Refactoring sparql templates for harvest optimization

pom.xml

@@ -45,13 +45,11 @@
 
   <dependencies>
 
-  <!-- https://mvnrepository.com/artifact/org.springframework.boot/spring-boot-devtools -->
-<dependency>
-    <groupId>org.springframework.boot</groupId>
-    <artifactId>spring-boot-devtools</artifactId>
-</dependency>
-
-
+    <!-- <dependency>
+        <groupId>org.springframework.boot</groupId>
+        <artifactId>spring-boot-devtools</artifactId>
+        <optional>true</optional>
+    </dependency> -->
     <dependency>
       <groupId>org.springframework.boot</groupId>
       <artifactId>spring-boot-configuration-processor</artifactId>

src/main/java/edu/tamu/scholars/middleware/discovery/service/component/jena/TriplestoreHarvester.java

@@ -75,13 +75,52 @@ public Flux<AbstractIndexDocument> harvest() {
         QueryExecution queryExecution = triplestore.createQueryExecution(query);
         Iterator<Triple> tripleIterator = queryExecution.execConstructTriples();
         Iterable<Triple> triples = () -> tripleIterator;
+
         return Flux.fromIterable(triples)
             .map(this::subject)
             .map(this::harvest)
             .doFinally(onFinally -> queryExecution.close());
     }
 
     public AbstractIndexDocument harvest(String subject) {
+        //
+
+        // sparql variable addition for jit inference
+
+        // scholars-discovery currently confined to adding only uri which can be a subject, predicate, or literal
+        // leaf nodes should have literals or pulling up intermediate nodes without any literals (flattening the predicatation)
+        // graph pruning process should strip all triples which do not go from a top level class identifier to property literal (sanatize and normalize types)
+
+        // Find language identifier API. provide text until it provides a confidence interval above 98%
+
+        // annotations within the discovery model encodes the minimal knowledge required to provide the requested views by product owner
+        // doing so provides the nested structure and index behavior
+
+        // top level classes of the VIVO ontology have been identified as the discovery model
+
+
+        // logical branch ?1
+        // regex capture {{}} and control group
+        // if within construct clause
+        // else within where clause
+
+
+        // refactor all construct sparql to use ?subject in its single construct clause
+        // regex extract single construct clause variable name
+        // use variable name prefix with ? for {{uri}}
+
+        // number of {{}} interpolations for logical constraints
+
+        // logical branch ?2
+        // if one occurance of {{}} then {}
+        // else if two occurances of {{}} then {}
+        // else if three occurances of {{}} then {}
+        // else {}
+
+        // backward compatability requires removing URI notation from template and wrap inputs with it if necessary
+
+        // query for all of a property first
+        // by passing in the subject of the construct {{subject}} with ?subject
         try {
             return createDocument(subject);
         } catch (Exception e) {
@@ -116,7 +155,12 @@ private void lookupProperties(AbstractIndexDocument document, String subject) {
             try {
                 FieldSource source = typeOp.getPropertySource();
                 Model model = queryForModel(source, subject);
+
+
+
+
                 List<Object> values = lookupProperty(typeOp, source, model);
+
                 populate(document, typeOp.getField(), values);
             } catch (Exception e) {
                 logger.error(String.format("Unable to populate document %s: %s", name(), parse(subject)));

src/main/java/edu/tamu/scholars/middleware/service/TemplateService.java

@@ -56,10 +56,12 @@ public String template(String template, Object data) {
         }
     }
 
-    public String templateSparql(String name, String uri) {
+    public String templateSparql(String name, String value) {
         String path = String.format("templates/sparql/%s.sparql", name);
         Map<String, String> data = new HashMap<String, String>();
-        data.put("uri", uri);
+        // cost of increasing context variables?
+        data.put("uri", value);
+        data.put("subject", value);
         Context context = Context.newBuilder(data).build();
         try {
             return handlebars.compileInline(resourceService.getTemplate(path)).apply(context);

src/main/resources/templates/sparql/collection/abbreviation.sparql

@@ -1,7 +1,7 @@
 PREFIX vivo: <http://vivoweb.org/ontology/core#>
 
 CONSTRUCT {
-    <{{uri}}> vivo:abbreviation ?abbreviation .
+    {{subject}} vivo:abbreviation ?abbreviation .
 } WHERE {
-    <{{uri}}> vivo:abbreviation ?abbreviation .
+    {{subject}} vivo:abbreviation ?abbreviation .
 }

