Add dbpedia triples

app.py

 from flask import Flask, Response
 from flask.json import jsonify
 
-from src.main import assemble_wikidata_groundtruth, assemble_dbpedia_groundtruth, assemble_wikidata_triples
+from src.main import assemble_wikidata_groundtruth, assemble_dbpedia_groundtruth, assemble_wikidata_triples, \
+    assemble_dbpedia_triples
 
 app = Flask(__name__)
 
 
+@app.route('/dbpedia/n3')
+def dbpedia_n3():
+    return jsonify(assemble_dbpedia_triples())
+
+
 @app.route('/wikidata/n3')
 def wikidata_n3():
     return jsonify(assemble_wikidata_triples())

src/dbpedia/keys.py

+from enum import Enum
+
+
+class ResultKeys(Enum):
+    movie = 'Movie'
+    title = 'Title'
+    director = 'Director'
+    author = 'Author'
+    cast = 'Cast'
+    published = 'Published'
+    subject = 'Subject'
+    genre = 'Genre'
+    duration = 'Duration'
+    description = 'Description'
+    distributor = 'Distributor'
+    production_companies = 'ProductionCompanies'
+    line_separator = '|'

src/main.py

@@ -8,6 +8,7 @@ from src.lib.reader import FileReader
 from src.lib.ntriple import NTriple
 from src.lib.writer import FileWriter
 from src.wikidata.keys import ResultKeys
+from src.dbpedia.keys import ResultKeys
 
 
 def assemble_wikidata_groundtruth() -> Dict:
@@ -42,7 +43,7 @@ def assemble_wikidata_triples() -> List:
     """
     This method assembles the N-Triples of wikidata.
 
-    :return:
+    :return: List of all triples.
     """
 
     data = FileReader(source='static/wikidata_groundtruth.txt').as_json()
@@ -73,7 +74,8 @@ def assemble_wikidata_triples() -> List:
                     triples += [NTriple(subject=movie, predicate=wdt + 'P161', value=cast).as_string()]
 
         if ResultKeys.published.value in result.keys():
-            for published in Decapper(result[ResultKeys.published.value]).unpack().split(ResultKeys.line_separator.value):
+            for published in Decapper(result[ResultKeys.published.value]).unpack().split(
+                    ResultKeys.line_separator.value):
                 if published:
                     triples += [NTriple(subject=movie, predicate=wdt + 'P577', value=published).as_string()]
 
@@ -88,12 +90,94 @@ def assemble_wikidata_triples() -> List:
                     triples += [NTriple(subject=movie, predicate=wdt + 'P2047', value=duration).as_string()]
 
         if ResultKeys.description.value in result.keys():
-            for description in Decapper(result[ResultKeys.description.value]).unpack().split(ResultKeys.line_separator.value):
+            for description in Decapper(result[ResultKeys.description.value]).unpack().split(
+                    ResultKeys.line_separator.value):
                 if description:
                     triples += [NTriple(subject=movie, predicate=schema + 'description', value=description).as_string()]
 
