rara-tools 0.6.0__tar.gz → 0.6.2__tar.gz

{rara_tools-0.6.0/rara_tools.egg-info → rara_tools-0.6.2}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: rara-tools
-Version: 0.6.0
+Version: 0.6.2
 Summary: Tools to support Kata's work.
 Classifier: Programming Language :: Python :: 3
 Classifier: Programming Language :: Python :: 3.10

rara_tools-0.6.2/VERSION

@@ -0,0 +1 @@
+0.6.2

rara_tools-0.6.2/rara_tools/constants/linker.py

@@ -0,0 +1,14 @@
+COMPONENT_KEY = "linker"
+
+
+class Task:
+    BASE = "base_linker_task"
+    VECTORIZE = "vectorize_text"
+    LINK_AND_NORMALIZE = "core_linker_with_normalization"
+    VECTORIZE_AND_INDEX = "core_vectorize_and_index"
+    RECEIVE_LINK_AND_NORMALIZE = "receive_link_and_normalize"
+
+
+class Queue:
+    LINKER = "linker"
+    VECTORIZER = "vectorizer"

rara_tools-0.6.2/rara_tools/constants/subject_indexer.py

@@ -0,0 +1,9 @@
+COMPONENT_KEY = "subject_indexer"
+
+
+class Tasks:
+    PIPELINE = "run_subject_indexer_pipeline"
+
+
+class Queue:
+    MAIN = "subject-indexer"

{rara_tools-0.6.0 → rara_tools-0.6.2}/rara_tools/normalizers/base.py

@@ -2,11 +2,11 @@ from datetime import datetime
 from pymarc import (Field, Subfield, JSONReader, Record)
 from typing import List, Optional, Iterator
 
-from rara_tools.constants import EMPTY_INDICATORS
 from rara_tools.normalizers.viaf import VIAFRecord, VIAFClient
 from rara_tools.constants.normalizers import (
     DEFAULT_VIAF_FIELD, ALLOWED_VIAF_FIELDS, ALLOWED_VIAF_WIKILINK_LANGS,
-    VIAF_SIMILARITY_THRESHOLD, VERIFY_VIAF_RECORD, MAX_VIAF_RECORDS_TO_VERIFY
+    VIAF_SIMILARITY_THRESHOLD, VERIFY_VIAF_RECORD, MAX_VIAF_RECORDS_TO_VERIFY,
+    EMPTY_INDICATORS
 )
 from glom import glom
 import logging

{rara_tools-0.6.0 → rara_tools-0.6.2}/rara_tools/normalizers/viaf.py

@@ -8,7 +8,7 @@ from requests.models import Response
 from rara_tools.parsers.tools.entity_normalizers import PersonalName, Normalizer
 from rara_tools.constants.normalizers import (
     DEFAULT_VIAF_FIELD, ALLOWED_VIAF_FIELDS, ALLOWED_VIAF_WIKILINK_LANGS,
-    VIAF_SIMILARITY_THRESHOLD
+    VIAF_SIMILARITY_THRESHOLD, VIAF_ALLOWED_SOURCES
 )
 
 import logging
@@ -20,9 +20,7 @@ class VIAFRecord:
     """
     def __init__(self,
             record: dict,
-            allowed_sources: List[str] = [
-                "LC", "DNB", "LNB", "NLL", "ERRR", "J9U"
-            ]
+            allowed_sources: List[str] = VIAF_ALLOWED_SOURCES
     ):
         """ Initializes VIAFRecord class.
 
@@ -108,14 +106,36 @@ class VIAFRecord:
             wikilink_lang = match.group()
         return wikilink_lang
 
-    def _get_marc_field(self, marc_dict: dict, subfield: str = "a") -> str:
+    def _get_marc_field(self, marc_dict: dict, subfield: str = "a",
+                        strict_subfield: bool = True
+    ) -> str:
+        """ Retrieve value from a MARC dict
+
+        Parameters
+        -----------
+        marc_dict: dict
+            MARC dictionaryself.
+        subfield: str
+            Subfield to extract
+        strict_subfield: bool
+            If set to True, data is extracted ONLY from
+            the subfield set with param `subfield`. If set to False,
+            data can be extracted from other subfields as well as long
+            there is only one subfield in the dict. This might be necessary
+            for uniformTitleWorks as sometimes the title is present in
+            subfield (t) while subfield (a) contains the author. However,
+            there are instances, where the title is present is subfield (a)
+            with no author.
+        """
         value = ""
         if marc_dict.get("dtype", "") == "MARC21":
             subfields = marc_dict.get("subfield", [])
             for _subfield in subfields:
-                if _subfield.get("code", "") == subfield:
+                if len(subfields) > 1 and _subfield.get("code", "") == subfield:
                     value = _subfield.get("value", "")
                     break
+                elif len(subfields) == 1 and not strict_subfield:
+                    value = _subfield.get("value", "")
         return value
 
