rara-tools 0.0.11__py3-none-any.whl → 0.0.13__py3-none-any.whl

rara_tools/converters.py

@@ -2,40 +2,80 @@ from .exceptions import SierraResponseConverterException
 
 
 class SierraResponseConverter:
-    """  Takes a JSON response from the Sierra API (https://tester.ester.ee/iii/sierra-api/swagger/index.html)
-    and converts it to MARC-in-JSON format.
-    
-    """
+    """Converts a JSON response from the Sierra API to MARC-in-JSON format."""
     
     def __init__(self, response: dict):
         if not isinstance(response, dict):
             raise SierraResponseConverterException("Please provide a valid JSON response.")
         self.response = response
+    
+    def _map_control_fields(self, field: dict) -> dict:
+        # for tags < 010, no subfields, instead one str value in "value" 
+        return {field["tag"]: field["value"]}
+    
+    def _map_data_fields(self, field: dict) -> dict:
+        """ Maps marc fields > 010. 
+
+        Args:
+            field (dict): Contains the marc tag and list with indicators and subfields.
+
+        Returns:
+            dict: standardised marc-in-json format.
+        """
+        
+        data = field["data"]
+        
+        # Order matters ind1, in2, subfields
+        field_data = {
+            "ind1": data.get("ind1", " "),
+            "ind2": data.get("ind2", " "),
+            "subfields": data.get("subfields", [])
+        }
+        
+        return {field["tag"]: field_data}
+    
+    def _is_marc21structured(self, field: dict) -> bool:
+        """Checks if the field is already structured according to MARC21 in JSON"""
+        return any(key.isdigit() for key in field.keys())
+        
+    
+    def _handle_field_type(self, field: dict) -> dict:
+        
+        if self._is_marc21structured(field):
+            return field
+        
+        if field.get("data"):
+            return self._map_data_fields(field)
         
-    def _map_field_data(self, field):
         tag = field.get("tag")
+        
         if not tag:
-            raise SierraResponseConverterException("Field is missing a valid 'tag'.")
-        data = field.get("data", {})
-        return {tag: data}
+            raise SierraResponseConverterException("Field is missing MARC21 tag.")
         
-    def _convert_response(self):
-        response = self.response
+        if tag < "010":
+            return self._map_control_fields(field)
+        else:
+            return self._map_data_fields(field)
         
-        entries = response.get("entries")
+    def _convert_response(self) -> list:
+        entries = self.response.get("entries")
         if not entries:
             raise SierraResponseConverterException("No entries found in the response.")
         
         try:
-            fields = [self._map_field_data(f) for e in entries for f in e["marc"]["fields"]]
+            return {"fields": [
+                {e["id"]: [
+                    self._handle_field_type(f) for f in e["marc"]["fields"]
+                    ]}
+                for e in entries
+            ]}
+            
         except KeyError as e:
-            raise SierraResponseConverterException(f"Missing expected MARC fields in the response: {e}")
+            raise SierraResponseConverterException(f"Malformed response: missing key {e}")
         
-        return {"fields": fields}
-            
-    def convert(self):
-        """Runner method, converts the response to MARC-in-JSON format with error handling."""
+    
+    def convert(self) -> list:
         try:
             return self._convert_response()
         except Exception as e:
-            raise SierraResponseConverterException(f"An unexpected error occurred during conversion: {e}")
+            raise SierraResponseConverterException(f"An unexpected error occurred: {e}")

rara_tools/elastic.py

@@ -263,7 +263,7 @@ class KataElastic:
         response = s.execute()
         return response
     
-    def execute_vector_search(
+    def execute_ann_vector_search(
             self,
             index: str,
             field: str,
@@ -281,6 +281,7 @@ class KataElastic:
         :param: query vector List[float]: Vector to search matches for.
         :param: k int: Number of nearest neighbors to return.
         :param: num_candidates int: Number of candidates considered before selecting k results.
+        :param: n_docs: int: Number of documents to return.
         :param: elastic_ids: List[str]: Elastic ID-s for restricting the search.
         """
 
@@ -308,6 +309,48 @@ class KataElastic:
         # Execute the search
         response = s.execute()
         return response
+    
+        
+    def execute_script_score_vector_search(
+            self,
+            index: str,
+            field: str,
+            query_vector: List[float],
+            n_docs: int = 10,
+            elastic_ids: List[str] = []
+    ) -> Response:
+        """ Execute a vector search.
+        NB! Requires different mapping than ANN!
+        
+        :param: index str: Index to search from.
+        :param: field str: Field containing vectorized data.
+        :param: query vector List[float]: Vector to search matches for.
+        :param: n_docs: int: Number of documents to return.
+        :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:
+            query = elasticsearch_dsl.Q("match_all")
+        # Apply script_score query
+        s = s.query(
+            "script_score",
+            query=query,
+            script={
+                "source": f"1.0 + cosineSimilarity(params.query_vector, '{field}')",
+                "params": {
+                    "query_vector": query_vector
+                }
+            }
+        )
+        # Set min_score and limit number of documents
+        s = s.extra(size=n_docs)
+
+        # Execute search
+        response = s.execute()
+        return response
 
 
     def __str__(self) -> str:

