rara-tools 0.6.15__py3-none-any.whl → 0.6.17__py3-none-any.whl

rara_tools/elastic.py

@@ -4,7 +4,7 @@ import elasticsearch_dsl
 from elastic_transport import ObjectApiResponse
 from elasticsearch import Elasticsearch
 from elasticsearch.helpers import bulk
-from elasticsearch_dsl import Index
+from elasticsearch_dsl import Index, Search, Q
 from elasticsearch_dsl.response import Response
 
 from .decorators import _elastic_connection
@@ -84,20 +84,30 @@ class KataElastic:
     def add_mapping(self, index_name: str, schema: dict):
         index = Index(name=index_name)
         return index.put_mapping(body=schema, using=self.elasticsearch)
-    
-    
+
+    @_elastic_connection
+    def delete_by_query(self, index: str, query_kwargs: dict, query_type: str = "term", wait_for_completion=True):
+        query = Q(query_type, **query_kwargs)
+        s = Search(using=self.elasticsearch, index=index).query(query)
+        response = self.elasticsearch.delete_by_query(
+            index=index,
+            body={"query": s.to_dict()["query"]},
+            wait_for_completion=True
+        )
+        return response
+
     @_elastic_connection
     def add_vector_mapping(
-            self, 
-            index_name: str, 
-            field: str, 
-            schema: Optional[dict] = None, 
+            self,
+            index_name: str,
+            field: str,
+            schema: Optional[dict] = None,
             dims: int = 1024
     ) -> dict:
         vector_mapping = {
             "properties": {
                 field: {
-                    "type": "dense_vector", 
+                    "type": "dense_vector",
                     "dims": dims
                 }
             }
@@ -105,22 +115,21 @@ class KataElastic:
         mapping = schema or vector_mapping
         index = Index(name=index_name)
         return index.put_mapping(body=mapping, using=self.elasticsearch)
-    
-    
+
     @_elastic_connection
     def add_ann_vector_mapping(
-            self, 
+            self,
             index_name: str,
             field: str,
-            schema: Optional[dict] = None, 
+            schema: Optional[dict] = None,
             dims: int = 1024
     ) -> dict:
         vector_mapping = {
             "properties": {
                 field: {
-                    "type": "dense_vector", 
-                    "dims": dims, 
-                    "similarity": "cosine", 
+                    "type": "dense_vector",
+                    "dims": dims,
+                    "similarity": "cosine",
                     "index": True
                 }
             }
@@ -131,15 +140,19 @@ class KataElastic:
 
     @_elastic_connection
     def add_vector(
-            self, 
+            self,
             index_name: str,
-            document_id: str, 
-            vector: List[float], 
-            field: str
+            document_id: str,
+            vector: List[float],
+            field: str,
+            refresh: str = "wait_for"
     ) -> dict:
         schema = {"doc": {field: vector}}
         return self.elasticsearch.update(
-            index=index_name, id=document_id, body=schema, refresh="wait_for"
+            index=index_name,
+            id=document_id,
+            body=schema,
+            refresh=refresh
         )
 
     @_elastic_connection
@@ -204,7 +217,7 @@ class KataElastic:
         actions = [{"_index": last_index_name, "_source": document} for document in documents]
         successful_count, error_count = bulk(actions=actions, client=self.elasticsearch, max_retries=3, refresh=refresh)
         return successful_count, error_count
-    
+
     @_elastic_connection
     def bulk_index_without_rollver(
             self,
@@ -240,16 +253,15 @@ class KataElastic:
             s.scan(), key=lambda doc: [getattr(doc, field) for field in sort_fields]
         )
         return documents
-    
-    
+
     @_elastic_connection
     def execute_fuzzy_search(
-            self, 
+            self,
             index: str,
             field: str,
-            entity: str, 
-            fuzziness: int = 2, 
-            prefix_length: int = 1, 
+            entity: str,
+            fuzziness: int = 2,
+            prefix_length: int = 1,
             max_expansions: int = 50
     ) -> Response:
         """Executes a fuzzy search.
@@ -261,7 +273,7 @@ class KataElastic:
         :param: max_expansion int: maximum number of terms the fuzzy query 
             will match before halting the search
         :return: Dict on search results.
-        """  
+        """
         query_params = {
             f"{field}.keyword": {
                 "value": entity,
@@ -274,7 +286,7 @@ class KataElastic:
         s = s.query("fuzzy", **query_params)
         response = s.execute()
         return response
-    
+
     def execute_ann_vector_search(
             self,
             index: str,
@@ -303,8 +315,8 @@ class KataElastic:
         s = s.extra(
             knn={
                 "field": field,
-                "query_vector": query_vector, 
-                "k": k, 
+                "query_vector": query_vector,
+                "k": k,
                 "num_candidates": num_candidates
             }
         )
@@ -314,15 +326,14 @@ class KataElastic:
             s = s.query(
                 elasticsearch_dsl.Q("terms", _id=elastic_ids)
             )
-            
+
         # Sort by score and return `n_docs` best-matching documents
         s = s.extra(size=n_docs)
 
         # Execute the search
         response = s.execute()
         return response
-    
-        
+
     def execute_script_score_vector_search(
             self,
             index: str,
@@ -341,7 +352,7 @@ class KataElastic:
         :param: elastic_ids: List[str]: Elastic ID-s for restricting the search.
         """
         s = elasticsearch_dsl.Search(using=self.elasticsearch, index=index)
-        
+
         if elastic_ids:
             query = elasticsearch_dsl.Q("terms", _id=elastic_ids)
         else:
@@ -364,6 +375,5 @@ class KataElastic:
         response = s.execute()
         return response
 
