Merge pull request #126 from bounswe/backend-wikidata-api

feat: add wikidata query endpoint

Conflicts solved in the other branch, ready to merge.

.gitignore

@@ -1,6 +1,9 @@
 
 .DS_Store
 .env
+/keypair
+.venv/**
+venv
 __pycache__/**
 *.pyc
 **/mydb/**

backend/project/app/qlever.py

@@ -0,0 +1,73 @@
+# This helper class is used to interact with the Wikidata API
+import requests
+import json
+from typing import List
+
+class QleverAPI:
+    def __init__(self):
+        self.endpoint_url = "https://qlever.cs.uni-freiburg.de/api/wikidata"
+        self.params = {
+            "action": "wbsearchentities",
+            "format": "json",
+            "language": "en",
+        }
+
+    # Send a semantic query to the Wikidata API
+    def execute_query(self, query):
+        try:
+            response = requests.get(
+                self.endpoint_url,
+                params={'query': query, 'format': 'json'}
+            )
+            response.raise_for_status()
+            print(response)
+            return response.json()
+        except requests.exceptions.RequestException as e:
+            print("Error:", e)
+            return None
+
+    def film_pattern_query(self, pattern, limit):
+
+        pattern = pattern.lower()
+        # remove spaces from the pattern
+        pattern = pattern.replace(" ", "")
+
+        SPARQL = f"""
+            PREFIX rdfs: <http://www.w3.org/2000/01/rdf-schema#>
+            PREFIX wd: <http://www.wikidata.org/entity/>
+            PREFIX wdt: <http://www.wikidata.org/prop/direct/>
+            SELECT DISTINCT ?film ?filmLabel ?filmId WHERE {{
+                {{
+                    SELECT ?film ?filmLabel ?filmId (1 as ?order) WHERE {{
+                        ?film wdt:P31 wd:Q11424;
+                            rdfs:label ?filmLabel.
+                        FILTER(LANG(?filmLabel) = "en")
+                        FILTER(STRSTARTS(REPLACE(LCASE(?filmLabel), " ", ""), "{pattern}"))
+                    }}
+                }}
+                UNION
+                {{
+                    SELECT ?film ?filmLabel ?filmId (2 as ?order) WHERE {{
+                        ?film wdt:P31 wd:Q11424;
+                            rdfs:label ?filmLabel.
+                        FILTER(LANG(?filmLabel) = "en")
+                        BIND(REPLACE(LCASE(?filmLabel), " ", "") AS ?formattedLabel)
+                        FILTER(REGEX(?formattedLabel, "{pattern}", "i"))
+                        FILTER (!STRSTARTS(REPLACE(LCASE(?filmLabel), " ", ""), "{pattern}"))
+                    }}
+                }}
+            }}
+            ORDER BY ?order
+            LIMIT {limit}
+        """
+
+        print(SPARQL)
+
+        results = self.execute_query(SPARQL)
+
+        return results
+
+
+
+
+

backend/project/app/serializers.py

@@ -86,3 +86,10 @@ class Meta:
         model = Actor
         fields = ['name', 'surname', 'description']
 
+
+class WikidataQuerySerializer(serializers.Serializer):
+    query = serializers.CharField()
+
+class FilmPatternWithLimitQuerySerializer(serializers.Serializer):
+    pattern = serializers.CharField()
+    limit = serializers.IntegerField()

backend/project/app/urls.py

@@ -3,17 +3,21 @@
 from app import views
 from django.urls import path
 from drf_spectacular.views import SpectacularAPIView, SpectacularSwaggerView
-from .views import film_api, film_detail_api, RegisterView, MyObtainTokenPairView, LogoutView, VerifyEmail
+from .views import film_api, film_detail_api, RegisterView, execute_query, query_film_pattern, MyObtainTokenPairView, LogoutView, VerifyEmail
 from rest_framework_simplejwt.views import TokenRefreshView
 
+
 urlpatterns = [
     path('film/schema/', SpectacularAPIView.as_view(), name='schema'),
     path('film/schema/swagger-ui/', SpectacularSwaggerView.as_view(url_name='schema'), name='swagger-ui'),
     path('film/', film_api, name='film-list'),
     path('film/<int:id>/', film_detail_api, name='film-detail'),
     path('login/', MyObtainTokenPairView.as_view(), name='token_obtain_pair'),
     path('login/refresh/', TokenRefreshView.as_view(), name='token_refresh'),
-    path('register/', RegisterView.as_view(), name='auth_register'),
+     path('register/', RegisterView.as_view(), name='auth_register'),
+    path('wikidata-query/', execute_query, name='wikidata-query'),
+    path('query-film-pattern/', query_film_pattern, name='query-film-pattern'),
     path('logout/', LogoutView.as_view(), name='logout'),
-    path('email-verify/', VerifyEmail.as_view(), name='email-verify'),
+    path('email-verify/', VerifyEmail.as_view(), name='email-verify')
 ]
+