{rara_tools-0.0.11.dist-info → rara_tools-0.0.13.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.2
 Name: rara-tools
-Version: 0.0.11
+Version: 0.0.13
 Summary: Tools to support Kata's work.
 Classifier: Programming Language :: Python :: 3
 Classifier: Programming Language :: Python :: 3.10

{rara_tools-0.0.11.dist-info → rara_tools-0.0.13.dist-info}/RECORD

@@ -1,7 +1,7 @@
-rara_tools/converters.py,sha256=JcS74VzV6jm12l3C6aqMJBY9nuVW_aevQeCe32KmfrE,1576
+rara_tools/converters.py,sha256=O769zNjde1VCfEUF2VU_49IAbm8NT-cG-VR0uPxixtE,2687
 rara_tools/decorators.py,sha256=MjOyvZ5nTkwxwx2JLFEGpKKBysvecFw6EN6UDrSvZLU,2187
 rara_tools/digar_schema_converter.py,sha256=k95U2iRlEA3sh772-v6snhHW6fju6qSTMnvWJ6DpzZk,14254
-rara_tools/elastic.py,sha256=LZfHZqeTDjCEb5YX4CLPJEFffRSZAcRq6AtyP49Fo0E,11575
+rara_tools/elastic.py,sha256=MgPHxZ3UbSTIL8_sT9gU5V4PLKJjo3aQ8CGyhXjRz6M,13065
 rara_tools/exceptions.py,sha256=BwNh4qWxau_ylr9RqZoYwd1KnExI6oWWWDno3jkh8q4,474
 rara_tools/s3.py,sha256=uNDu2HzMYHAWh33RcHeyPFK7gdQfQPxsdfohyIKezEY,4467
 rara_tools/task_reporter.py,sha256=WCcZts9dAUokPc4vbrG3-lNAFLnWaMgE3b3iaUB7mr8,3256
@@ -9,8 +9,8 @@ rara_tools/utils.py,sha256=9vSbmuWYU5ydr4lXBKlUKa0xzDccFsaJv4T-XwgUfuY,2578
 rara_tools/constants/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 rara_tools/constants/digitizer.py,sha256=gJ3jOMwuZfKcLqgOAxTyB266VYsskLabJiMUiSz3xX4,297
 rara_tools/constants/general.py,sha256=E9Jaw-YxocS_tOZw9QBoxO3e9KK5EMbLoM0R7D4Iflw,171
-rara_tools-0.0.11.dist-info/LICENSE.md,sha256=hkZVnIZll7e_KNEQzeY94Y9tlzVL8iVZBTMBvDykksU,35142
-rara_tools-0.0.11.dist-info/METADATA,sha256=pDcladCQ1A9O9Wh4UDSh0eHwNyqcGY1BOwPxSJKpLFk,3895
-rara_tools-0.0.11.dist-info/WHEEL,sha256=In9FTNxeP60KnTkGw7wk6mJPYd_dQSjEZmXdBdMCI-8,91
-rara_tools-0.0.11.dist-info/top_level.txt,sha256=JwfB5b8BAtW5OFKRln2AQ_WElTRyIBM4nO0FKN1cupY,11
-rara_tools-0.0.11.dist-info/RECORD,,
+rara_tools-0.0.13.dist-info/LICENSE.md,sha256=hkZVnIZll7e_KNEQzeY94Y9tlzVL8iVZBTMBvDykksU,35142
+rara_tools-0.0.13.dist-info/METADATA,sha256=0Aipkuodi_CzCTUMkVqKOI__n5mN2r8hEGJ49-MjpMo,3895
+rara_tools-0.0.13.dist-info/WHEEL,sha256=nn6H5-ilmfVryoAQl3ZQ2l8SH5imPWFpm1A5FgEuFV4,91
+rara_tools-0.0.13.dist-info/top_level.txt,sha256=JwfB5b8BAtW5OFKRln2AQ_WElTRyIBM4nO0FKN1cupY,11
+rara_tools-0.0.13.dist-info/RECORD,,

{rara_tools-0.0.11.dist-info → rara_tools-0.0.13.dist-info}/WHEEL

@@ -1,5 +1,5 @@
 Wheel-Version: 1.0
-Generator: setuptools (75.8.0)
+Generator: setuptools (75.8.1)
 Root-Is-Purelib: true
 Tag: py3-none-any