rara-tools 0.7.15__tar.gz → 0.7.17__tar.gz

{rara_tools-0.7.15/rara_tools.egg-info → rara_tools-0.7.17}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: rara-tools
-Version: 0.7.15
+Version: 0.7.17
 Summary: Tools to support Kata's work.
 Classifier: Programming Language :: Python :: 3
 Classifier: Programming Language :: Python :: 3.10

rara_tools-0.7.17/VERSION

@@ -0,0 +1 @@
+0.7.17

{rara_tools-0.7.15 → rara_tools-0.7.17}/rara_tools/normalizers/base.py

@@ -3,6 +3,8 @@ from pymarc import (Field, Subfield, JSONReader, Record)
 from typing import List, Optional, Iterator
 from rara_tools.normalizers.reader import SafeJSONReader
 
+from rara_tools.parsers.tools.validators import filter_names
+
 from rara_tools.normalizers.viaf import VIAFRecord, VIAFClient
 from rara_tools.constants.normalizers import (
     DEFAULT_VIAF_FIELD, ALLOWED_VIAF_FIELDS, ALLOWED_VIAF_WIKILINK_LANGS,
@@ -51,10 +53,23 @@ class RecordNormalizer:
         If one linked entity found, we create an updated record from the linked entity data.
         """
         linked_records = []
+        
+        def handle_create_new_record(entity, idx):
+            logger.info(f"No linked entities found for {entity}, Creating new record.")
+            linked_records.append({
+                "leader": self.DEFAULT_LEADER,
+                "fields": []
+            })
+            self.records_extra_data.append({
+                "entity": entity,
+                "classified_fields": classified_fields[idx] if idx < len(classified_fields) else [],
+                "edited": False,
+            })
     
         for idx, linked in enumerate(linking_results or []):
             
             if not isinstance(linked, dict):
+                logger.debug(f"Skipping invalid linked result: {linked}")
                 continue
 
             entity = linked.get("original_entity")
@@ -62,46 +77,32 @@ class RecordNormalizer:
 
             if not isinstance(linked_info, list) or not linked_info:
                 # No linked entities found, create new record
-                logger.info(
-                    f"No linked entities found for {entity}, Creating new record.")
-                linked_records.append({
-                    "leader": self.DEFAULT_LEADER,
-                    "fields": []
-                })
-                self.records_extra_data.append({
-                    "entity": entity,
-                    "classified_fields": classified_fields[idx] if idx < len(classified_fields) else [],
-                    "edited": False
-                })
+                handle_create_new_record(entity, idx)
                 continue
             
             elif len(linked_info) > 1:
                 # Multiple linked entities found, create new record
-                logger.info(
-                    f"Multiple linked entities found for {entity}. Creating new record.")
-                linked_records.append({
-                    "leader": self.DEFAULT_LEADER,
-                    "fields": []
-                })
-                self.records_extra_data.append({
-                    "entity": entity,
-                    "classified_fields": classified_fields[idx] if idx < len(classified_fields) else [],
-                    "edited": False
-                })
+                handle_create_new_record(entity, idx)
                 continue   
 
             elif len(linked_info) == 1:
+                # one record match found, we update existing record
+                
                 linked_item = linked_info[0]
                 if not isinstance(linked_item, dict):
                     continue
                 
+                # handle case where we have linked an entity without a record
+                if not linked_item.get("json", None):
+                    handle_create_new_record(entity, idx)
+                    continue
+                
                 linked_records.append(linked_item.get("json", {}))
                 
                 self.records_extra_data.append({
                     "entity": entity,
                     "viaf": linked_item.get("viaf", {}),
                     "classified_fields": classified_fields[idx] if idx < len(classified_fields) else [],
-                    "type": "linked",
                     "edited": True
                 })
                 continue
@@ -109,7 +110,6 @@ class RecordNormalizer:
         self.records_extra_data.extend(
             {
                 "sierraID": obj.get("sierraID"),
-                "type": "sierra",
                 "edited": True
             }
             for obj in (sierra_data or [])
@@ -313,25 +313,34 @@ class RecordNormalizer:
         if viaf_record:
             self._include_name_variations(record, viaf_record)
             
