Add some file metadata in linux file explorer

0_file_explorer_add_metadata.md

Many PDF file names are not explicit, but their title is. I was looking for a mean to show PDF title in file explorer column (when using detailed view), I posted on stackexchange but did not found any solution I was using Thunar, and I changed file explorer to thunar because some python bindings exists. It is Python nautilus project Just install it, on arch it is python2-nautilus package

After, you have just to read /usr/share/doc/nautilus-python/README And check it works adding copying /usr/share/doc/nautilus-python/examples/block-size-column.py to ~/.local/share/nautilus-python/extensions/ and selectring block-size in column list nautilus preferences.

Now I was sure it was possible some custom columns in nautilus detailed view. It was time to show pdf title. I've tested different python packages to get pdf title :

• python2-pdfrw package, it cannot decrypt encrypted titles... see issue
• python2-pdfminer.six, it was very slow to get PDF title for large files, slowing file explorer

I got a simple tool pdfinfo, it is very fast to get PDF title and shows encrypted title, so I used it from python to get PDF name as you can see in pdf-name-column.py

pdf-name-column.py

# Nautilus python extension that shows pdf name in a nautilus
# detailed view column - Python 3 script 

import os
from urllib.parse import unquote
import subprocess
from os.path  import splitext
from gi.repository import GObject,Nautilus

TITLE_PDFINFO = 'Title:'

class ExifColumnExtension(GObject.GObject, Nautilus.ColumnProvider, Nautilus.InfoProvider):
    def __init__(self):
        pass
    
    def get_columns(self):
        return Nautilus.Column(name="NautilusPython::pdf-name-column",
                               attribute="pdf_name",
                               label="PDF name",
                               description="Get pdf name"),

    def update_file_info(self, file):
        if file.get_uri_scheme() != 'file':
            return
        
        filename = unquote(file.get_uri()[7:])
        if splitext(filename)[1].lower() == '.pdf' :
            res = subprocess.check_output(['pdfinfo', filename]).decode('utf-8')
            for line in res.splitlines():
                if TITLE_PDFINFO in line :
                    file.add_string_attribute('pdf_name', line.split(TITLE_PDFINFO)[1].strip())
                    return
        file.add_string_attribute('pdf_name', '')

Hi there, this is probably a long shot but do you know if there would be any way to modify this script to display PDF keyword metadata in the file explorer.

Very convenient. Here are my notes.
In ubuntu, to install the necessary bindings, simply do sudo apt install python-nautilus. This package seems to have been configured for python2, whereas the script above is written for python3. No worries, simply change the line

from urllib.parse import unquote

to

from urlparse import unquote

After placing the script in the correct location, you need to restart nautilus. Running pkill nautilus and starting it again should do the job. Starting nautilus from the command line will show you any errors from the script, and helps to debug.