-
     def __str__(self) -> str:
         return self.elasticsearch_url

rara_tools/normalizers/viaf.py

@@ -672,6 +672,7 @@ class VIAFClient:
             f"Allowed VIAF sources are: {self.allowed_viaf_sources}."
         )
         viaf_record = None
+        verified_record = None
         viaf_ids = self.get_viaf_ids_by_search_terms(
             search_term=search_term,
             field=field,

{rara_tools-0.6.15.dist-info → rara_tools-0.6.17.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: rara-tools
-Version: 0.6.15
+Version: 0.6.17
 Summary: Tools to support Kata's work.
 Classifier: Programming Language :: Python :: 3
 Classifier: Programming Language :: Python :: 3.10

{rara_tools-0.6.15.dist-info → rara_tools-0.6.17.dist-info}/RECORD

@@ -1,7 +1,7 @@
 rara_tools/converters.py,sha256=a1dEMa0TwcO9UmjuSBkiuc7LGmH0d_dB6wwoTLpdZhI,4040
 rara_tools/decorators.py,sha256=MjOyvZ5nTkwxwx2JLFEGpKKBysvecFw6EN6UDrSvZLU,2187
 rara_tools/digar_schema_converter.py,sha256=k95U2iRlEA3sh772-v6snhHW6fju6qSTMnvWJ6DpzZk,14254
-rara_tools/elastic.py,sha256=7HvDmFKpQbGnnzYyiCKOg0uvubnv2TpCASRrnPP8DcQ,13540
+rara_tools/elastic.py,sha256=4D9yoyMy6AJIKwhSi2H1usffDHAh2A_IZfv5BtYnBKg,13992
 rara_tools/exceptions.py,sha256=YQyaueUbXeTkJYFDEuN6iWTXMI3eCv5l7PxGp87vg5I,550
 rara_tools/formatters.py,sha256=LTliadjIPZTO4s-44NsumaUdlQlEvqetvWz4bEvwf90,3418
 rara_tools/s3.py,sha256=9ziDXsLjBtFAvsjTPxFddhfvkpA8773rzPJqO7y1N5Q,6415
@@ -20,7 +20,7 @@ rara_tools/normalizers/__init__.py,sha256=_NqpS5w710DhaURytHq9JpEt8HgYpSPfRDcOtO
 rara_tools/normalizers/authorities.py,sha256=IDtcm0yNZNhv1f-WcdqWFSRzZk_CoKuBFsk6hEPddWM,4513
 rara_tools/normalizers/base.py,sha256=6tLfNdF6FZo8M6j_Q61lXoaF1HdIB1c0SKMatTc-Z64,12014
 rara_tools/normalizers/bibs.py,sha256=4DTS6k37z8qR5B3n7aiCXsT5Z49rLTvQ60lKKr5dyLs,2352
-rara_tools/normalizers/viaf.py,sha256=IA_dIJBA47IBnt0O1E2vKD6UbqhJdrSyg080CXEikKA,24187
+rara_tools/normalizers/viaf.py,sha256=LIeqbJoKtVt_0H1o7XMmhSE0BjF4l-jdAJgX_8Gg9Z4,24218
 rara_tools/parsers/marc_parsers/base_parser.py,sha256=Kdw4aivJf2FkWgIK7pJtHtVXF_G1pjHVQ7IcFItSqy8,1649
 rara_tools/parsers/marc_parsers/ems_parser.py,sha256=LFuhZcVwmHMcJknX9p4ZkO8RdjPdQZ4APGbw8KV6BIs,2024
 rara_tools/parsers/marc_parsers/location_parser.py,sha256=dSU9dQoGV5z0ajhLI1bn3AAghkOr79qKIrX7sO0_4lA,1873
@@ -35,8 +35,8 @@ rara_tools/parsers/marc_records/title_record.py,sha256=XrtJ4gj7wzSaGxNaPtPuawmqq
 rara_tools/parsers/tools/entity_normalizers.py,sha256=VyCy_NowCLpOsL0luQ55IW-Qi-J5oBH0Ofzr7HRFBhM,8949
 rara_tools/parsers/tools/marc_converter.py,sha256=LgSHe-7n7aiDrw2bnsB53r3fXTRFjZXTwBYfTpL0pfs,415
 rara_tools/parsers/tools/russian_transliterator.py,sha256=5ZU66iTqAhr7pmfVqXPAI_cidF43VqqmuN4d7H4_JuA,9770
-rara_tools-0.6.15.dist-info/licenses/LICENSE.md,sha256=hkZVnIZll7e_KNEQzeY94Y9tlzVL8iVZBTMBvDykksU,35142
-rara_tools-0.6.15.dist-info/METADATA,sha256=WOK0svJOvm14qHwAjnGHi9daTaaDOIZi2__URFTw44c,4080
-rara_tools-0.6.15.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
-rara_tools-0.6.15.dist-info/top_level.txt,sha256=JwfB5b8BAtW5OFKRln2AQ_WElTRyIBM4nO0FKN1cupY,11
-rara_tools-0.6.15.dist-info/RECORD,,
+rara_tools-0.6.17.dist-info/licenses/LICENSE.md,sha256=hkZVnIZll7e_KNEQzeY94Y9tlzVL8iVZBTMBvDykksU,35142
+rara_tools-0.6.17.dist-info/METADATA,sha256=_E9Ml7OSHn29YTsh5V4DytlzwRTSf-xeeQRZJNDzbos,4080
+rara_tools-0.6.17.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
+rara_tools-0.6.17.dist-info/top_level.txt,sha256=JwfB5b8BAtW5OFKRln2AQ_WElTRyIBM4nO0FKN1cupY,11
+rara_tools-0.6.17.dist-info/RECORD,,