src/main/resources/templates/sparql/collection/abstract.sparql

@@ -1,7 +1,7 @@
 PREFIX bibo: <http://purl.org/ontology/bibo/>
 
 CONSTRUCT {
-    <{{uri}}> bibo:abstract ?abstract .
+    {{subject}} bibo:abstract ?abstract .
 } WHERE {
-    <{{uri}}> bibo:abstract ?abstract .
+    {{subject}} bibo:abstract ?abstract .
 }

src/main/resources/templates/sparql/collection/author.sparql

@@ -6,7 +6,7 @@ PREFIX rdf: <http://www.w3.org/1999/02/22-rdf-syntax-ns#>
 CONSTRUCT {
     ?author rdfs:label ?labelWithId .
 } WHERE {
-    <{{uri}}> vivo:relatedBy ?relatedBy .
+    {{subject}} vivo:relatedBy ?relatedBy .
     ?relatedBy rdf:type vivo:Authorship .
     ?relatedBy vivo:relates ?author .
     ?author rdf:type ?type .

src/main/resources/templates/sparql/collection/doi.sparql

@@ -1,7 +1,7 @@
 PREFIX bibo: <http://purl.org/ontology/bibo/>
 
 CONSTRUCT {
-    <{{uri}}> bibo:doi ?doi .
+    {{subject}} bibo:doi ?doi .
 } WHERE {
-    <{{uri}}> bibo:doi ?doi .
+    {{subject}} bibo:doi ?doi .
 }

src/main/resources/templates/sparql/collection/editor.sparql

@@ -6,7 +6,7 @@ PREFIX rdf: <http://www.w3.org/1999/02/22-rdf-syntax-ns#>
 CONSTRUCT {
     ?editor rdfs:label ?labelWithId .
 } WHERE {
-    <{{uri}}> vivo:relatedBy ?relatedBy .
+    {{subject}} vivo:relatedBy ?relatedBy .
     ?relatedBy a vivo:Editorship .
     ?relatedBy vivo:relates ?editor .
     ?editor rdf:type ?type .

src/main/resources/templates/sparql/collection/eissn.sparql

@@ -1,7 +1,7 @@
 PREFIX bibo: <http://purl.org/ontology/bibo/>
 
 CONSTRUCT {
-    <{{uri}}> bibo:eissn ?eissn .
+    {{subject}} bibo:eissn ?eissn .
 } WHERE {
-    <{{uri}}> bibo:eissn ?eissn .
+    {{subject}} bibo:eissn ?eissn .
 }

src/main/resources/templates/sparql/collection/feature.sparql

@@ -4,7 +4,7 @@ PREFIX rdfs: <http://www.w3.org/2000/01/rdf-schema#>
 CONSTRUCT {
     ?feature rdfs:label ?labelWithId .
 } WHERE {
-    <{{uri}}> vivo:features ?feature .
+    {{subject}} vivo:features ?feature .
     ?feature rdfs:label ?label .
     BIND( CONCAT( STR(?label), REPLACE(STR(?feature), "(^.*/)", "::") ) AS ?labelWithId ) .
 }

src/main/resources/templates/sparql/collection/hasSubjectArea.sparql

@@ -4,7 +4,7 @@ PREFIX rdfs: <http://www.w3.org/2000/01/rdf-schema#>
 CONSTRUCT {
     ?subjectArea rdfs:label ?labelWithId .
 } WHERE {
-    <{{uri}}> vivo:hasSubjectArea ?subjectArea .
+    {{subject}} vivo:hasSubjectArea ?subjectArea .
     ?subjectArea rdfs:label ?label .
     BIND( CONCAT( STR(?label), REPLACE(STR(?subjectArea), "(^.*/)", "::") ) AS ?labelWithId ) .
 }

src/main/resources/templates/sparql/collection/isAbout.sparql

@@ -4,7 +4,7 @@ PREFIX rdfs: <http://www.w3.org/2000/01/rdf-schema#>
 CONSTRUCT {
     ?isAbout rdfs:label ?labelWithId .
 } WHERE {
-    <{{uri}}> obo:IAO_0000136 ?isAbout .
+    {{subject}} obo:IAO_0000136 ?isAbout .
     ?isAbout rdfs:label ?label .
     BIND( CONCAT( STR(?label), REPLACE(STR(?isAbout), "(^.*/)", "::") ) AS ?labelWithId ) .
 }

src/main/resources/templates/sparql/collection/issn.sparql

@@ -1,7 +1,7 @@
 PREFIX bibo: <http://purl.org/ontology/bibo/>
 
 CONSTRUCT {
-    <{{uri}}> bibo:issn ?issn .
+    {{subject}} bibo:issn ?issn .
 } WHERE {
-    <{{uri}}> bibo:issn ?issn .
+    {{subject}} bibo:issn ?issn .
 }

