U

�R�]��@s�dd
lZddl
mZ
ddlmZ
ddlmZ
ddlmZ
ddlmZ
ddlm	Z	
dd	lm
Z

dd
lmZ
ddl
mZ
ddl
mZ
dd
lZe�e�
Zed
�
Zed�
ZGdd�de�Zd
S)�N�
)
�settings)
�LIT)
�PDFObjectNotFound)
�resolve1)
�	int_value)
�
list_value)
�
dict_value)
�	PDFParser)
�PDFDocument)
�PDFTextExtractionNotAllowedZPage�Pagesc@sJeZ
dZd
Zdd�Zdd�Zedddd	g�
Zed
d��
Z	eddd�
�
Z
dS)�PDFPagea!An object that holds the information about a page.

    A PDFPage object is merely a convenience class that has a set
    of keys and values, which describe the properties of a page
    and point to its contents.

    Attributes:
      doc: a PDFDocument object.
      pageid: any Python object that can uniquely identify the page.
      attrs: a dictionary of page attributes.
      contents: a list of PDFStream objects that represents the page content.
      lastmod: the last modified time of the page.
      resources: a list of resources used by the page.
      mediabox: the physical size of the page.
      cropbox: the crop rectangle of the page.
      rotate: the page rotation (in degree).
      annots: the page annotations.
      beads: a chain that represents natural reading order.
    cCs�|
|_||_
t|�
|_t|j�d
�
�
|_t|j�dt���
|_t|jd�
|_	d|jkrjt|jd�
|_
n|j	|_
t|j�dd��
dd|_|j�d�
|_
|j�d	�
|_d
|jkr�t|jd
�
}ng}t|t�s�|g
}||_dS)z�Initialize a page object.

        doc: a PDFDocument object.
        pageid: any Python object that can uniquely identify the page.
        attrs: a dictionary of page attributes.
        ZLastModified�	Resources�MediaBox�CropBox�Rotater
ih
ZAnnots�
BZContentsN)�doc�pageidr	�attrsr�getZlastmod�dict�	resources�mediaboxZcropboxr�rotateZannotsZbeads�
isinstance�list�contents)�selfrrrr�r �2/usr/lib/python3/dist-packages/pdfminer/pdfpage.py�__init__.s&



















zPDFPage.__init__c

Csd
|j|j
fS)Nz$<PDFPage: Resources=%r, MediaBox=%r>)rr)
rr r r!�__repr__Ks
zPDFPage.__repr__rrrrc
#s���
�fd
d��d}d�jkrL��jd�j�D]\}}�
�||�V
d}q0|s��j
D]\}|��D]N}z4��|�
}t|t�r�|�d�
tkr��
�||�V
Wqbtk
r�


YqbXqbqVdS)Nc	3s�t|t
�r"|}t��|�
�
��}n|j}t|�
��}t�|
�
D]"\}}|�
jkr>||kr>|||<q>|�	d
�
}|dkr�t
js�|�	d�
}|tkr�d|kr�t
�d|d�
t|d�
D]}�||�D]
}|V
q�q�n|tkr�t
�d|�
||fV
dS)N�Type�typeZKidszPages: Kids=%rzPage: %r)r�intr	�getobj�copy�objid�sixZ	iteritems�INHERITABLE_ATTRSrrZSTRICT�
LITERAL_PAGES�log�infor�LITERAL_PAGE)	�obj�parentr)�tree�
k�
vZ	tree_type�
c�
x��document�klass�searchr r!r:Rs&



















z$PDFPage.create_pages.<locals>.searchFr
Tr$)	ZcatalogZxrefsZ
get_objidsr'rrrr/r)r9r8Zpagesr)r2Zxrefr0r r7r!�create_pagesPs 
















zPDFPage.create_pagesNr
�Tccspt|
�
}t
|||d
�}|r,|js,td|
�
�
t|�|�
�
D]0\}	}
|rP|	|krPq:|
V
|r:||	dk
r:
qlq:dS)N)�password�cachingz"Text extraction is not allowed: %rr)r
rZis_extractabler�	enumerater;)r9�fpZpagenosZmaxpagesr=r>Zcheck_extractable�parserrZpagenoZpager r r!�	get_pageszs







zPDFPage.get_pages)Nr
r<TT)�__name__�
__module__�__qualname__�__doc__r"r#�setr+�classmethodr;rBr r r r!rs

)
�r)Zloggingr<rZpsparserrZpdftypesrrrrr	Z	pdfparserr
Zpdfdocumentrrr*Z	getLoggerrCr-r/r,�objectrr r r r!�<module>s