backend/project/app/views.py

@@ -6,6 +6,9 @@
 from django.conf import settings
 from django.http.response import JsonResponse
 from django.contrib.auth.models import User
+from app.models import Film, Genre, Director, Actor
+from app.serializers import *
+#from app.serializers import UserSerializer, FilmSerializer, GenreSerializer, DirectorSerializer, ActorSerializer,WikidataQuerySerializer, FilmPatternWithLimitQuerySerializer, MyTokenObtainPairSerializer, LogoutSerializer
 from rest_framework import permissions, status , viewsets, generics
 from rest_framework.response import Response
 from rest_framework_simplejwt.tokens import RefreshToken
@@ -14,9 +17,13 @@
 from rest_framework.permissions import AllowAny, IsAuthenticated
 import jwt
 from rest_framework_simplejwt.views import TokenObtainPairView
-
-from app.models import Film, Genre, Director, Actor
-from app.serializers import UserSerializer, FilmSerializer, GenreSerializer, DirectorSerializer, ActorSerializer, RegisterSerializer, LogoutSerializer, MyTokenObtainPairSerializer
+from rest_framework.permissions import IsAuthenticated
+from .serializers import RegisterSerializer
+from rest_framework import generics
+from app.wikidata import WikidataAPI
+from app.qlever import QleverAPI
+from rest_framework.response import Response
+from rest_framework import status
 from .utils import Util
 
 
@@ -169,3 +176,70 @@ def film_detail_api(request, id):
         return JsonResponse("Film Deleted Successfully", safe=False)
 
 
+# A simple endpoint for sending a semantic query to the Wikidata API
+@extend_schema(
+    description="API endpoint for sending a semantic query to the Wikidata API.",
+    methods=['POST'],
+    request=WikidataQuerySerializer,
+)
+@api_view(['POST'])
+def execute_query(request):
+    """
+    Allow users to send a semantic query to the Wikidata API.
+    """
+    if request.method == 'POST':
+        serializer = WikidataQuerySerializer(data=request.data)
+        if serializer.is_valid():
+            query_text = serializer.validated_data.get('query')
+
+            # Execute the query using the WikidataAPI class
+            wikidata_api = WikidataAPI()
+            results = wikidata_api.execute_query(query_text)
+
+            print(results)
+
+            return Response(results)
+        else:
+            return Response(serializer.errors, status=status.HTTP_400_BAD_REQUEST)
+
+
+# Find films with a pattern string and a limit value
+@extend_schema(
+    description="API endpoint for finding films with a pattern string and a limit value.",
+    methods=['POST'],
+    request=FilmPatternWithLimitQuerySerializer,
+)
+@api_view(['POST'])
+def query_film_pattern(request):
+    """
+    Find films with a pattern string and a limit value using Qlever.
+    """
+    if request.method == 'POST':
+        serializer = FilmPatternWithLimitQuerySerializer(data=request.data)
+        if serializer.is_valid():
+            pattern = serializer.validated_data.get('pattern')
+            limit = serializer.validated_data.get('limit')
+
+            # Execute the query using the Qlever class
+            qlever = QleverAPI()
+            results = qlever.film_pattern_query(pattern, limit)
+
+            print(results)
+
+            # change response format
+            # get only film ids and labels
+            results = results['results']['bindings']
+            films = []
+            for result in results:
+                film = {
+                    'id': result['film']['value'],
+                    'label': result['filmLabel']['value']
+                }
+                films.append(film)
+            return Response(films)
+
+        else:
+            return Response(serializer.errors, status=status.HTTP_400_BAD_REQUEST)
+
+
+

backend/project/app/wikidata.py

@@ -0,0 +1,37 @@
+# This helper class is used to interact with the Wikidata API
+import requests
+import json
+from typing import List
+
+QUERY = """
+SELECT ?item ?itemLabel ?itemDescription ?itemAltLabel WHERE {
+    ?item wdt:P31 wd:Q11424.
+    SERVICE wikibase:label { bd:serviceParam wikibase:language "[AUTO_LANGUAGE],en". }
+    }
+"""
+
+class WikidataAPI:
+    def __init__(self):
+        self.endpoint_url = "https://query.wikidata.org/sparql"
+        self.params = {
+            "action": "wbsearchentities",
+            "format": "json",
+            "language": "en",
+        }
+
+    # Send a semantic query to the Wikidata API
+    def execute_query(self, query):
+        try:
+            response = requests.get(
+                self.endpoint_url,
+                params={'query': query, 'format': 'json'}
+            )
+            response.raise_for_status()
+            print(response)
+            return response.json()
+        except requests.exceptions.RequestException as e:
+            print("Error:", e)
+            return None
+
+
+