-    def _include_name_variations(self, record: Record, viaf_record: VIAFRecord) -> None:
+    def _include_name_variations(self, record: Record, viaf_record: VIAFRecord, filter_variations=True) -> None:
         """ Include name variations from VIAF record as 400|t fields """
         
         if not viaf_record or not viaf_record.name_variations:
             return
         
         existing_name_variations = record.get_fields("400")
-        existing_variations = [sf.value for field in existing_name_variations for sf in field.get_subfields("t")]
+        existing_variations = [sf.value for field in existing_name_variations for sf in field.get_subfields("a")]
+        
+        if filter_variations:
+            allowed_variations = filter_names(viaf_record.name_variations)
+            logger.debug(
+                f"filtered out {len(viaf_record.name_variations) - len(allowed_variations)} name variations for '{viaf_record.name}'"
+            )
+            
+        else:
+            allowed_variations = viaf_record.name_variations
         
         fields = []
         
-        for variation in viaf_record.name_variations:
+        for variation in allowed_variations:
             if variation not in existing_variations:
                 fields.append(
                     Field(
                         tag="400",
                         indicators=EMPTY_INDICATORS,
                         subfields=[
-                            Subfield("t", variation)
+                            Subfield("a", variation)
                         ]
                     )
                 )
@@ -465,6 +474,8 @@ class RecordNormalizer:
                         verify=verify,
                         threshold=threshold
                     )
+                    if viaf_record:
+                        logger.debug(f"VIAF {search_term}, linked to ID: {viaf_record.viaf_id}")
 
         except Exception as e:
             logger.error(
@@ -473,7 +484,8 @@ class RecordNormalizer:
         return viaf_record
 
     def _normalize_record(self, record: Record, sierraID: str,
-                          viaf_record: VIAFRecord, is_editing_existing_record: bool, original_entity: str) -> Record:
+                          viaf_record: VIAFRecord, is_editing_existing_record: bool, 
+                          original_entity: str) -> Record:
         return record
     
     def get_record(self, index: int) -> Record:
@@ -500,19 +512,26 @@ class RecordNormalizer:
         return next(iter(self))
       
     def __iter__(self) -> Iterator:
-        viaf_id_path = "viaf.queryResult.records.record.0.recordData.VIAFCluster.viafID"
+        # viaf_id_path = "viaf.original.queryResult.viafID"
+        viaf_id_path = "viaf.parsed.viaf_id"
+        
         sierra_id_path = "sierraID"
-
+        
         for record, extra_data in zip(self.records, self.records_extra_data):
             
             sierra_id = glom(extra_data, sierra_id_path, default="")
             viaf_id = glom(extra_data, viaf_id_path, default=None)
-            classified_fields = extra_data.get("classified_fields", [])
             
+            classified_fields = extra_data.get("classified_fields", [])
             entity = extra_data.get("entity")
             is_editing_existing_record = extra_data.get("edited") == True
 
             viaf_record = self._get_viaf_record(record, viaf_id, entity)
+            if viaf_record:
+                logger.debug(
+                    f"linked VIAF record with ID {viaf_record.viaf_id} for entity '{entity}'"
+                )
+            
             record = self._normalize_common(record, is_editing_existing_record, classified_fields)
 
             normalized_record = self._normalize_record(

{rara_tools-0.7.15 → rara_tools-0.7.17}/rara_tools/normalizers/bibs.py

@@ -73,26 +73,13 @@ class BibRecordNormalizer(RecordNormalizer):
             
