o �e�2�@svUdZddlmZddlZddlZddlmZddlmZddl m Z ddlmZm Z mZmZddlmZdd lmZdd lmZmZmZddlmZddlmZmZe�e�Zd Z ddddd�Z!de"d<Gdd�dej#�Z$dZ%Gdd�de�Z&dMdd�Z'dNdd�Z(dOd d!�Z)dPd&d'�Z*dQd+d,�Z+dRd2d3�Z,dSd5d6�Z-dTd9d:�Z.dUd;d<�Z/dVd@dA�Z0dWdFdG�Z1dXdIdJ�Z2dYdKdL�Z3dS)Zz!Interface to Tesseract executable�)�annotationsN)�pi)�fspath)�Path)�PIPE�STDOUT�CalledProcessError�TimeoutExpired)�Version)�Image)�MissingDependencyError�SubprocessOutputError�TesseractConfigError)�OrientationConfidence)�get_version�runaj

��)�autoZotsuz adaptive-otsuZsauvolazdict[str, int]�TESSERACT_THRESHOLDING_METHODSc@seZdZdZdd�ZdS)�TesseractLoggerAdapterz6Prepend [tesseract] to messages emitted from tesseractcCs|j|d<d|��|fS)N�extraz[tesseract] �r)�self�msg�kwargs�r�HD:\Projects\ConvertPro\env\Lib\site-packages\ocrmypdf/_exec/tesseract.py�process9s zTesseractLoggerAdapter.processN)�__name__� __module__�__qualname__�__doc__rrrrrr6sra� v? (?: (?:(?P[0-9]+)!)? # epoch (?P[0-9]+(?:\.[0-9]+)*) # release segment (?P

                                          # pre-release
            [-_\.]?
            (?P(a|b|c|rc|alpha|beta|pre|preview))
            [-_\.]?
            (?P[0-9]+)?
        )?
        (?P                                         # post release
            (?:-(?P[0-9]+))
            |
            (?:
                [-_\.]?
                (?Ppost|rev|r)
                [-_\.]?
                (?P[0-9]+)?
            )
        )?
        (?P                                          # dev release
            [-_\.]?
            (?Pdev)
            [-_\.]?
            (?P[0-9]+)?
        )?
        (?P
            [-_\.]
            (?:20[0-9][0-9] [0-1][0-9] [0-3][0-9])       # yyyy mm dd
        )?
        (?P
            [-_\.]?
            [0-9]+
        )?
        (?P
            [-_\.]?
            g[0-9a-f]{2,10}
        )?
    )
    (?:\+(?P[a-z0-9]+(?:[-_\.][a-z0-9]+)*))?       # local version
c@s,eZdZdZe�dedejejB�Z	dS)�TesseractVersionzLModify standard packaging.Version regex to support Tesseract idiosyncracies.z^\s*z\s*$N)
rr r!r"�re�compile�TESSERACT_VERSION_PATTERN�VERBOSE�
IGNORECASE�_regexrrrrr#js
�r#�return�strcCstddd�S)N�	tesseractztesseract\s(.+))�regex)rrrrr�versionqsr.�boolcCs
t�dkS)z6Does Tesseract have -c thresholding method capability?z5.0)r.rrrr�has_thresholdingus
r0�set[str]c
Cs�dd�}ddg}zt|dttddd�}|j}Wnty+}zt||j��|�d}~ww|��D]
}|�d�r=t||���q0|��^}}dd	�|D�S)
NcSsd}||7}|S)NzSTesseract failed to report available languages.
Output from Tesseract:
-----------
r)�outputrrrr�
lang_error{s�z!get_languages..lang_errorr,z--list-langsT)�text�stdout�stderrZlogs_errors_to_stdout�check�ErrorcSsh|]}|���qSr)�strip)�.0�langrrr�	�sz get_languages..)	rrrr5rrr2�
splitlines�
startswith)r3Z	args_tess�procr2�e�lineZ_header�restrrr�
get_languageszs,	�
��
�rC�langs�	list[str]�engine_mode�
int | NonecCs<dg}|r|�dd�|�g�|dur|�dt|�g�|S)Nr,z-l�+z--oem)�extend�joinr+)rDrF�argsrrr�tess_base_args�srL�
binary_output�bytes�dict[str, str]cs�fdd�}t|��S)Nc3sT������D] }|��}|jddd�}t|�dkr'|d��|d��fVqdS)N�:r)�maxsplitrr)�decoder=r9�split�len)rA�parts�rMrr�gen�s���z$_parse_tesseract_output..gen)�dict)rMrWrrVr�_parse_tesseract_output�s
rY�
input_filer�timeout�floatrc	
Cs�tdg|�ddt|�dg}zt|tt|dd�}Wn=ty(tddd	�YStyV}z#t|j	�t|j
�d
|jvsCd|jvrNtdd�WYd}~St�|�d}~wwt
|j	�}t|�dd��}t|t|�d
d��d	�}|S)N�osd�--psm�0r5T�r5r6r[r7r�)�angle�
confidences&Too few characters. Skipping this page�Image too largezOrientation in degreeszOrientation confidence)rLrrrrr	rr�tesseract_log_outputr5r6r2r
rY�int�getr\)	rZrFr[�args_tesseract�pr@r]rbZorient_confrrr�get_orientation�s2
�



��

�rj�	languagesc

Cs�t||�ddt|�dg}zt|tt|dd�}Wn9ty"YdStyQ}z$t|j�t|j	�d|j
vsB|j
dkrI|jd	krIWYd
}~dSt�|�d
}~wwt
|j�}t|�dd��}d
t|}	|	S)z+Gets angle to deskew this page, in degrees.r^�2r5Tr`ra�Empty page!!�rNzDeskew angler�)rLrrrrr	rrer5r6r2�
returncoder
rYr\rgr)
rZrkrFr[rhrir@�parsedZdeskew_radiansZdeskew_degreesrrr�
get_deskew�s.�


��

rr�stream�NonecCsZttttd�r
tjndd�}|sdSz|��}Wnty'|�dd�}Ynw|��}|D]|}|�d�r6q.|�d�rry�lowerrwr9rSr�info)rsZtlogr4�linesrA�problemrrrre�sH��


�recCs|dkrdSt�d�dS)Nrz+[tesseract] took too long to OCR - skipping)rzry)r[rrr�
page_timedoutsr��output_hocr�output_text�imagecCs\t�|��
}|j\}}Wd�n1swY|jt�||�dd�|jddd�dS)zjProduce a .hocr file that reports no text detected on a page that is
    the same size as the input image.Nru��encoding�[skipped page])r�open�size�
write_text�
HOCR_TEMPLATE�format)r�r�r�Zim�w�hrrr�_generate_null_hocrs
�r�rf�
tessconfig�pagesegmode�thresholdingc
Csb|�d�}t||�}|dur|�dt|�g�|dkr(t�r(|�dd|��g�|	r1|�d|	g�|
r:|�d|
g�|�t|�t|�d	d
g�|�|�zt|tt|dd�}
|
j	}Wn?t
ynt|�t|||�YdSt
y�}z t|j�d
|jvs�d|jvr�t|||�WYd}~dSt�|�d}~wwt|�|�d���r�|�d��|�dSdS)z5Generate a hOCR file, which must be converted to PDF.�Nr^r�-c�thresholding_method=�--user-words�--user-patternsZhocr�txtTr`rdrm�.txt)�with_suffixrLrIr+r0rrrrr5r	r�r�rrer2r
�exists�replace)rZr�r�rkrFr�r[r�r��
user_words�
user_patterns�prefixrhrir5r@rrr�
generate_hocr$s<




���r��
output_pdfcCs|jddd�|�d�dS)Nr�rur�rn)r��write_bytes)r�r�rrr�
use_skip_page_sr�c
Csnt||�}|dur|�dt|�g�|�ddg�|dkr*t�r*|�dd|��g�|	r3|�d|	g�|
r<|�d|
g�|jt|j�}|�t|�t|�d	d
g�|�|�zt|t	t
|dd�}
|
j}|�d
��
�rr|�d
��|�Wn=ty�t|�t||�YdSty�}zt|j�d|jvs�d|jvr�t||�WYd}~dSt�|�d}~wwt|�dS)z�Generate a PDF using Tesseract's internal PDF generator.

    We specifically a text-only PDF which is more suitable for combining with
    the input page.
    Nr^r�ztextonly_pdf=1rr�r�r�Zpdfr�Tr`r�rdrm)rLrIr+r0�parentr�stemrrrrr5r�r�r�r	r�r�rrer2r
)rZr�r�rkrFr�r[r�r�r�r�rhr�rir5r@rrr�generate_pdffs>

�

��r�)r*r+)r*r/)r*r1)rDrErFrGr*rE)rMrNr*rO)rZrrFrGr[r\r*r)
rZrrkrErFrGr[r\r*r\)rsrNr*rt)r[r\r*rt)r�rr�rr�rr*rt)rZrr�rr�rrkrErFrfr�rEr[r\r�rfr�rfr*rt)r�rr�rr*rt)rZrr�rr�rrkrErFrfr�rEr[r\r�rfr�rfr*rt)4r"�
__future__r�loggingr$�mathr�osr�pathlibr�
subprocessrrrr	Zpackaging.versionr
ZPILrZocrmypdf.exceptionsrr
rZocrmypdf.pluginspecrZocrmypdf.subprocessrr�	getLoggerrrzr�r�__annotations__�
LoggerAdapterrr&r#r.r0rCrLrYrjrrrer�r�r�r�r�rrrr�sH
�,




	

 

(



;