A tool for removing malicious content from input data before saving data into database. It takes input containing HTML with XSS scripts and returns valid HTML in the output. It is a wrapper around Python’s bleach library to easily integrate it with Django framework and it implements whitelist based approach to remove harmful content.

Setup

  1. Install input-sanitizer via pip:

    pip install input-sanitizer

  2. Add input-sanitizer to your INSTALLED_APPS:

    INSTALLED_APPS = [
    # ...
    'input_sanitizer',
    # ...
]

  3. Add default configurations for allowed tags, etc in settings.py. These configurations are optional and will defaults to using the bleach defaults. Refer to bleach documentation for their use:

    # tags which are allowed
BLEACH_ALLOWED_TAGS = ["div", "section", "a", "i"]

# remove all tags from input
BLEACH_STRIP_TAGS = True

# remove comments, or leave them in
BLEACH_STRIP_COMMENTS = True

Usage

In Django Models

input-sanitizer provides two custom model fields SanitizedCharField and SanitizedTextField to automatically remove malicious content from input before saving data into database, but keep in mind that it won’t work with bulk update, bulk create, etc as these operations are done at the database level. You can still manually sanitize input data to use for bulk update, bulk create, etc operations.

# in models.py
from django import models
from input_sanitizer import sanitized_models

class User(models.Model):
    username = sanitized_models.SanitizedCharField()
    info = sanitized_models.SanitizedTextField()

SanitizedCharField and SanitizedTextField may take following arguments to alter cleaning behaviour. Please, refer to bleach documentation for their use:

  • allowed_tags: Tags which are allowed
  • strip_comments: Remove comments from data
  • strip_tags: Remove all tags from data

SanitizedCharField is a extension of Django model’s CharField and therefore, it will accept all normal CharField arguments.

SanitizedTextField is a extension of Django model’s TextField and therefore, it will accept all normal TextField arguments.

In Django Forms

SanitizedCharField and SanitizedTextField fields can be used to clean XSS content from form fields while validating and saving the form data.

# in forms.py
from django import forms
from input_sanitizer import sanitized_forms

class User(forms.ModelForm):
    username = sanitized_forms.SanitizedCharField()
    info = sanitized_forms.SanitizedTextField()

SanitizedCharField and SanitizedTextField may take following arguments to alter cleaning behaviour. Please, refer to bleach documentation for their use:

  • allowed_tags: Tags which are allowed
  • strip_comments: Remove comments from data
  • strip_tags: Remove all tags from data

SanitizedCharField and SanitizedTextField fields will return validation errors if these fields are required. You can provide following arguments to customize error messages. f_name takes precedence over f_name while returning error message.

  • FIELD_ERROR: Error message
  • FIELD_NAME: Field name

SanitizedCharField is a extension of Django form’s CharField. It will accept all normal CharField arguments.

SanitizedTextField is a extension of Django form’s TextField. It will accept all normal TextField arguments.

In DRF Serializers

SanitizedCharField and SanitizedTextField fields can be used to clean XSS content from serializer fields while validating and saving the serializer data.

# in serializers.py
from rest_framework import serializers
from input_sanitizer import sanitized_serializers

class User(serializers.ModelSerializer):
    username = sanitized_serializers.SanitizedCharField()
    info = sanitized_serializers.SanitizedTextField()

SanitizedCharField and SanitizedTextField may take following arguments to alter cleaning behaviour. Please, refer to bleach documentation for their use:

  • allowed_tags: Tags which are allowed
  • strip_comments: Remove comments from data
  • strip_tags: Remove all tags from data

SanitizedCharField and SanitizedTextField fields will return validation errors if these fields are required. You can provide following arguments to customize error messages. f_name takes precedence over f_name while returning error message.

  • FIELD_ERROR: Error message
  • FIELD_NAME: Field name

SanitizedCharField is a extension of DRF serializer’s CharField. It will accept all normal CharField arguments.

SanitizedTextField is a extension of DRF serializer’s TextField. It will accept all normal TextField arguments.

In Views

To manually sanitize data, you can use sanitize_data function. It can be used to sanitize data to be used for bulk update, bulk create, etc.

from input_sanitizer import sanitizers
cleaned_data = sanitizers.sanitize_data(data, bleach_kwargs={})

bleach_kwargs arguments are optional and will default to using the bleach defaults. You may pass following arguments to alter cleaned output as per your requirement.

  • allowed_tags: Tags which are allowed
  • strip_comments: Remove comments from data
  • strip_tags: Remove all tags from data

GitHub

View Github