     def _get_marc_tag(self, marc_dict: dict) -> str:
@@ -127,7 +147,11 @@ class VIAFRecord:
     def _get_names(self, marc_dicts: List[dict]) -> List[str]:
         names_d = defaultdict(int)
         for marc_dict in marc_dicts:
-            name = self._get_marc_field(marc_dict, self.subfield_indicator)
+            name = self._get_marc_field(
+                marc_dict=marc_dict,
+                subfield=self.subfield_indicator,
+                strict_subfield=False
+            )
             names_d[name]+=1
         name_list = sorted(
             list(names_d.items()),
@@ -154,8 +178,8 @@ class VIAFRecord:
         # Strip "." only if the last token is not an initial,
         # e.g: "Meri, Lennart." -> Strip
         # "Meri, L." -> Do not strip.
-        ent_tokens = entity.split()
-        if len(ent_tokens[-1]) > 2:
+        ent_tokens = [t.strip() for t in entity.split() if t.strip()]
+        if ent_tokens and len(ent_tokens[-1]) > 2:
             entity = entity.strip(".")
         return entity
 
@@ -166,6 +190,14 @@ class VIAFRecord:
         _entity = re.sub(r"[(][^)][)]", "", entity)
         return _entity.strip()
 
+    @property
+    def record(self) -> dict:
+        return self.__record
+
+    @property
+    def record_data(self) -> dict:
+        return self.__record_data
+
     @property
     def subfield_indicator(self) -> str:
         if not self.__subfield_indicator:
@@ -239,7 +271,6 @@ class VIAFRecord:
             vars_3 = [Normalizer.clean_entity(v) for v in _vars]
 
             vars = _vars + vars_3
-            #print(vars)
             self.__name_variations = list(set(vars))
         return self.__name_variations
 
@@ -369,7 +400,10 @@ class VIAFRecord:
 
 
 class VIAFClient:
-    def __init__(self, viaf_api_url: str = "https://viaf.org/api"):
+    def __init__(self,
+            viaf_api_url: str = "https://viaf.org/api",
+            allowed_viaf_sources: List[str] = VIAF_ALLOWED_SOURCES
+        ):
         self.root_url: str = viaf_api_url.strip("/")
         self.record_url: str = f"{self.root_url}/cluster-record"
         self.search_url: str = f"{self.root_url}/search"
@@ -377,6 +411,7 @@ class VIAFClient:
             "Accept": "application/json",
             "Content-Type": "application/json"
         }
+        self.allowed_viaf_sources: List[str] = allowed_viaf_sources
 
     def check_search_term_query(self) -> bool:
         """ Function for checking, if VIAF search term
@@ -448,6 +483,9 @@ class VIAFClient:
                 score: float
                     Similarity score of the most similar record.
         """
+        logger.debug(
+            f"Verifying if '{viaf_record.name}' is sufficiently similar to '{entity}'."
+        )
         # might not always be personal name, but shouldn't break anything
         if len(entity.split()) > 1:
             pn = PersonalName(entity)
@@ -464,7 +502,7 @@ class VIAFClient:
                     max_similarity = score
                     most_similar_record = var
                 if score >= threshold:
-                    logger.info(
+                    logger.debug(
                         f"Verification successful! '{name_form}' sufficiently " \
                         f"similar to '{var}'! Score = {score}."
                     )
@@ -486,6 +524,10 @@ class VIAFClient:
         """ Takes in n VIAFRecords found while searching the term `search_term`.
         Returns the most similar VIAFRecord.
         """
+        logger.debug(
+            f"Retrieving a single verified record from VIAF search results. " \
+            f"search term = '{search_term}'."
+        )
         verified_record = None
         max_score = 0
         most_similar_record = ""
@@ -517,6 +559,7 @@ class VIAFClient:
      ) -> Response:
         """ Query VIAF records by search term.
         """
+        logger.debug(f"Retriecing VIAF records for search term '{search_term}'.")
         if field and field not in ALLOWED_VIAF_FIELDS:
             logger.error(
                 f"Field '{field}' is not allowed. Defaulting to '{DEFAULT_VIAF_FIELD}'. " \
@@ -541,6 +584,7 @@ class VIAFClient:
     def get_records_by_viaf_id(self, record_id: str) -> Response:
         """ Query VIAF records by ID.
         """
+        logger.debug(f"Retrieving VIAF records for ID {record_id}.")
         data = {
             "reqValues": {
                 "recordId": str(record_id)
@@ -552,6 +596,7 @@ class VIAFClient:
     def extract_viaf_ids(self, search_query_response: Response) -> List[str]:
         """ Parse VIAF ID-s from search query response.
         """
+        logger.debug("Extracting VIAF IDs from VIAF search query results.")
         try:
             records = search_query_response.json()["queryResult"]["records"]["record"]
         except Exception as e:
@@ -591,7 +636,6 @@ class VIAFClient:
 
