym ],
+        [ a owl:Restriction ;
+            owl:minCardinality 0 ;
+            owl:onProperty infores:agent_type ],
+        [ a owl:Restriction ;
+            owl:minCardinality 0 ;
+            owl:onProperty infores:xref ],
         linkml:ClassDefinition ;
     skos:altLabel "knowledgebase" ;
     skos:definition "A database or knowledgebase and its supporting ecosystem of interfaces  and services that deliver content to consumers (e.g. web portals, APIs,  query endpoints, streaming services, data downloads, etc.). A single Information Resource by this definition may span many different datasets or databases, and include many access endpoints and user interfaces. Information Resources include project-specific resources such as a Translator Knowledge Provider, and community knowledgebases like ChemBL, OMIM, or DGIdb." ;

project/shacl/information_resource_registry.shacl.ttl

@@ -19,49 +19,49 @@ infores:InformationResource a sh:NodeShape ;
     sh:closed true ;
     sh:description "A database or knowledgebase and its supporting ecosystem of interfaces  and services that deliver content to consumers (e.g. web portals, APIs,  query endpoints, streaming services, data downloads, etc.). A single Information Resource by this definition may span many different datasets or databases, and include many access endpoints and user interfaces. Information Resources include project-specific resources such as a Translator Knowledge Provider, and community knowledgebases like ChemBL, OMIM, or DGIdb." ;
     sh:ignoredProperties ( rdf:type ) ;
-    sh:property [ sh:description "The type of agent that supports an edge or node.  This is a general categorization of the type of agent that supports a statement, and is not intended to be a comprehensive description of the agent.  For example, a statement may be supported by a single publication, but that publication may contain multiple types of evidence, such as a computational prediction and a manual curation.  In this case, the agent type would be \"publication\", and the evidence would be described in more detail in the evidence graph." ;
-            sh:in ( "manual_agent" "not_provided" "automated_agent" "data_analysis_pipeline" "computational_model" "text_mining_agent" "image_processing_agent" "manual_validation_of_automated_agent" ) ;
+    sh:property [ sh:description "The level of knowledge that supports an edge or node.  This is a general categorization of the type of evidence that supports a statement, and is not intended to be a comprehensive description of the evidence.  For example, a statement may be supported by a single publication, but that publication may contain multiple types of evidence, such as a computational prediction and a manual curation.  In this case, the knowledge level would be \"curated\", and the evidence would be described in more detail in the evidence graph." ;
+            sh:in ( "knowledge_assertion" "statistical_association" "prediction" "observation" "not_provided" "logical_entailment" "mixed" "other" ) ;
             sh:maxCount 1 ;
-            sh:order 7 ;
-            sh:path infores:agent_type ],
+            sh:order 6 ;
+            sh:path infores:knowledge_level ],
+        [ sh:datatype xsd:string ;
+            sh:description "A free-text description of an entity or attribute." ;
+            sh:maxCount 1 ;
+            sh:nodeKind sh:Literal ;
+            sh:order 5 ;
+            sh:path rdfs:comment ],
         [ sh:datatype xsd:string ;
             sh:description "A database cross reference or alternative identifier for a NamedThing or edge between two  NamedThings.  This property should point to a database record or webpage that supports the existence of the edge, or  gives more detail about the edge. This property can be used on a node or edge to provide multiple URIs or CURIE cross references." ;
             sh:nodeKind sh:Literal ;
             sh:order 3 ;
             sh:path infores:xref ],
+        [ sh:description "The type of agent that supports an edge or node.  This is a general categorization of the type of agent that supports a statement, and is not intended to be a comprehensive description of the agent.  For example, a statement may be supported by a single publication, but that publication may contain multiple types of evidence, such as a computational prediction and a manual curation.  In this case, the agent type would be \"publication\", and the evidence would be described in more detail in the evidence graph." ;
+            sh:in ( "manual_agent" "not_provided" "automated_agent" "data_analysis_pipeline" "computational_model" "text_mining_agent" "image_processing_agent" "manual_validation_of_automated_agent" ) ;
+            sh:maxCount 1 ;
+            sh:order 7 ;
+            sh:path infores:agent_type ],
         [ sh:datatype xsd:string ;
-            sh:description "A unique identifier for an entity. Must be either a CURIE shorthand for a URI or a complete URI" ;
+            sh:description "A human-readable name for an attribute or entity." ;
             sh:maxCount 1 ;
-            sh:minCount 1 ;
             sh:nodeKind sh:Literal ;
-            sh:order 2 ;
-            sh:path infores:id ],
+            sh:order 1 ;
+            sh:path rdfs:label ],
         [ sh:datatype xsd:string ;
             sh:description "Alternate human-readable names for a thing" ;
             sh:nodeKind sh:Literal ;
             sh:order 4 ;
             sh:path infores:synonym ],
-        [ sh:description "the status of the infores identifier, the default is \"released\"" ;
-            sh:in ( "released" "deprecated" "draft" "modified" ) ;
-            sh:maxCount 1 ;
-            sh:order 0 ;
-            sh:path infores:status ],
-        [ sh:description "The level of knowledge that supports an edge or node.  This is a general categorization of the type of evidence that supports a statement, and is not intended to be a comprehensive description of the evidence.  For example, a statement may be supported by a single publication, but that publication may contain multiple types of evidence, such as a computational prediction and a manual curation.  In this case, the knowledge level would be \"curated\", and the evidence would be described in more detail in the evidence graph." ;
-            sh:in ( "knowledge_assertion" "statistical_association" "prediction" "observation" "not_provided" "logical_entailment" "mixed" "other" ) ;
-            sh:maxCount 1 ;
-            sh:order 6 ;
-            sh:path infores:knowledge_level ],
         [ sh:datatype xsd:string ;
-            sh:description "A free-text description of an entity or attribute." ;
+            sh:description "A unique identifier for an entity. Must be either a CURIE shorthand for a URI or a complete URI" ;
             sh:maxCount 1 ;
+            sh:minCount 1 ;
             sh:nodeKind sh:Literal ;
-            sh:order 5 ;
-            sh:path rdfs:comment ],
-        [ sh:datatype xsd:string ;
-            sh:description "A human-readable name for an attribute or entity." ;
+            sh:order 2 ;
+            sh:path infores:id ],
+        [ sh:description "the status of the infores identifier, the default is \"released\"" ;
+            sh:in ( "released" "deprecated" "draft" "modified" ) ;
             sh:maxCount 1 ;
-            sh:nodeKind sh:Literal ;
-            sh:order 1 ;
-            sh:path rdfs:label ] ;
+            sh:order 0 ;
+            sh:path infores:status ] ;
     sh:targetClass infores:InformationResource .
 

