3
�gaL � @ sl d dl mZ d dlmZ d dlmZ dZeeef Z e e
d�dd�Z e e
e
d�d d
�Z e
e
ed�d d �Z
d
S )� )�Union)�
from_bytes)�TOO_SMALL_SEQUENCEzutf-8)�content�returnc C s. t }t| �tkr*tt| ��j� }|r*|j}|S )ap
We default to UTF-8 if text too short, because the detection
can return a random encoding leading to confusing results
given the `charset_normalizer` version (< 2.0.5).
>>> too_short = ']"foo"'
>>> detected = from_bytes(too_short.encode()).best().encoding
>>> detected
'ascii'
>>> too_short.encode().decode(detected)
']"foo"'
)�UTF8�lenr r �bytesZbest�encoding)r r
�match� r � /usr/lib/python3.6/encoding.py�detect_encoding s
r )r r
r c C s |s t | �}| j|d�S )z�Decode `content` using the given `encoding`.
If no `encoding` is provided, the best effort is to guess it from `content`.
Unicode errors are replaced.
�replace)r �decode)r r
r r r
� smart_decode s r c C s | j |d�S )zTEncode `content` using the given `encoding`.
Unicode errors are replaced.
r )�encode)r r
r r r
� smart_encode, s r N)Ztypingr Zcharset_normalizerr Zcharset_normalizer.constantr r � bytearrayr Z ContentBytes�strr r r r r r r
�<module> s