Update app.py

app.py

@@ -8,19 +8,20 @@ from unstructured.chunking.title import chunk_by_title
 import cohere
 import requests
 import json
+import os
 
 co_embed = cohere.ClientV2(os.environ.get("COHERE_API_KEY"))
 co_rerank = cohere.ClientV2(os.environ.get("COHERE_API_KEY"))
 
 vectored = None
 
-dataid = requests.get("https://www.ryhintl.com/dbjson/getjson?sqlcmd=select * from company_matters")
+#dataid = requests.get("https://www.ryhintl.com/dbjson/getjson?sqlcmd=select * from company_matters")
 
-data_str = dataid.content.decode('utf-8')
-data = json.loads(data_str)
+#data_str = dataid.content.decode('utf-8')
+#data = json.loads(data_str)
 
 # デコード関数
-def decode_text(data):
+'''def decode_text(data):
     for item in data:
         try:
             # latin1 でデコードし、utf-8に変換
@@ -28,7 +29,7 @@ def decode_text(data):
         except UnicodeDecodeError as e:
             print(f"エラー: {e}")
             item['text'] = "[デコード失敗]"
-    return data
+    return data'''
 
 
 
@@ -151,8 +152,8 @@ vectorstore = Vectorstore(raw_documents)
 # Gradioの関数
 def search(query):
     results = vectorstore.retrieve(query)
-    decoded_data = decode_text(results)
-    print("decoded_data:",decoded_data)
+    #decoded_data = decode_text(results)
+    #print("decoded_data:",decoded_data)
     return "\n\n".join([f"**Title**: {r['title']}\n**Text**: {r['text']}\n**URL**: {r['url']}" for r in decoded_data])
 
 # Gradioインターフェース