Convert PDF files to Microsoft Office Word compatible doc/docx files, using LibreOffice's command line interface.

pdf2docx.py

#!C:/Python27/python.exe
#
# Convert PDF files to Microsoft Office Word compatible doc/docx files, 
# using LibreOffice's command line interface.
#
# http://stackoverflow.com/questions/26358281/convert-pdf-to-doc-python-bash
# http://ask.libreoffice.org/en/question/20111/converting-files-using-soffice-convert-to-with-embedded-images-html-to-doc/
# http://cgit.freedesktop.org/libreoffice/core/tree/filter/source/config/fragments/filters
#

import os
import sys
import subprocess

# pdf source file(s) and target paths
basedir = 'C:/path/to'
pdfdir = os.path.normpath(basedir + '/pdf')
docdir = os.path.normpath(basedir + '/doc')
docxdir = os.path.normpath(basedir + '/docx')

# absolute path to libre office writer application
lowriter = 'C:/Progra~2/LibreO~1/program/swriter.exe'

# output-filter for conversion
#outfilter = ':"Office Open XML Text"'
#outfilter = ':"MS Word 2003 XML"'
#outfilter = ':"MS Word 2007 XML"'
#outfilter = ':"MS Word 97"'
outfilter = ''

i = 0
for top, dirs, files in os.walk(pdfdir):
    for filename in files:
        if filename.endswith('.pdf'):
            i = i + 1
            abspath_pdf = os.path.normpath(os.path.join(top, filename))

            print 'Converting {0} into .doc format..'.format(abspath_pdf)
            subprocess.call('{0} --invisible --convert-to doc{1} --outdir "{2}" "{3}"'
                            .format(lowriter, outfilter, docdir, abspath_pdf), shell=True)

            print 'Converting {0} into .docx format..'.format(abspath_pdf)
            subprocess.call('{0} --invisible --convert-to docx{1} --outdir "{2}" "{3}"'
                            .format(lowriter, outfilter, docxdir, abspath_pdf), shell=True)

    print '|-------------------------------------------------------|'
    print 'Done. Converted {0} pdf files.'.format(i)

using LibreOffice's command line interface???

Anyone active here ?? I need to ask if there is any way of doing the same thing NodeJS

@alimuqaddas The actual work here is done by Libreoffice. The Python code can be replaced by any other language. Try something like this:

    const exec = require('child_process').exec;
    const lowriter = 'C:/Progra~2/LibreO~1/program/swriter.exe';
    
    const outDir = 'C:/abs_outpath/';
    const srcDir = 'C:/abs_srcpath/';

    let files = ['file1.pdf', 'file2.pdf'];
    files.map(pdffile => {
        let cmd = lowriter + ' --invisible --convert-to docx --outdir "' + outDir + '" "' + srcDir + pdffile '"';
        exec(cmd, function(error, stdout, stderr) {
            // command output is in stdout
        });
    });

without using liberoffice is there any other ways using python ?

I need the answer for this too please.. Anyone could achieve this without Liberoffice?

Hi, I'm try to convert a pdf which contains both text and images into a word document. Is this approach suitable for it?
Also, is there any alternative to using libreoffice?

Converting /home/user/PycharmProjects/Test/myfile.pdf into .doc format..
/bin/sh: 1: /usr/share/man/man1/lowriter.1.gz: Exec format error

Converting /home/user/PycharmProjects/Test/myfile.pdf into .docx format..
Done. Converted 1 pdf files.
/bin/sh: 1: /usr/share/man/man1/lowriter.1.gz: Exec format error

I was trying to convert a pdf file to doc following this post...
The above mentioned error appeared when I executed the program..

Plz help me with this...
Thank you.....

Can anyone help me for doing the same in ubuntu16.x? I want to convert pdf to docx without loosing font, font-size, bold etc..
Please help for this.
Thank you.

Can anyone help me for doing the same i. I want to convert pdf to docx without loosing format and layout
Please help for this.
Thank you.

@yashgupta any solution you got?

@pptyasar no as of now i dnt have any solution .do you have any solution for this .Basically i have scanned invoices in pdf and want to convert that to docx format in text format with retaining all the fonts,format and layout

@yashugupta786 try java solution
https://www.pdfonline.com/downloads/easyconvertersdk/demos_easyconvertersdk-word.asp

Apart from running on the command line can I run this on jupyter or pycharm?

I tried converting pdf to docx and it didn't work. Its working from docx to pdf and pdf to ODG.

https://stackoverflow.com/questions/49739245/having-trouble-using-python-and-libreoffice-to-convert-pdf-to-docx-and-doc-to-do

Based on the above link, it seems like libreoffice 6 and higher versions are not supported to convert to doc/docx. Anyone have idea about this?
Please share if we have any other ways to convert pdf to docx using python.

Thank you.

Please try to use lid pdf2docx.
Easy to use.
https://dothinking.github.io/pdf2docx/quickstart.convert.html