kolzchut-ragbot 1.1.3__tar.gz → 1.3.0__tar.gz

{kolzchut_ragbot-1.1.3 → kolzchut_ragbot-1.3.0}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.2
 Name: kolzchut-ragbot
-Version: 1.1.3
+Version: 1.3.0
 Summary: A search engine using machine learning models and Elasticsearch for advanced document retrieval.
 Home-page: https://github.com/shmuelrob/ragbot
 Author: Shmuel Robinov

{kolzchut_ragbot-1.1.3 → kolzchut_ragbot-1.3.0}/kolzchut_ragbot/engine.py

@@ -178,7 +178,7 @@ class Engine:
         all_docs_by_model = self.elastic_model.search(query_embeddings)
         all_docs = []
         ids_for_fusion = []
-        all_docs_and_scores = []
+        all_docs_and_scores = {}
 
         for key, values in all_docs_by_model.items():
             print(f"\nFound {len(values)} documents for model\n")
@@ -190,7 +190,7 @@ class Engine:
                 all_docs.append(doc)
                 scores_for_model.append({"doc": doc["_source"]["title"], "score": doc["_score"]})
             ids_for_fusion.append(model_ids)
-            all_docs_and_scores.append({f'{key}': scores_for_model})
+            all_docs_and_scores[f'{key}'] = scores_for_model
         print(f"\nFusing {len(ids_for_fusion)} results\n")
         fused_ids = self.reciprocal_rank_fusion(ids_for_fusion, k=top_k)
         top_k_documents = []

{kolzchut_ragbot-1.1.3 → kolzchut_ragbot-1.3.0}/kolzchut_ragbot/model.py

@@ -93,23 +93,18 @@ class Model:
         index = index_from_page_id(int(doc_id))
 
         if update:
-            hits = []
             try:
-                existing_doc = self.es_client.search(
-                    index=index,
-                    body={
-                        "query": {
-                            "term": {f"{identifier}": {"value": doc_id}}}
-                    })
-                hits = existing_doc["hits"]["hits"]
+                query = {
+                    "query": {
+                        "match": {
+                            f"{identifier}": doc_id
+                        }
+                    }
+                }
+                self.es_client.delete_by_query(index=f"{EMBEDDING_INDEX}*", body=query)
 
             except Exception as e:
                 logging.error(f"Error while searching for existing document: {e}")
-            if len(hits) > 0:
-                print("delete existing documents")
-                for hit in hits:
-                    print("delete existing document", hit["_id"])
-                    self.es_client.delete(index=index, id=hit["_id"])
 
         for i, doc_dict in enumerate(paragraphs_dicts):
             print(f"saving paragraph {i}")

{kolzchut_ragbot-1.1.3 → kolzchut_ragbot-1.3.0}/kolzchut_ragbot.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.2
 Name: kolzchut-ragbot
-Version: 1.1.3
+Version: 1.3.0
 Summary: A search engine using machine learning models and Elasticsearch for advanced document retrieval.
 Home-page: https://github.com/shmuelrob/ragbot
 Author: Shmuel Robinov

{kolzchut_ragbot-1.1.3 → kolzchut_ragbot-1.3.0}/setup.py

@@ -1,8 +1,12 @@
 from setuptools import setup, find_packages
 
+# to deploy the package to PyPI, run the following command:
+# python setup.py sdist bdist_wheel
+# twine upload dist/*
+
 setup(
     name='kolzchut-ragbot',
-    version='1.1.3',
+    version='1.3.0',
     author='Shmuel Robinov',
     author_email='shmuel_robinov@webiks.com',
     description='A search engine using machine learning models and Elasticsearch for advanced document retrieval.',