Add Triple Writer; Add Triple-Representation for Wikidata data

.gitignore

@@ -61,7 +61,6 @@ dist/
 downloads/
 eggs/
 .eggs/
-lib/
 lib64/
 parts/
 sdist/

app.py

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

src/main.py

-from typing import Dict
+from tokenize import String
+from typing import Dict, List
 
 from src import WIKIDATA_ENTRYPOINT, DBPEDIA_ENTRYPOINT
+from src.lib.decapper import Decapper
 from src.lib.engine import SPARQLEngine
+from src.lib.reader import FileReader
+from src.lib.ntriple import NTriple
 from src.lib.writer import FileWriter
+from src.wikidata.keys import ResultKeys
 
 
 def assemble_wikidata_groundtruth() -> Dict:
@@ -31,3 +36,64 @@ def assemble_dbpedia_groundtruth() -> Dict:
     writer = FileWriter(destination='static/dbpedia_groundtruth.txt', data=data)
     writer.as_json()
     return data
+
+
+def assemble_wikidata_triples() -> List:
+    """
+    This method assembles the N-Triples of wikidata.
+
+    :return:
+    """
+
+    data = FileReader(source='static/wikidata_groundtruth.txt').as_json()
+    wdt = 'http://www.wikidata.org/prop/direct/'
+    schema = 'http://schema.org/'
+    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=wdt + 'P1476', 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=wdt + 'P57', 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=wdt + 'P58', 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=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):
+                if published:
+                    triples += [NTriple(subject=movie, predicate=wdt + 'P577', value=published).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=wdt + 'P136', 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=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):
+                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):
+                if production_company:
+                    triples += [NTriple(subject=movie, predicate=wdt + 'P272', value=production_company).as_string()]
+    return triples