src/main/resources/templates/sparql/collection/keyword.sparql

@@ -1,7 +1,7 @@
 PREFIX vivo: <http://vivoweb.org/ontology/core#>
 
 CONSTRUCT {
-    <{{uri}}> vivo:freetextKeyword ?freetextKeyword .
+    {{subject}} vivo:freetextKeyword ?freetextKeyword .
 } WHERE {
-    <{{uri}}> vivo:freetextKeyword ?freetextKeyword .
+    {{subject}} vivo:freetextKeyword ?freetextKeyword .
 }

src/main/resources/templates/sparql/collection/mention.sparql

@@ -4,7 +4,7 @@ PREFIX rdfs: <http://www.w3.org/2000/01/rdf-schema#>
 CONSTRUCT {
     ?mention rdfs:label ?labelWithId .
 } WHERE {
-    <{{uri}}> obo:IAO_0000142 ?mention .
+    {{subject}} obo:IAO_0000142 ?mention .
     ?mention rdfs:label ?label .
     BIND( CONCAT( STR(?label), REPLACE(STR(?mention), "(^.*/)", "::") ) AS ?labelWithId ) .
 }

src/main/resources/templates/sparql/collection/name.sparql

@@ -1,8 +1,8 @@
 PREFIX rdfs: <http://www.w3.org/2000/01/rdf-schema#>
 
 CONSTRUCT {
-    <{{uri}}> rdfs:label ?label .
+    {{subject}} rdfs:label ?label .
 } WHERE {
-    <{{uri}}> rdfs:label ?label .
+    {{subject}} rdfs:label ?label .
     FILTER (lang(?label) = '')
 }

src/main/resources/templates/sparql/collection/oclcnum.sparql

@@ -1,7 +1,7 @@
 PREFIX bibo: <http://purl.org/ontology/bibo/>
 
 CONSTRUCT {
-    <{{uri}}> bibo:oclcnum ?oclcnum .
+    {{subject}} bibo:oclcnum ?oclcnum .
 } WHERE {
-    <{{uri}}> bibo:oclcnum ?oclcnum .
+    {{subject}} bibo:oclcnum ?oclcnum .
 }

src/main/resources/templates/sparql/collection/outputOfProcessOrEvent.sparql

@@ -4,7 +4,7 @@ PREFIX rdfs: <http://www.w3.org/2000/01/rdf-schema#>
 CONSTRUCT {
     ?outputOfProcessOrEvent rdfs:label ?labelWithId .
 } WHERE {
-    <{{uri}}> obo:RO_0002353 ?outputOfProcessOrEvent .
+    {{subject}} obo:RO_0002353 ?outputOfProcessOrEvent .
     ?outputOfProcessOrEvent rdfs:label ?label .
     BIND( CONCAT( STR(?label), REPLACE(STR(?outputOfProcessOrEvent), "(^.*/)", "::") ) AS ?labelWithId ) .
 }

src/main/resources/templates/sparql/collection/participatesIn.sparql

@@ -4,7 +4,7 @@ PREFIX rdfs: <http://www.w3.org/2000/01/rdf-schema#>
 CONSTRUCT {
     ?participatesIn rdfs:label ?labelWithId .
 } WHERE {
-    <{{uri}}> obo:RO_0000056 ?participatesIn .
+    {{subject}} obo:RO_0000056 ?participatesIn .
     ?participatesIn rdfs:label ?label .
     BIND( CONCAT( STR(?label), REPLACE(STR(?participatesIn), "(^.*/)", "::") ) AS ?labelWithId ) .
 }

src/main/resources/templates/sparql/collection/publicationDate.sparql

@@ -3,6 +3,6 @@ PREFIX vivo: <http://vivoweb.org/ontology/core#>
 CONSTRUCT {
     ?publicationDate vivo:dateTime ?dateTime .
 } WHERE {
-    <{{uri}}> vivo:dateTimeValue ?publicationDate .
+    {{subject}} vivo:dateTimeValue ?publicationDate .
     ?publicationDate vivo:dateTime ?dateTime .
 }

src/main/resources/templates/sparql/collection/publicationVenueFor.sparql

@@ -4,7 +4,7 @@ PREFIX rdfs: <http://www.w3.org/2000/01/rdf-schema#>
 CONSTRUCT {
     ?publicationVenueFor rdfs:label ?labelWithId .
 } WHERE {
-    <{{uri}}> vivo:publicationVenueFor ?publicationVenueFor .
+    {{subject}} vivo:publicationVenueFor ?publicationVenueFor .
     ?publicationVenueFor rdfs:label ?label .
     BIND( CONCAT( STR(?label), REPLACE(STR(?publicationVenueFor), "(^.*/)", "::") ) AS ?labelWithId ) .
 }

