大文字小文字の正規化に対応した

examples/example.py

@@ -7,3 +7,16 @@
 print('`京都`で検索', index['京都'])
 print()
 print('`京都府`で検索', index['京都府'])
+print()
+
+index_2 = naivesearch('./examples/fruits.txt')
+print('`apple`で検索', index_2['apple'])  # Apple
+print()
+print('`APPLE`で検索', index_2['APPLE'])  # Apple
+print()
+print('`Orange`で検索', index_2['Orange'])  # Orange
+print()
+print('`ORANGE`で検索', index_2['ORANGE'])  # Orange
+print()
+print('`京都`で検索', index_2['京都'])  # ヒットしない
+print()

examples/fruits.txt

@@ -9,7 +9,8 @@ Grape
 Mango
 Blueberry
 Pomegranate
-Carambola(U.K) – starfruit (U.S)
+Carambola
+starfruit
 Plum
 Banana
 Raspberry
@@ -25,4 +26,4 @@ Grapefruit
 Melon
 Coconut
 Avocado
-Peachモン
+Peach

naivesearch/indexer/formatter.py

@@ -3,6 +3,12 @@
 
 Formatter = Callable[[str], str]
 
+
 class UnicodeNormalizer:
-    def __call__(self, x :str):
+    def __call__(self, x: str):
         return unicodedata.normalize('NFKC', x)
+
+
+class LowerCaseNormalizer:
+    def __call__(self, x: str):
+        return x.lower()

naivesearch/indexer/inverted_index.py

@@ -11,11 +11,12 @@
 class Reader(Iterable[str]):
     pass
 
+
 class InvertedIndex:
-    index: Dict[str, List[str]] = defaultdict(list)
     chunkers: List[Chunker]
 
     def __init__(self, reader: Reader, chunkers: List[Chunker]):
+        self.index: Dict[str, List[str]] = defaultdict(list)
         self.chunkers = chunkers
 
         logger.info('Start indexing.')
@@ -40,4 +41,3 @@ def __getitem__(self, q):
             result = result & chain
 
         return list(result)
-

naivesearch/indexer/test_formatter.py

@@ -1,6 +1,7 @@
 from naivesearch import InvertedIndex
 from typing import Callable, List, Optional
-from .formatter import UnicodeNormalizer
+from .formatter import UnicodeNormalizer, LowerCaseNormalizer
+
 
 class TestFormatter:
     def test_unicode_normalizer(self):
@@ -10,3 +11,9 @@ def test_unicode_normalizer(self):
         rhs = '⼈⼝'
         assert lhs != rhs
         assert formatter(lhs) == formatter(rhs)
+
+    def test_lower_case_normalizer(self):
+        formatter = LowerCaseNormalizer()
+        lhs = 'UPPER'
+        rhs = 'upper'
+        assert formatter(lhs) == formatter(rhs)

naivesearch/main.py

@@ -1,8 +1,9 @@
 from naivesearch.indexer import InvertedIndex
-from naivesearch.indexer.formatter import UnicodeNormalizer
+from naivesearch.indexer.formatter import UnicodeNormalizer, LowerCaseNormalizer
 from naivesearch.indexer.converter import BigramConverter
 from naivesearch.indexer.chunker import CharacterChunker
 
+
 def naivesearch(filepath: str):
 
     def file_reader(filepath):
@@ -13,8 +14,8 @@ def file_reader(filepath):
     index = InvertedIndex(
         file_reader(filepath),
         [
-            BigramConverter(CharacterChunker(UnicodeNormalizer()))
+            # BigramConverter(CharacterChunker(UnicodeNormalizer())),
+            BigramConverter(CharacterChunker(LowerCaseNormalizer()))
         ]
     )
     return index
-