Update cleaner.py

## Change to support <data> instead of data. For bidDatasets

cleaner.py

@@ -57,7 +57,7 @@ class Cleaner:
         O espaço antes e depois do token é por conta do padrão que retira os espaços antes e depois do texto.
         """
         paragraphs = re.sub(r'((\s)-)+', ' - ', paragraphs)
-        return re.sub(r'[-–]+', '-', paragraphs)
+        return re.sub(r'(([-]){2,})', '-', paragraphs)
     
     @staticmethod
     def _remove_multiples_dots(paragraphs):
@@ -167,7 +167,7 @@ class Corretor:
         texts = self._tokenizaEmail(texts)
         texts = self._tokenizaData(texts)
         texts = self._tokenizaHora(texts)
-        texts = self._tokenizaNumero(texts)
+        # texts = self._tokenizaNumero(texts)
         texts = self._tokenizaNumeroRomano(texts)
         texts = self._reduzNumeros(texts)
         texts = self._removeHifenInicial(texts)
@@ -436,49 +436,49 @@ class Corretor:
         """
         Tokeniza URL.
         """
-        return re.sub(r'(((https?:\/\/)(www\.))|(www\.)|(https?:\/\/))[-a-zA-Z0-9@:%.\+~#=]{1,256}\.[a-zA-Z@0-9()]{1,6}\b([-a-zA-Z0-9()@:%\+.~#?&\/=]*)', ' url ', paragraphs)
+        return re.sub(r'(((https?:\/\/)(www\.))|(www\.)|(https?:\/\/))[-a-zA-Z0-9@:%.\+~#=]{1,256}\.[a-zA-Z@0-9()]{1,6}\b([-a-zA-Z0-9()@:%\+.~#?&\/=]*)', '<URL>', paragraphs)
     
     @staticmethod
     def _tokenizaEmail(paragraphs):
         """
         Tokeniza email.
         """
-        return re.sub(r'(\b([a-zA-Z0-9._%+-]+@[a-zA-Z0-9.-]+\.[a-zA-Z]{2,})\b)', ' mail ', paragraphs)
+        return re.sub(r'(\b([a-zA-Z0-9._%+-]+@[a-zA-Z0-9.-]+\.[a-zA-Z]{2,})\b)', '<EMAIL>', paragraphs)
 
     @staticmethod
     def _tokenizaData(paragraphs):
         """
         Tokeniza data.
         """
-        return re.sub(r'(\b([0-3][0-9]\/[0-1][0-9]\/(([0-9]{2})|([0-2][0-9]{3})))\b)', ' date ', paragraphs)
+        return re.sub(r'(\b([0-3][0-9]\/[0-1][0-9]\/(([0-9]{2})|([0-2][0-9]{3})))\b)', '<DATA>', paragraphs)
 
     @staticmethod
     def _tokenizaHora(paragraphs):
         """
         Tokeniza hora.
         """
-        return re.sub(r'(\b(([0-1][0-9])|(2[0-3]))(\:|h)([0-5][0-9])?\b)', ' hour ', paragraphs)
+        return re.sub(r'(\b(([0-1][0-9])|(2[0-3]))(\:|h)([0-5][0-9])?\b)', '<HORA>', paragraphs)
     
     @staticmethod
     def _tokenizaNumero(paragraphs):
         """
         Tokeniza número.
         """
-        return re.sub(r'([0-9])+', ' number ', paragraphs)
+        return re.sub(r'(\b([0-9]+)\b)', '<NUMERO>', paragraphs)
     
     @staticmethod
     def _tokenizaNumeroRomano(paragraphs):
         """
         Tokeniza número romano.
         """
-        return re.sub(r"(\s|\.|\,|\;|\:|^)(?=[XVIΙ])(XC|XL|L?X{0,3})([IΙ]X|[IΙ]V|V?[IΙ]{0,3})(\s|\.|\,|\;|\:|$)", ' number ', paragraphs, flags=re.IGNORECASE)
+        return re.sub(r"(\s|\.|\,|\;|\:|^)(?=[XVIΙ])(XC|XL|L?X{0,3})([IΙ]X|[IΙ]V|V?[IΙ]{0,3})(\s|\.|\,|\;|\:|$)", '<NUMERO>', paragraphs, flags=re.IGNORECASE)
     
     @staticmethod
     def _reduzNumeros(paragraphs):
         """
         Reduz números.
         """
-        return re.sub(r'(([\.\\\/\;\:\s])*(number([\-–\.\\\/\;\:\,\s])*)+)', ' number ', paragraphs)
+        return re.sub(r'(([\.\\\/\;\:\s])*(<NUMERO>([\-–\.\\\/\;\:\,\s])*)+)', ' <NUMERO> ', paragraphs)
 
     @staticmethod
     def _removeHifenInicial(paragraphs):
@@ -583,7 +583,6 @@ class Remover:
             return ''
         else:
             return paragraphs
-        
 
 def executaLimpeza(dataframe: pd.DataFrame, column: str, cased, accents) -> pd.DataFrame:
     """