Clean media by deleting those files which are no more referenced by any FileField (django command)

clean_media.py

# -*- coding: utf-8 -*-

import os
import sys

from six.moves import input

from django.conf import settings
from django.core.management.base import BaseCommand, CommandError
from django.apps import apps
from django.db import models

class Command(BaseCommand):
    """"""
    help = 'Clean media by deleting those files which are no more ' \
           'referenced by any FileField'

    def add_arguments(self, parser):
        """"""
        parser.add_argument('--noinput',
            action = 'store_true',
            default = False,
            help = 'Clean media by deleting those files which are no more '
                   'referenced by any FileField without asking for confirmation'
        )

    def handle(self, *args, **options):
        """"""
        #print(options)

        # Exclude cache folder by default since generally it's used by
        # third-party apps
        exclude_paths = getattr(settings, 'CLEANUP_EXCLUDE_PATHS', ['cache'])
        exclude_roots = [
            os.path.abspath(os.path.normpath(os.path.join(settings.MEDIA_ROOT,
                                                          path)))
            for path in exclude_paths
        ]

        # Build a list of all media files used by our model
        models_files = set([])

        for model in apps.get_models():

            model_objects = list(model.objects.all())
            model_file_fields = []
            model_files = []

            for field_name in model._meta.get_all_field_names():
                field = model._meta.get_field(field_name)

                if isinstance(field, models.FileField):
                    model_file_fields.append(str(field_name))

            if not len(model_file_fields):
                continue

            #print(u'')
            #print(u'%s %s' % (model.__name__, model_file_fields,))

            for obj in model_objects:
                for field in model_file_fields:
                    file = getattr(obj, field, '')

                    if file:
                        #print(file.path)
                        models_files.add(os.path.abspath(file.path))

        unref_files = set([])

        for root, dirs, files in os.walk(os.path.abspath(settings.MEDIA_ROOT)):
            exclude_root = False

            for excluded_root in exclude_roots:
                if root.find(excluded_root) == 0:
                    exclude_root = True
                    break

            if exclude_root:
                continue

            for file in files:
                file_path = os.path.abspath(os.path.join(root, file))

                if not file_path in models_files:
                    unref_files.add(file_path)

        for file in unref_files:
            print('Unreferenced file found: %s' % (file))

        num_unref_files = len(unref_files)

        print('%s unreferenced file%s found'
              '' % (num_unref_files, 's' if num_unref_files else '',))

        if num_unref_files:

            remove_unref_files = True \
                if options['noinput'] \
                else (input('Remove all unreferenced files? (y/N) '
                            '').lower().find('y') == 0)

            if remove_unref_files:
                for file in unref_files:
                    try:
                        os.remove(file)
                        print('Removed unreferenced file: %s' % (file))

                    except Exception as err:
                        print('Failed to remove unreferenced file: '
                              '%s due to %s' % (file, str(err)))
                        continue

        return

Summary of changes (from original):

• Fixes for Python 3.
• Improved performance (replaced list with set) for which the in works much faster.
• Use absolute paths everywhere for cases when media directory is located outside of the Django project.
• Allow listing of the paths to exclude in CLEANUP_EXCLUDE_PATHS of django settings. Fall back to ['cache'] by default.

Thanks for the fixed, by the way this script doesn't work anymore with django >= 1.10:
AttributeError: 'Options' object has no attribute 'get_all_field_names'