src/main/resources/templates/sparql/collection/publisher.sparql

@@ -4,7 +4,7 @@ PREFIX rdfs: <http://www.w3.org/2000/01/rdf-schema#>
 CONSTRUCT {
     ?publisher rdfs:label ?labelWithId .
 } WHERE {
-    <{{uri}}> vivo:publisher ?publisher .
+    {{subject}} vivo:publisher ?publisher .
     ?publisher rdfs:label ?label .
     BIND( CONCAT( STR(?label), REPLACE(STR(?publisher), "(^.*/)", "::") ) AS ?labelWithId ) .
 }

src/main/resources/templates/sparql/collection/publisherType.sparql

@@ -4,7 +4,7 @@ PREFIX vitro: <http://vitro.mannlib.cornell.edu/ns/vitro/0.7#>
 CONSTRUCT {
     ?publisher vitro:mostSpecificType ?mostSpecificTypeWithId .
 } WHERE {
-    <{{uri}}> vivo:publisher ?publisher .
+    {{subject}} vivo:publisher ?publisher .
     ?publisher vitro:mostSpecificType ?mostSpecificType .
     BIND( CONCAT( STR(?mostSpecificType), REPLACE(STR(?publisher), "(^.*/)", "::") ) AS ?mostSpecificTypeWithId ) .
 }

src/main/resources/templates/sparql/collection/specifiedOutputOf.sparql

@@ -4,7 +4,7 @@ PREFIX rdfs: <http://www.w3.org/2000/01/rdf-schema#>
 CONSTRUCT {
     ?specifiedOutputOf rdfs:label ?labelWithId .
 } WHERE {
-    <{{uri}}> obo:OBI_0000312 ?specifiedOutputOf .
+    {{subject}} obo:OBI_0000312 ?specifiedOutputOf .
     ?specifiedOutputOf rdfs:label ?label .
     BIND( CONCAT( STR(?label), REPLACE(STR(?specifiedOutputOf), "(^.*/)", "::") ) AS ?labelWithId ) .
 }

src/main/resources/templates/sparql/collection/supportedBy.sparql

@@ -4,7 +4,7 @@ PREFIX rdfs: <http://www.w3.org/2000/01/rdf-schema#>
 CONSTRUCT {
     ?informationResource rdfs:label ?labelWithId .
 } WHERE {
-    <{{uri}}> vivo:informationResourceSupportedBy ?informationResource .
+    {{subject}} vivo:informationResourceSupportedBy ?informationResource .
     ?informationResource rdfs:label ?label .
     BIND( CONCAT( STR(?label), REPLACE(STR(?informationResource), "(^.*/)", "::") ) AS ?labelWithId ) .
 }

src/main/resources/templates/sparql/collection/translator.sparql

@@ -4,7 +4,7 @@ PREFIX rdfs: <http://www.w3.org/2000/01/rdf-schema#>
 CONSTRUCT {
     ?translator rdfs:label ?labelWithId .
 } WHERE {
-    <{{uri}}> bibo:translator ?translator .
+    {{subject}} bibo:translator ?translator .
     ?translator rdfs:label ?label .
     BIND( CONCAT( STR(?label), REPLACE(STR(?translator), "(^.*/)", "::") ) AS ?labelWithId ) .
 }

src/main/resources/templates/sparql/common/geographicFocus.sparql

@@ -4,7 +4,7 @@ PREFIX rdfs: <http://www.w3.org/2000/01/rdf-schema#>
 CONSTRUCT {
     ?geographicFocus rdfs:label ?labelWithId .
 } WHERE {
-    <{{uri}}> vivo:geographicFocus ?geographicFocus .
+    {{subject}} vivo:geographicFocus ?geographicFocus .
     ?geographicFocus rdfs:label ?label .
     BIND( CONCAT( STR(?label), REPLACE(STR(?geographicFocus), "(^.*/)", "::") ) AS ?labelWithId ) .
 }

src/main/resources/templates/sparql/common/image.sparql

@@ -3,7 +3,7 @@ PREFIX vitro-public: <http://vitro.mannlib.cornell.edu/ns/vitro/public#>
 CONSTRUCT {
     ?downloadLocation vitro-public:directDownloadUrl ?directDownloadUrl .
 } WHERE {
-    <{{uri}}> vitro-public:mainImage ?mainImage .
+    {{subject}} vitro-public:mainImage ?mainImage .
     ?mainImage vitro-public:downloadLocation ?downloadLocation .
     ?downloadLocation vitro-public:directDownloadUrl ?directDownloadUrl .
 }

