import re
from docx2python import docx2python # pip install doc2xpython
from pdfminer.high_level import extract_text # pip install pdfminer.six

# For textract to work with .doc format we need the antiword package, LibreOffice and a symlink hack
# sudo ln -sf /Applications/LibreOffice.app/Contents/MacOS/soffice /usr/local/bin/libreoffice
import textract # pip install textract, pip install antiword

def read_doc(filename):
    try:
        text = textract.process(filename).decode()
        return_result = text
    except Exception as e:
        return_result = "fail"        
    return return_result

def read_docx(filename):
    try:
        return_result = docx2python(filename)
        return_result = return_result.text
    except Exception as e:
        return_result = "fail"        
    return return_result

def read_pdf(filename):
    try:
        return_result = repr(extract_text(filename))
    except Exception as e:
        return_result = "fail"
    return return_result

def file_parser(the_file):
    # Parse your file if needed.
    pass

# List all docx and pdf files in the folder
results_doc = []
results_docx = []
results_pdf = []
docfolder = "files"

for f in os.listdir(docfolder):
    if re.search('.doc$', f):
        results_doc += [f]  
    if re.search('.docx', f):
        results_docx += [f]
    if re.search('.pdf', f):
        results_pdf += [f]

for file in results_doc:
    the_file = read_doc(docfolder + "/" + fil)
    if (the_file == "fail"):
        continue
    if (the_file != "fail"):
        file_parser(the_file)

for file in results_docx:
    the_file = read_docx(docfolder + "/" + fil)
    if (the_file == "fail"):
        continue
    if (the_file != "fail"):
        file_parser(the_file)

for file in results_pdf:
    the_file = read_pdf(docfolder + "/" + fil)
    if (the_file == "fail"):
        continue
    if (the_file != "fail"):
        file_parser(the_file)