File: //usr/lib/python3/dist-packages/pdfminer/__pycache__/encodingdb.cpython-38.pyc
U
�R�]�
� @ st d dl Z d dlZd dlZddlmZ ddlmZ ddlmZ e� d�Z
e � e �Z
dd� Zd d
� ZG d d � d e�ZdS )
� N� )�glyphname2unicode)�ENCODING)� PSLiteralz [0-9a-fA-F]+c s* | � d�d } | � d�}t|�dkr4d�tt|��S | tkrFt�| �S | �d�r�| �d�� t �
� �r�t� �d dkr�� fdd �t dt� �d�D �}|D ] }t |� q�tt
j|�}d�|�S n^| �d
��r| �d
�}t �
|��rdt|� kr�d k�rn n t|d d
�}t |� t
�|�S td| ��dS )a� Converts Adobe glyph names to Unicode numbers.
In contrast to the specification, this raises a KeyError instead of return an empty string when the key is unknown.
This way the caller must explicitly define what to do when there is not a match.
Reference: https://github.com/adobe-type-tools/agl-specification#2-the-mapping
:returns unicode character if name resembles something, otherwise a KeyError
�.r �_r � Zuni� c s$ g | ] }t � ||d � dd��qS )r � ��base)�int)�.0�i�Zname_without_uni� �5/usr/lib/python3/dist-packages/pdfminer/encodingdb.py�
<listcomp>' s z name2unicode.<locals>.<listcomp>�u� r
r zXCould not convert unicode name "%s" to character because it does not match specificationN)�split�len�join�map� name2unicoder �get�
startswith�strip� HEXADECIMAL�match�range�#raise_key_error_for_invalid_unicode�sixZunichrr
�KeyError)�nameZ
componentsZunicode_digitsZdigitZ
charactersZname_without_u�
unicode_digitr r r r s*
*
r c C s( d| k rdk r$n n t d| ��dS )z�Unicode values should not be in the range D800 through DFFF because that is used for surrogate pairs in UTF-16
:raises KeyError if unicode digit is invalid
i�� i � zHUnicode digit %d is invalid because it is in the range D800 through DFFFN)r# )r% r r r r! 8 s r! c @ s� e Zd Zi Zi Zi Zi ZeD ]F\ZZ Z
Z Z e
e�Ze r>eee < e
rJeee
< e rVeee < e r eee < q eeeed�Zeddd��ZdS )�
EncodingDB)ZStandardEncodingZMacRomanEncodingZWinAnsiEncodingZPDFDocEncodingNc C s� | j �|| j�}|r�|�� }d}|D ]j}t|t�r8|}q$t|t�r$zt|j�||< W n0 t k
r� } zt
� t |�� W 5 d }~X Y nX |d7 }q$|S )Nr r )
� encodingsr � std2unicode�copy�
isinstancer
r r r$ r# �log�debug�str)�klassr$ ZdiffZ cid2unicodeZcid�x�er r r � get_encodingY s
zEncodingDB.get_encoding)N)�__name__�
__module__� __qualname__r( Z mac2unicodeZ win2unicodeZ pdf2unicoder r$ ZstdZmac�winZpdfr �cr' � classmethodr1 r r r r r& A s*
�r& )Zlogging�rer" Z glyphlistr Z latin_encr Zpsparserr �compiler Z getLoggerr2 r+ r r! �objectr& r r r r �<module> s
)