src/main/resources/templates/sparql/common/modTime.sparql

@@ -1,7 +1,7 @@
 PREFIX vitro: <http://vitro.mannlib.cornell.edu/ns/vitro/0.7#>
 
 CONSTRUCT {
-    <{{uri}}> vitro:modTime ?modTime .
+    {{subject}} vitro:modTime ?modTime .
 } WHERE {
-    <{{uri}}> vitro:modTime ?modTime .
+    {{subject}} vitro:modTime ?modTime .
 }

src/main/resources/templates/sparql/common/sameAs.sparql

@@ -4,7 +4,7 @@ PREFIX rdfs: <http://www.w3.org/2000/01/rdf-schema#>
 CONSTRUCT {
     ?sameAs rdfs:label ?labelWithId .
 } WHERE {
-    <{{uri}}> owl:sameAs ?sameAs .
+    {{subject}} owl:sameAs ?sameAs .
     ?sameAs rdfs:label ?label .
     BIND( CONCAT( STR(?label), REPLACE(STR(?sameAs), "(^.*/)", "::") ) AS ?labelWithId ) .
 }

src/main/resources/templates/sparql/common/thumbnail.sparql

@@ -3,7 +3,7 @@ PREFIX vitro-public: <http://vitro.mannlib.cornell.edu/ns/vitro/public#>
 CONSTRUCT {
     ?downloadLocation vitro-public:directDownloadUrl ?directDownloadUrl .
 } WHERE {
-    <{{uri}}> vitro-public:mainImage ?mainImage .
+    {{subject}} vitro-public:mainImage ?mainImage .
     ?mainImage vitro-public:thumbnailImage ?thumbnailImage .
     ?thumbnailImage vitro-public:downloadLocation ?downloadLocation .
     ?downloadLocation vitro-public:directDownloadUrl ?directDownloadUrl .   

src/main/resources/templates/sparql/common/type.sparql

@@ -1,7 +1,7 @@
 PREFIX vitro: <http://vitro.mannlib.cornell.edu/ns/vitro/0.7#>
 
 CONSTRUCT {
-    <{{uri}}> vitro:mostSpecificType ?mostSpecificType .
+    {{subject}} vitro:mostSpecificType ?mostSpecificType .
 } WHERE {
-    <{{uri}}> vitro:mostSpecificType ?mostSpecificType .
+    {{subject}} vitro:mostSpecificType ?mostSpecificType .
 }

src/main/resources/templates/sparql/common/website.sparql

@@ -5,7 +5,7 @@ PREFIX rdfs: <http://www.w3.org/2000/01/rdf-schema#>
 CONSTRUCT {
     ?hasURL rdfs:label ?labelWithId .
 } WHERE {
-    <{{uri}}> obo:ARG_2000028 ?vcard .
+    {{subject}} obo:ARG_2000028 ?vcard .
     ?vcard vcard:hasURL ?hasURL .
     ?hasURL rdfs:label ?label .
     BIND( CONCAT( STR(?label), REPLACE(STR(?hasURL), "(^.*/)", "::") ) AS ?labelWithId ) .

src/main/resources/templates/sparql/common/websiteUrl.sparql

@@ -4,7 +4,7 @@ PREFIX vcard: <http://www.w3.org/2006/vcard/ns#>
 CONSTRUCT {
     ?hasURL vcard:url ?urlWithLabel .
 } WHERE {
-    <{{uri}}> obo:ARG_2000028 ?vcard .
+    {{subject}} obo:ARG_2000028 ?vcard .
     ?vcard vcard:hasURL ?hasURL .
     ?hasURL vcard:url ?url .
     BIND( CONCAT( STR(?url), REPLACE(STR(?hasURL), "(^.*/)", "::") ) AS ?urlWithLabel ) .

src/main/resources/templates/sparql/concept/associatedDepartment.sparql

@@ -5,7 +5,7 @@ PREFIX rdfs: <http://www.w3.org/2000/01/rdf-schema#>
 CONSTRUCT {
     ?organization rdfs:label ?labelWithId .
 } WHERE {
-    <{{uri}}> vivo:researchAreaOf ?researchAreaOf .
+    {{subject}} vivo:researchAreaOf ?researchAreaOf .
     ?researchAreaOf a foaf:Person .
     ?researchAreaOf vivo:relates ?position .
     ?position a vivo:Position .