Updated pyMetadata with new source

pyMetadata/app.py

@@ -10,6 +10,7 @@ from fuzzywuzzy import fuzz
 from sources.result import DataResult, Metadata
 from sources.anii import metadata as AniiMetadata
 from sources.imdb import metadata as ImdbMetadata
+from sources.mal import metadata as MalMetadata
 
 # Konfigurer Kafka-forbindelsen
 bootstrap_servers = os.environ.get("KAFKA_BOOTSTRAP_SERVER") or "127.0.0.1:9092"
@@ -144,21 +145,36 @@ class MessageHandlerThread(threading.Thread):
     def perform_action(self, title) -> DataResult:
         anii = AniiMetadata(title)
         imdb = ImdbMetadata(title)
+        mal = MalMetadata(title)
 
+        mal_result = mal.lookup()
         anii_result = anii.lookup()
         imdb_result = imdb.lookup()
 
         # Sammenlign resultater basert på likheter og sammenhenger med tittelen
-        if anii_result.statusType == "SUCCESS" and imdb_result.statusType == "SUCCESS":
+        if anii_result.statusType == "SUCCESS" and imdb_result.statusType == "SUCCESS" and mal_result.statusType == "SUCCESS":
             # Begge registrene ga suksessresultater, bruk fuzzy matching for å gjøre en vurdering
             title_similarity_anii = fuzz.ratio(title.lower(), anii_result.data.title.lower())
             title_similarity_imdb = fuzz.ratio(title.lower(), imdb_result.data.title.lower())
+            title_similarity_mal = fuzz.ratio(title.lower(), mal_result.data.title.lower())
 
-            # Sammenlign likheter mellom tittel og registertitler
+            alt_titles_anii = anii_result.data.altTitle
-            if title_similarity_anii > title_similarity_imdb:
+            alt_titles_imdb = imdb_result.data.altTitle
-                most_likely_result = anii_result
+            alt_titles_mal = mal_result.data.altTitle
-            else:
+
+            # Sammenlign likheter mellom tittel og registertitler, inkludert alternative titler
+            if (
+                title_similarity_anii * 0.8 + sum(fuzz.ratio(title.lower(), alt_title.lower()) for alt_title in alt_titles_anii) * 0.2
+                < title_similarity_mal * 0.8 + sum(fuzz.ratio(title.lower(), alt_title.lower()) for alt_title in alt_titles_mal) * 0.2
+            ):
+                most_likely_result = mal_result
+            elif (
+                title_similarity_imdb * 0.8 + sum(fuzz.ratio(title.lower(), alt_title.lower()) for alt_title in alt_titles_imdb) * 0.2
+                > title_similarity_anii * 0.8 + sum(fuzz.ratio(title.lower(), alt_title.lower()) for alt_title in alt_titles_anii) * 0.2
+            ):
                 most_likely_result = imdb_result
+            else:
+                most_likely_result = anii_result
 
         elif anii_result.statusType == "SUCCESS":
             # AniList ga suksessresultat, bruk det som det mest sannsynlige
@@ -168,13 +184,19 @@ class MessageHandlerThread(threading.Thread):
             # IMDb ga suksessresultat, bruk det som det mest sannsynlige
             most_likely_result = imdb_result
 
+        elif mal_result.statusType == "SUCCESS":
+            # MAL ga suksessresultat, bruk det som det mest sannsynlige
+            most_likely_result = mal_result
+
         else:
-            # Begge registrene feilet, håndter etter eget behov
+            # Ingen resultater, bruk AniList hvis tilgjengelig
-            most_likely_result = DataResult(statusType="ERROR", errorMessage="No Result")
+            most_likely_result = anii_result
 
         # Returner det mest sannsynlige resultatet
         return most_likely_result
 
+
+
     def compose_message(self, referenceId: str, result: DataResult) -> ProducerDataValueSchema:
         return ProducerDataValueSchema(
             referenceId=referenceId,

pyMetadata/requirements.txt

@@ -3,4 +3,5 @@ AnilistPython>=0.1.3
 kafka-python>=2.0.2
 fuzzywuzzy>=0.18.0
 requests>=2.31.0
-python-Levenshtein>=0.21.1
+python-Levenshtein>=0.21.1
+mal-api>=0.5.3

pyMetadata/sources/anii.py

@@ -15,7 +15,7 @@ class metadata():
 
             meta = Metadata(
                 title = result.get("name_english", None),
-                altTitle = result.get("name_romaji", None),
+                altTitle = [result.get("name_romaji", None)],
                 cover = result.get("cover_image", None),
                 summary = result.get("desc", None),
                 type = 'movie' if result.get('airing_format', '').lower() == 'movie' else 'serie',

pyMetadata/sources/imdb.py

@@ -17,7 +17,7 @@ class metadata():
             result = self.imdb.get_movie(imdbId)
             meta = Metadata(
                 title = result.get("title", None),
-                altTitle = result.get("localized title", None),
+                altTitle = [result.get("localized title", None)],
                 cover = result.get("cover url", None),
                 summary = result.get("plot outline", None),
                 type = 'movie' if result.get('kind', '').lower() == 'movie' else 'serie',

pyMetadata/sources/result.py

@@ -4,7 +4,7 @@ from dataclasses import dataclass, asdict
 @dataclass
 class Metadata:
     title: str
-    altTitle: str
+    altTitle: List[str]
     cover: str
     type: str  # Serie/Movie
     summary: str