           
     def _normalize_viaf(self, record: Record, viaf_record: VIAFRecord, original_entity: str) -> None:
-        
         if not viaf_record:
             # viaf record not found, include original entity as 100|t
             self._add_author(record, viaf_record=None, original_entity=original_entity)
             return record
-
-        viaf_id = viaf_record.viaf_id
-        fields = [
-            Field(
-                tag="035",
-                indicators=EMPTY_INDICATORS,
-                subfields=[
-                    Subfield("a", viaf_id)
-                ]
-            )
-        ]
-
-        self._add_fields_to_record(record, fields)
+        
         self._add_author(record, viaf_record, original_entity=original_entity)
-
+        
     def _normalize_record(self, record: Record, sierraID: str,
                           viaf_record: VIAFRecord, is_editing_existing_record: bool, original_entity: str) -> Record:
 

rara_tools-0.7.17/rara_tools/parsers/tools/validators.py

@@ -0,0 +1,54 @@
+import regex as re
+from typing import List
+
+def has_valid_chars(entity: str, allow_cyrillic: bool = True) -> bool:
+    """ Checks if entity contains any valid characters in latin
+    or in cyrillic, if the latter is enabled
+
+    Parameters
+    ------------
+    entity: str
+        String to validate.
+    allow_cyrillic: bool
+        Allow strings in cyrillic?
+
+    Returns
+    ------------
+    bool
+        Boolean value indicating, if the string
+        contains any valid characters.
+
+    """
+    # Check for latin characters
+    is_valid = bool(re.search(r"[a-züõöäA-ZÜÕÖÄ]", entity))
+
+    if allow_cyrillic and not is_valid:
+        # If cyrillic characters are allowed,
+        # check for them as well
+        is_valid = bool(re.search(r"[а-яА-Я]", entity))
+
+    return is_valid
+
+
+def filter_names(names: List[str], allow_cyrillic: bool = True) -> List[str]:
+    """ Filters out names not in allowed encodings (latin / cyrillic).
+
+    Parameters
+    ------------
+    names: List[str]
+        Names to filters.
+    allow_cyrillic: bool
+        Allow strings in cyrillic?
+
+    Returns
+    ------------
+    List[str]
+        List of filtered names.
+
+    """
+    filtered_names = [
+        name for name in names
+        if has_valid_chars(entity=name, allow_cyrillic=allow_cyrillic)
+    ]
+    return filtered_names
+    

{rara_tools-0.7.15 → rara_tools-0.7.17/rara_tools.egg-info}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: rara-tools
-Version: 0.7.15
+Version: 0.7.17
 Summary: Tools to support Kata's work.
 Classifier: Programming Language :: Python :: 3
 Classifier: Programming Language :: Python :: 3.10

{rara_tools-0.7.15 → rara_tools-0.7.17}/rara_tools.egg-info/SOURCES.txt

@@ -49,6 +49,7 @@ rara_tools/parsers/marc_records/title_record.py
 rara_tools/parsers/tools/entity_normalizers.py
 rara_tools/parsers/tools/marc_converter.py
 rara_tools/parsers/tools/russian_transliterator.py
+rara_tools/parsers/tools/validators.py
 tests/test_digar_schema_converter.py
 tests/test_elastic.py
 tests/test_elastic_vector_and_search_operations.py
@@ -61,4 +62,5 @@ tests/test_s3_file_operations.py
 tests/test_sierra_converters.py
 tests/test_task_reporter.py
 tests/test_utils.py
+tests/test_validators.py
 tests/test_viaf_client.py

{rara_tools-0.7.15 → rara_tools-0.7.17}/tests/test_normalization.py

@@ -5,6 +5,7 @@ from rara_tools.constants import YYMMDD_FORMAT
 from rara_tools.normalizers import (BibRecordNormalizer, AuthoritiesRecordNormalizer)
 from tests.test_utils import (get_linker_res_example, get_formatted_sierra_response,
                         check_record_tags_sorted, check_no_dupe_tag_values, check_record_tags_have_values)
+from rara_tools.normalizers.viaf import VIAFRecord
 