     def fetch_viaf_clusters(self, viaf_ids: List[str]) -> Dict[str, dict]:
         results = {}
-
         for viaf_id in viaf_ids:
             try:
                 response = self.get_records_by_viaf_id(viaf_id)
@@ -605,9 +649,13 @@ class VIAFClient:
 
     def get_normalized_data_by_ids(self, record_ids: List[str]) -> List[VIAFRecord]:
         """ Fetch data required for normalization from VIAF. """
+        logger.debug(f"Fetching VIAFRecords for the following IDs: {record_ids}.")
         response = self.fetch_viaf_clusters(record_ids)
         viaf_records = [
-            VIAFRecord(response[record_id])
+            VIAFRecord(
+                record=response[record_id],
+                allowed_sources=self.allowed_viaf_sources
+            )
             for record_id in record_ids
         ]
         return viaf_records
@@ -618,6 +666,11 @@ class VIAFClient:
         viaf_index: str = "VIAF"
     ) -> VIAFRecord | None:
         """ Fetch data required for normalization from VIAF. """
+        logger.debug(
+            f"Finding VIAFRecords with search term '{search_term}' " \
+            f"using VIAF field='{field}', verify={verify}, threshold={threshold}. " \
+            f"Allowed VIAF sources are: {self.allowed_viaf_sources}."
+        )
         viaf_record = None
         viaf_ids = self.get_viaf_ids_by_search_terms(
             search_term=search_term,
@@ -637,17 +690,3 @@ class VIAFClient:
                 records = self.get_normalized_data_by_ids(viaf_ids[:1])
                 verified_record = records[0] if records else None
         return verified_record
-
-
-
-if __name__ == "__main__":
-    from pprint import pprint
-    vc = VIAFClient()
-    entity="Kevade"
-    record = vc.get_normalized_data_by_search_term(entity,  field="local.uniformTitleWorks", max_records=5, verify=True)
-    #pprint(record.record_data)
-    if record:
-        pprint(record.all_fields)
-        #pprint(record.record_data)
-    else:
-        print(f"Couldn't detect a verified record for entity '{entity}' :(.")

{rara_tools-0.6.0 → rara_tools-0.6.2/rara_tools.egg-info}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: rara-tools
-Version: 0.6.0
+Version: 0.6.2
 Summary: Tools to support Kata's work.
 Classifier: Programming Language :: Python :: 3
 Classifier: Programming Language :: Python :: 3.10

{rara_tools-0.6.0 → rara_tools-0.6.2}/tests/test_viaf_client.py

@@ -69,3 +69,20 @@ def test_subfield_based_main_field_extraction():
         verify=True
     )
     assert record.name == "Luts, Oskar"
+
+def test_changing_allowed_sources():
+    client = VIAFClient(allowed_viaf_sources=["PLWABN"])
+    record = client.get_normalized_data_by_search_term(
+        search_term="Anora",
+        field="local.uniformTitleWorks",
+        verify=False
+    )
+    assert record.name == "Anora (film)"
+
+    client = VIAFClient(allowed_viaf_sources=["LC"])
+    record = client.get_normalized_data_by_search_term(
+        search_term="Anora",
+        field="local.uniformTitleWorks",
+        verify=False
+    )
+    assert record.name == "Anora (Motion picture)"

rara_tools-0.6.0/VERSION

@@ -1 +0,0 @@
-0.6.0

rara_tools-0.6.0/rara_tools/constants/linker.py

@@ -1,11 +0,0 @@
-COMPONENT_KEY = "linker"
-
-
-class Task:
-    BASE = "base_linker_task"
-    VECTORIZE = "vectorize_text"
-
-
-class Queue:
-    LINKER = "linker"
-    VECTORIZER = "vectorizer"

rara_tools-0.6.0/rara_tools/constants/subject_indexer.py

@@ -1,9 +0,0 @@
-COMPONENT_KEY = "subject_indexer"
-
-
-class Tasks:
-    PIPELINE  = "run_subject_indexer_pipeline"
-
-
-class Queue:
-    MAIN = "subject-indexer"