Extracting Text from Docx, Doc and Pdf files Using Python

import os
import docx # pip install python-docx # Does not support .pdf and .doc
import PyPDF2

from docx import Document

SUPPORTED_FORMATS = ['pdf', 'doc', 'docx']
WORD_FORMATS = ['doc', 'docx']

For DOCX

f_list = []
for dirpath, subdirs, files in os.walk("."):
    for f in files:
        if f.split(".")[1] == "docx" and f[0] != "~":
            
            f_list.append(os.path.join(dirpath, f))

d_list = []
for f in f_list:
    d_list.append(Document(f))
    
t_list = []
for d in d_list:
    para_text = ""
    for para in d.paragraphs:
        para_text = para_text + " " + para.text
    t_list.append(para_text) 
    
FOR PDF
f_list = []
for dirpath, subdirs, files in os.walk("."):
    for f in files:
        if f.split(".")[1] == "pdf" and f[0] != "~":
            #print(f, dirpath)
            f_list.append(os.path.join(dirpath, f))

d_list = []
t_list = []

for f in f_list:
    pdfFileObj = open(f, 'rb')
    pdfReader = PyPDF2.PdfFileReader(pdfFileObj)
    d_list.append(pdfReader)
    
    text = ""
    for pageObj in pdfReader.pages:
        text = text + " " + pageObj.extractText()

    t_list.append(text)
    
    pdfFileObj.close()


For DOC

import win32com.client

word = win32com.client.Dispatch("Word.Application")
word.visible = False
wb = word.Documents.Open(r'D:\xyz.doc')
doc = word.ActiveDocument
print(doc.Range().Text) 
Tags: Technology,Python,Natural Language Processing,