fix emblems analyzer test

backend/addcorpus/tests/test_language_analysis.py

@@ -103,25 +103,36 @@ def test_language_analyzer_valid(analyzer_class: Type[LanguageAnalyzer]):
 def tokens(analyzed_response):
     return [token['token'] for token in analyzed_response['tokens']]
 
+
+
+def create_analyzer_test_corpus(es_client: Elasticsearch, analyzer: LanguageAnalyzer) -> str:
+    index_name = f'test-analyzers-{analyzer.code}'
+    settings = es_settings([analyzer.code])
+    es_client.indices.create(index=index_name, settings=settings)
+    return index_name
+
+
+def analyzer_test_output(es_client: Elasticsearch, index_name: str, analyzer_name: str, text: str):
+    response = es_client.indices.analyze(
+        index=index_name,
+        analyzer=analyzer_name,
+        text=text,
+    )
+    return tokens(response.body)
+
+
 @pytest.mark.parametrize('data', analyzer_test_cases,
     ids=[case['analyzer'].__name__ for case in analyzer_test_cases]
 )
 def test_language_analyzer_output(es_client: Elasticsearch, test_index_cleanup, data):
     analyzer: LanguageAnalyzer = data['analyzer']()
 
-    index_name = f'test-analyzers-{analyzer.code}'
-    settings = es_settings([analyzer.code])
-    es_client.indices.create(index=index_name, settings=settings)
+    index_name = create_analyzer_test_corpus(es_client, analyzer)
 
     text = data['input']
 
     def analyzer_output(analyzer_name):
-        response = es_client.indices.analyze(
-            index=index_name,
-            analyzer=analyzer_name,
-            text=text,
-        )
-        return tokens(response.body)
+        return analyzer_test_output(es_client, index_name, analyzer_name, text)
 
     assert analyzer_output(analyzer.standard_analyzer_name) == data['standard']
 

backend/corpora/emblems/tests/test_emblems_analyzer.py

@@ -1,5 +1,6 @@
 from elasticsearch import Elasticsearch
 from corpora.emblems.analyzer import CustomEmblemsAnalyzer
+from addcorpus.tests.test_language_analysis import analyzer_test_output
 
 def test_custom_emblems_analyzer():
     analyzer = CustomEmblemsAnalyzer()
@@ -10,12 +11,24 @@ def test_custom_emblems_analyzer():
     ('hæc', ['haec']),
 ]
 
-def test_custom_emblems_analyzer_matching(es_client: Elasticsearch):
+
+def test_custom_emblems_analyzer_matching(es_client: Elasticsearch, test_index_cleanup):
     analyzer = CustomEmblemsAnalyzer()
+    index_name = f'test-analyzers-{analyzer.code}'
+    settings = {
+        'index': {'number_of_shards': 1, 'number_of_replicas': 1},
+        'analysis': {
+            'char_filter': analyzer.char_filters(),
+            'filter': analyzer.token_filters(),
+            'analyzer': analyzer.analyzers(),
+        },
+    }
+    es_client.indices.create(index=index_name, settings=settings)
     for content, expected in _analysis_cases:
-        result = es_client.indices.analyze(
-            text=content,
-            **analyzer._standard_analyzer(),
+        tokens = analyzer_test_output(
+            es_client,
+            index_name,
+            analyzer.standard_analyzer_name,
+            content
         )
-        tokens = [token['token'] for token in result.body['tokens']]
         assert tokens == expected