backend/project/project/settings.py

@@ -44,6 +44,12 @@
         'PASSWORD': 'password',  # MySQL password
         'HOST': 'db',  # Host where MySQL is running (in this case, Docker container)
         'PORT': '3306',  # Port where MySQL is running (in this case, Docker container)
+
+    }
+}
+"""
+
+"""
 DATABASES = {
     'default': {
         'ENGINE': 'django.db.backends.mysql',

backend/project/docker-compose.yml → docker-compose.yml

@@ -17,11 +17,12 @@ services:
       - backend-network
 
   backend:
+    image: semanticflix_backend
     build: 
-      context: .
+      context: ./backend/project
       dockerfile: Dockerfile
     container_name: semanticflix_backend
-    command: sh -c "python3 manage.py migrate --noinput && python3 manage.py collectstatic --noinput && python manage.py runserver 0.0.0.0:8000"
+    command: sh -c "python backend/project/manage.py migrate --noinput && python backend/project/manage.py collectstatic --noinput && python backend/project/manage.py runserver 0.0.0.0:8000"
     restart: always
     volumes:
       - .:/app

sample_queries.txt

@@ -0,0 +1,54 @@
+======= Films played by dicaprio
+{
+  "query": "SELECT ?film ?filmLabel WHERE { ?film wdt:P31 wd:Q11424; wdt:P161 wd:Q38111. SERVICE wikibase:label { bd:serviceParam wikibase:language \"[AUTO_LANGUAGE],en\". } }"
+}
+=======
+
+
+===== Film ids starting with the given string 
+{
+  "query": "SELECT DISTINCT ?film ?filmLabel ?filmId WHERE { ?film wdt:P31 wd:Q11424; rdfs:label ?filmLabel. FILTER(STRSTARTS(?filmLabel, \"OPP\")) OPTIONAL { ?film wdt:P345 ?filmId. } SERVICE wikibase:label { bd:serviceParam wikibase:language \"[AUTO_LANGUAGE],en\". } } LIMIT 10"
+}
+
+=== Better version of it 
+SELECT DISTINCT ?film ?filmLabel ?filmId WHERE {
+  ?film wdt:P31 wd:Q11424;    # Instance of film
+        rdfs:label ?filmLabel. # Label of the film
+  FILTER(LANG(?filmLabel) = "en")   # Filter out non-English labels
+  FILTER(STRSTARTS(REPLACE(LCASE(?filmLabel), " ", ""), "opportunityk"))  # Matches films with labels starting with "A" (case-insensitive and ignoring spaces)
+}
+LIMIT 3
+====
+
+
+====
+Final film listing sparql
+
+PREFIX rdfs: <http://www.w3.org/2000/01/rdf-schema#>
+PREFIX wd: <http://www.wikidata.org/entity/>
+PREFIX wdt: <http://www.wikidata.org/prop/direct/>
+SELECT DISTINCT ?film ?filmLabel ?filmId WHERE {
+  {
+    SELECT ?film ?filmLabel ?filmId (1 as ?order) WHERE {
+      ?film wdt:P31 wd:Q11424;    # Instance of film
+            rdfs:label ?filmLabel. # Label of the film
+      FILTER(LANG(?filmLabel) = "en")   # Filter out non-English labels
+      FILTER(STRSTARTS(REPLACE(LCASE(?filmLabel), " ", ""), "oppen"))  # Matches labels starting with "oppen" (case-insensitive and ignoring spaces)
+    }
+  }
+  UNION
+  {
+    SELECT ?film ?filmLabel ?filmId (2 as ?order) WHERE {
+      ?film wdt:P31 wd:Q11424;    # Instance of film
+            rdfs:label ?filmLabel. # Label of the film
+      FILTER(LANG(?filmLabel) = "en")   # Filter out non-English labels
+      FILTER(REGEX(?filmLabel, "oppen", "i"))  # Matches labels containing "oppen" (case-insensitive)
+      FILTER (!STRSTARTS(REPLACE(LCASE(?filmLabel), " ", ""), "oppen")) # Ensure it's not already matched by starts with
+    }
+  }
+}
+ORDER BY ?order
+LIMIT 3
+
+
+====