         if ResultKeys.production_company.value in result.keys():
-            for production_company in Decapper(result[ResultKeys.production_company.value]).unpack().split(ResultKeys.line_separator.value):
+            for production_company in Decapper(result[ResultKeys.production_company.value]).unpack().split(
+                    ResultKeys.line_separator.value):
                 if production_company:
                     triples += [NTriple(subject=movie, predicate=wdt + 'P272', value=production_company).as_string()]
     return triples
+
+
+def assemble_dbpedia_triples() -> List:
+    """
+    This method assembles the N-Triples of dbpedia.
+
+    :return: List of all triples.
+    """
+
+    data = FileReader(source='static/dbpedia_groundtruth.txt').as_json()
+    foaf = 'http://xmlns.com/foaf/0.1/'
+    dbo = 'http://dbpedia.org/ontology/'
+    dct = 'http://purl.org/dc/terms/'
+    dbp = 'http://dbpedia.org/property/'
+
+    triples: List[String] = []
+    for result in data:
+        movie = Decapper(result[ResultKeys.movie.value]).unpack()
+
+        if ResultKeys.title.value in result.keys():
+            for title in Decapper(result[ResultKeys.title.value]).unpack().split(ResultKeys.line_separator.value):
+                if title:
+                    triples += [NTriple(subject=movie, predicate=foaf + 'name', value=title).as_string()]
+
+        if ResultKeys.director.value in result.keys():
+            for director in Decapper(result[ResultKeys.director.value]).unpack().split(ResultKeys.line_separator.value):
+                if director:
+                    triples += [NTriple(subject=movie, predicate=dbo + 'director', value=director).as_string()]
+
+        if ResultKeys.author.value in result.keys():
+            for author in Decapper(result[ResultKeys.author.value]).unpack().split(ResultKeys.line_separator.value):
+                if author:
+                    triples += [NTriple(subject=movie, predicate=dbo + 'author', value=author).as_string()]
+
+        if ResultKeys.cast.value in result.keys():
+            for cast in Decapper(result[ResultKeys.cast.value]).unpack().split(ResultKeys.line_separator.value):
+                if cast:
+                    triples += [NTriple(subject=movie, predicate=dbo + 'starring', value=cast).as_string()]
+
+        if ResultKeys.published.value in result.keys():
+            for published in Decapper(result[ResultKeys.published.value]).unpack().split(
+                    ResultKeys.line_separator.value):
+                if published:
+                    triples += [NTriple(subject=movie, predicate=dbo + 'releaseDate', value=published).as_string()]
+
+        if ResultKeys.subject.value in result.keys():
+            for subject in Decapper(result[ResultKeys.subject.value]).unpack().split(ResultKeys.line_separator.value):
+                if subject:
+                    triples += [NTriple(subject=movie, predicate=dct + 'subject', value=subject).as_string()]
+
+        if ResultKeys.genre.value in result.keys():
+            for genre in Decapper(result[ResultKeys.genre.value]).unpack().split(ResultKeys.line_separator.value):
+                if genre:
+                    triples += [NTriple(subject=movie, predicate=dbo + 'genre', value=genre).as_string()]
+
+        if ResultKeys.duration.value in result.keys():
+            for duration in Decapper(result[ResultKeys.duration.value]).unpack().split(ResultKeys.line_separator.value):
+                if duration:
+                    triples += [NTriple(subject=movie, predicate=dbo + 'runtime', value=duration).as_string()]
+
+        if ResultKeys.description.value in result.keys():
+            for description in Decapper(result[ResultKeys.description.value]).unpack().split(
+                    ResultKeys.line_separator.value):
+                if description:
+                    triples += [NTriple(subject=movie, predicate=dbo + 'abstract', value=description).as_string()]
+
+        if ResultKeys.distributor in result.keys():
+            for distributor in Decapper(result[ResultKeys.distributor.value]).unpack().split(
+                    ResultKeys.line_separator.value):
+                if distributor:
+                    triples += [NTriple(subject=movie, predicate=dbo + 'distributor',
+                                        value=distributor).as_string()]
+
+        if ResultKeys.production_companies in result.keys():
+            for production_companies in Decapper(result[ResultKeys.production_companies.value]).unpack().split(
+                    ResultKeys.line_separator.value):
+                if production_companies:
+                    triples += [NTriple(subject=movie, predicate=dbp + 'productionCompanies',
+                                        value=production_companies).as_string()]
+    return triples

static/dbpedia_groundtruth.sparql

@@ -4,8 +4,9 @@ PREFIX dbo: <http://dbpedia.org/ontology/>
 PREFIX dct: <http://purl.org/dc/terms/>
 PREFIX xsd: <http://www.w3.org/2001/XMLSchema#>
 PREFIX foaf: <http://xmlns.com/foaf/0.1/>
+PREFIX dbp:	<http://dbpedia.org/property/>
 
-SELECT ?movie
+SELECT DISTINCT (?movie AS ?Movie)
 
 (GROUP_CONCAT(DISTINCT ?name;separator="|") AS ?Title)
 (GROUP_CONCAT(DISTINCT ?director;separator="|") AS ?Director)