pyproject.toml

@@ -23,7 +23,7 @@ extract_consume_info = "information_resource_registry.translator_dataflow.extrac
 generate_sankey = "information_resource_registry.translator_dataflow.generate_sankey:main"
 molepro_consume = "information_resource_registry.translator_dataflow.molepro_consume:main"
 rtx_kg2 = "information_resource_registry.translator_dataflow.rtx_kg2:main"
-
+unsecret = "information_resource_registry.translator_dataflow.unsecret:main"
 
 [tool.poetry-dynamic-versioning]
 enable = true

src/information_resource_registry/datamodel/information_resource_registry.py

@@ -1,5 +1,5 @@
 # Auto generated from information_resource_registry.yaml by pythongen.py version: 0.0.1
-# Generation date: 2024-10-28T22:09:42
+# Generation date: 2024-10-28T15:23:02
 # Schema: Information-Resource-Registry-Schema
 #
 # id: https://w3id.org/biolink/information_resource_registry.yaml

src/information_resource_registry/translator_dataflow/unsecret.py

@@ -17,7 +17,7 @@ def main():
         print(f"Error downloading file: {response.status_code}")
         exit()
 
-    json.dump(infores_dict, open('data/unsecret.json', 'w'), indent=2, sort_keys=True)
+    json.dump(infores_dict, open('src/information_resource_registry/translator_dataflow/data/unsecret.json', 'w'), indent=2, sort_keys=True)
 
 # Call main() if this script is executed directly
 if __name__ == "__main__":

translator_dataflow.Makefile

@@ -19,7 +19,8 @@ generate_consumes_annotations:
 	$(RUN) rtx_kg2
 	$(RUN) bte_sp
 	$(RUN) arax_kps
-	$(RUN) aragorn_kps  # currently does not run b/c of pydantic v1 requirement
+	$(RUN) aragorn_kps
+	$(RUN) unsecret
 	$(RUN) molepro_consume
 
 merge_into_infores_catalog: