keyword annoation module

text_annotator.py

@@ -0,0 +1,45 @@
+import re
+
+def generate_annotated_text(text, keyw_list):
+    '''
+    :param text: str
+    :param keyw_list: list of str
+    :return: str
+
+    e.g. input1 = 'I like an apple. Do you like apples too?'
+         input2 = ["like", "apple"]
+         output = annotated_text("I ", ("like", ""), " an ", ("apple", ""), ". Do you ", ("like", ""), " ", ("apple", ""), "s too?")
+    '''
+    def find_keyword_index(text, keyw_list):
+        indices = []
+
+        for keyword in keyw_list:
+            for match in re.finditer(keyword, text):
+                indices.append((keyword, match.start()))
+
+        return indices
+
+    indices = find_keyword_index(text, keyw_list)
+
+    # 1. indices를 index 값에 따라 오름차순으로 정렬
+    sorted_indices = sorted(indices, key=lambda x: x[1])
+
+    output = 'annotated_text('
+    last_index = 0
+
+    # 2. input1에서 각 키워드의 위치를 활용하여 문자열 분할 및 재처리
+    for word, start_idx in sorted_indices:
+        output += f'"{text[last_index:start_idx]}", ("{word}", ""), '
+        last_index = start_idx + len(word)
+
+    # 3. 결과 문자열 생성
+    output += f'"{text[last_index:]}"' + ')'
+
+    # 4. 연속 문자열 대치
+    string_cont_type1 = re.compile(r'", ""\), " ", \("')
+    string_cont_type2 = re.compile(r'", ""\), "", \("')
+    output = string_cont_type1.sub(' ', output)
+    output = string_cont_type2.sub(' ', output)
+
+    return output
+