 from rara_tools.constants.linker import EntityType
 
@@ -212,7 +213,7 @@ def test_missing_fields_created_bibrecord_normalization():
     for record in normalizer_entities_only:
         check_record_tags_have_values(
             record, ["008",  # Sierra related, always with bibs
-                     "035", "100",  # VIAf enriched
+                      "100",  # VIAf enriched
                      ] + REQUIRED_FIELDS
         )
         validate_bibrecord_normalized(record, has_viaf_data=True)
@@ -752,7 +753,7 @@ def test_classified_fields_added_to_linked_record():
                     }
                 ]
             }
-        }
+        } 
     ]
     ]
     # Case 1 - no 670 exists, should be added to linked record
@@ -765,7 +766,7 @@ def test_classified_fields_added_to_linked_record():
         assert len(fields_670) == 1
         assert fields_670[0].get_subfields("a")[0] == "Päikesekiri, 2021"
     
-    # Case 1 - existing record with 670 should not update (same behavior for both normalizers)
+    # Case 2 - existing record with 670 should not update (same behavior for both normalizers)
     linker_res = get_linker_res_example(
         "oneFound.json")
     linking_results = [linker_res]
@@ -779,7 +780,54 @@ def test_classified_fields_added_to_linked_record():
         fields_670 = record.get_fields("670")
         assert len(fields_670) == 1
         assert fields_670[0].get_subfields("a")[0] == "Eesti kirjarahva leksikon, 1995."
+    
+    def get_046_field(year: str) -> dict:
+        return {
+            "046": {
+                "ind1": " ",
+                "ind2": " ",
+                "subfields": [
+                    {"k": year }
+                ]
+            }
+        }    
         
+    # Case 3 - 046 $k - publication date Passed for bib
+    classified_fields = [
+        [get_046_field("2021")], 
+        [get_046_field("1999")], 
+        [get_046_field("2022")]
+    ]
+    
+    
+    mock_046_exists = MOCK_LINKER_ONE_FOUND.copy()
+    mock_046_exists["linked_info"][0]["json"]["fields"].append(get_046_field("2000"))
+    
+    # for new record should get included
+    linking_results = [MOCK_LINKER_NOT_FOUND, # new record
+                       MOCK_LINKER_ONE_FOUND, # new record
+                       MOCK_LINKER_NOT_FOUND] # editing existing record
+    
+    normalizer = BibRecordNormalizer(linking_results=linking_results, classified_fields=classified_fields)
+    
+    # for i, record in enumerate(normalizer):
+        # first two should have 046 from classified data
+    record1 = normalizer.get_record(0)
+    fields_046 = record1.get_fields("046")
+    assert len(fields_046) == 1
+    assert fields_046[0].get_subfields("k")[0] == "2021"
+    
+    record2 = normalizer.get_record(1)
+    fields_046 = record2.get_fields("046")
+    assert len(fields_046) == 1
+    # should be unchanged, aka 2000
+    assert fields_046[0].get_subfields("k")[0] == "2000"
+    
+    record3 = normalizer.get_record(2)
+    fields_046 = record3.get_fields("046")
+    assert len(fields_046) == 1
+    assert fields_046[0].get_subfields("k")[0] == "2022"
+    
 def test_classified_data_with_multiple_records():
     """ Test classified data with multiple records - should match by sierraID """
     
