
    R<
iw                     8    d dl Z d dlZd dlZd dlZd dlZd Zd Zy)    Nc                 b    t        j                  dd|       } t        j                  dd|       } | S )Nz
(\d)\s(\d)z\1\2zR\$\s*(\d+)\s*,\s*(\d{2})zR$\1,\2)resub)values    ?C:\Users\bmuel\python-api\convertPdfToCsv\convert_pdf_to_csv.pyfix_currencyr      s,    FF='51EFF/UCEL    c                 b   t         j                  j                  t         j                  j                  t                    }t         j                  j                  |d      }t        | d      5 }t        j                  |      }|j                  d   }|j                         j                  d      d   }d d d        t        j                  | |dd       t        |dd	
      5 }t        t        j                   |            }	d d d        	dd  D ](  }
t#        |
      D ]  \  }}d|v st%        |      |
|<    * t        |ddd	      5 }t        j&                  |      }|j)                  g       |j+                  |	       d d d        |S # 1 sw Y   xY w# 1 sw Y   xY w# 1 sw Y   |S xY w)Nzextracted_text.txtrbr   .csvall)output_formatpagesrzutf-8)encoding   zR$w )newliner   )ospathdirnameabspath__file__joinopenPyPDF2	PdfReaderr   extract_textsplittabulaconvert_intolistr   reader	enumerater   writerwriterow	writerows)	file_path
output_csv
script_dir
output_txtfile
pdf_reader
first_pagefirst_sentencecsv_filedatarowicolr'   s                 r   process_pdfr7      sd   !:;Jj*>?J	i	$%%d+
%%a(
#00288=a@ 

 	:U%P	j#	0HCJJx() 
1 ABxnFAss{%c*A % 
 
j#rG	<H%() 
=
 ) 
	 
1	0 
=
 s%   -AF#F	9F$FF!$F.)r"   r   r   r   r   r   r7    r	   r   <module>r9      s      
 	 	
r	   