Script to extract text from a pdf using Apache Tika.

pdf_extraction_tika.py

# imports
from tika import parser
import os
import datetime

# class for extracting tika files
class TikaExtract(object):
    # initialize the object
    def __init__(self, source_directory, target_directory_name):
        # assigned variables for source_directory and target_directory_name
        self.dir = source_directory
        self.target = str(target_directory_name)
    
    # define recursive function to walk through directory and convert pdfs    
    def extract_text_from_pdfs_recursively(self):
        for root, dirs, files in os.walk(self.dir):
            for file in files:
                path_to_pdf = os.path.join(root, file)
                [stem, ext] = os.path.splitext(path_to_pdf)
                if ext == '.pdf':
                    print("Processing " + path_to_pdf)
                    # use tika to parse contents from file
                    pdf_contents = parser.from_file(path_to_pdf)
                    # project specific - convert to raw
                    raw_text = r'{}'.format(pdf_contents['content'])
                    # project specific - replace new lines with spaces
                    raw_text = raw_text.replace("\n"," ")
                    # project specific - replace double new lines with spaces
                    raw_text = raw_text.replace("\n\n" , " ")
                    # project specific - replace tabs with spaces
                    raw_text = raw_text.replace("\t"," ")
                    path_to_txt = stem + '.txt'
                    # check if target directory exists
                    if not os.path.exists(str(os.getcwd()) + self.target):
                        os.makedirs(str(os.getcwd()) + self.target)
                    # write the text file to the target directory
                    # names of the files will be the same, except have the .txt extension
                    with open(str(os.getcwd()) + self.target + str(file[:-4]) + ".txt", 'w') as txt_file:
                        print("Writing contents to " + str(os.getcwd()) + self.target + str(file[:-4]) + ".txt")
                        txt_file.write(raw_text)
                       
# implementation
tikaextract = TikaExtract(source_directory=str(os.getcwd())+'/source_directory/',
                         target_directory_name='/target_directory/')
# run the function
tikaextract.extract_text_from_pdfs_recursively()