@@ -842,3 +890,62 @@ def test_classified_data_with_multiple_records():
         assert len(record.get_fields("670")) == 1
         fields_670 = record.get_fields("670")[0]
         assert fields_670.get_subfields("a")[0] == "Teine kirjeldus, 2022"
+        
+        
+def test_viaf_name_variations():
+    """ Test adding alternative name forms from VIAF to 4XX fields. Should skip some variants """
+    
+    normalizer = AuthoritiesRecordNormalizer()
+    record = Record()
+    
+    viaf_record: VIAFRecord = normalizer._get_viaf_record(
+        record,
+        entity="Jaan Kaplinski"
+    )
+    
+    assert viaf_record is not None
+    assert len(viaf_record.name_variations) > 0
+    
+    normalizer._add_author(record, viaf_record)
+    
+    fields_4xx = record.get_fields("400") + record.get_fields("410") + record.get_fields("430")
+    
+    unfiltered_name_variations = viaf_record.name_variations
+    
+    assert len(fields_4xx) > 0
+    assert len(fields_4xx) < len(unfiltered_name_variations)
+    
+def test_existing_record_linked_to_viaf_record():
+    """ Test existing record linked to VIAF record - should enrich with VIAF data """
+    
+    base_path = "tests/test_data/marc_records/json/"
+    with open(os.path.join(base_path, "imbi.json"), "r", encoding="utf-8") as f, \
+        open(os.path.join(base_path, "ernits.json"), "r", encoding="utf-8") as f2, \
+        open(os.path.join(base_path, "rowling.json"), "r", encoding="utf-8") as f3:
+        imbi = json.load(f)
+        ernits = json.load(f2)
+        rowling = json.load(f3)
+        
+    linking_results = [
+        imbi, 
+        ernits,
+        rowling
+    ]
+    
+    normalizer = AuthoritiesRecordNormalizer(
+        linking_results=linking_results,
+    )
+    
+    def get_viaf_url(record: Record):
+        field_024 = record.get_fields("024")
+        if len(field_024) == 0:
+            return None
+        return field_024[0].get_subfields("0")[0]
+    
+    viaf_base_url = "http://viaf.org/viaf"
+    assert get_viaf_url(normalizer.get_record(0)) == f"{viaf_base_url}/167120147/"
+    assert get_viaf_url(normalizer.get_record(1)) == f"{viaf_base_url}/22458146/"
+    assert get_viaf_url(normalizer.get_record(1)) == f"{viaf_base_url}/22458146/"
+    assert get_viaf_url(normalizer.get_record(2)) == f"{viaf_base_url}/116796842/"
+    
+ 

rara_tools-0.7.17/tests/test_validators.py

@@ -0,0 +1,55 @@
+from rara_tools.parsers.tools.validators import filter_names 
+import pytest
+
+are_equal = lambda x, y: not bool(set(x).difference(set(y)))
+
+names_to_validate = [
+    "ליסט, פראנץ",
+    "Liszt, Franz",
+    "Lißt, Franz",
+    "ליסט, פרנץ",
+    "Liszt, Ferencz",
+    "Лист, Франц",
+    "Listz",
+    "Lißzt, Franz",
+    "Lists, Francis",
+    "List, Ferenc",
+    "List, Frants리스",
+    "List, Ferents",
+    "李斯特，弗朗西斯庫斯",
+    "ᓕᔅᑦ, ᕗᕌᓐᓯᔅᑲᔅ",
+    "리스트, 프란치스코"
+]
+
+valid_names_1 = [
+    "Liszt, Franz",
+    "Lißt, Franz",
+    "Liszt, Ferencz",
+    "Лист, Франц",
+    "Listz",
+    "Lißzt, Franz",
+    "Lists, Francis",
+    "List, Ferenc",
+    "List, Frants리스",
+    "List, Ferents"
+]
+
+valid_names_2 = [
+    "Liszt, Franz",
+    "Lißt, Franz",
+    "Liszt, Ferencz",
+    "Listz",
+    "Lißzt, Franz",
+    "Lists, Francis",
+    "List, Ferenc",
+    "List, Frants리스",
+    "List, Ferents"
+]
+
+def test_filtering_latin_cyrillic():
+    filtered_names = filter_names(names_to_validate, allow_cyrillic=True)
+    assert are_equal(filtered_names, valid_names_1)
+
+def test_filtering_latin():
+    filtered_names = filter_names(names_to_validate, allow_cyrillic=False)
+    assert are_equal(filtered_names, valid_names_2)

rara_tools-0.7.15/VERSION

@@ -1 +0,0 @@
-0.7.15