faker-file

Create files with fake data. In many formats. With no efforts.

Prerequisites

All of core dependencies of this package are MIT licensed. Most of optional dependencies of this package are MIT licensed, while a few are BSD-, Apache 2-, GPL or HPND licensed.

All licenses are mentioned below between the brackets.

• Core package requires Python 3.9, 3.10, 3.11 or 3.12.
• Faker (MIT) is the only required dependency.
• Django (BSD) integration with factory_boy (MIT) has been tested with Django starting from version 2.2 to 4.2 (although only maintained versions of Django are currently being tested against).
• BMP, GIF and TIFF file support requires either just Pillow (HPND), or a combination of WeasyPrint (BSD), pdf2image (MIT), Pillow (HPND) and poppler (GPLv2).
• DOCX file support requires python-docx (MIT).
• EPUB file support requires xml2epub (MIT) and Jinja2 (BSD).
• ICO, JPEG, PNG, SVG and WEBP files support requires either just Pillow (HPND), or a combination of imgkit (MIT) and wkhtmltopdf (LGPLv3).
• MP3 file support requires gTTS (MIT) or edge-tts (GPLv3).
• PDF file support requires either Pillow (HPND), or a combination of pdfkit (MIT) and wkhtmltopdf (LGPLv3), or reportlab (BSD).
• PPTX file support requires python-pptx (MIT).
• ODP and ODT file support requires odfpy (Apache 2).
• ODS file support requires tablib (MIT) and odfpy (Apache 2).
• XLSX file support requires tablib (MIT) and openpyxl (MIT).
• PathyFileSystemStorage storage support requires pathy (Apache 2).
• AWSS3Storage storage support requires pathy (Apache 2) and boto3 (Apache 2).
• AzureCloudStorage storage support requires pathy (Apache 2) and azure-storage-blob (MIT).
• GoogleCloudStorage storage support requires pathy (Apache 2) and google-cloud-storage (Apache 2).
• SFTPStorage storage support requires paramiko (LGLPv2.1).
• AugmentFileFromDirProvider provider requires either a combination of textaugment (MIT) and nltk (Apache 2) or a combination of nlpaug (MIT), PyTorch (BSD), transformers (Apache 2), numpy (BSD), pandas (BSD), tika (Apache 2) and Apache Tika (Apache 2).

Documentation

• Documentation is available on Read the Docs.
• For bootstrapping check the Quick start.
• For various ready to use code examples see the Recipes.
• For tips on PDF creation see Creating PDF.
• For tips on DOCX creation see Creating DOCX.
• For tips on ODT creation see Creating ODT.
• For tips on images creation see Creating images.
• For CLI options see the CLI.
• Read the Methodology.
• For guidelines on contributing check the Contributor guidelines.

Online demos and related projects

Check the demo(s) and related projects below:

• REST API demo (based on faker-file-api REST API)
• UI frontend demo (based on faker-file-ui UI frontend)
• WASM frontend demo (based on faker-file-wasm WASM frontend)
• faker-file-qt GUI application (based on PyQT5).

Installation

Latest stable version from PyPI

WIth all dependencies

pip install faker-file'[all]'

Only core

pip install faker-file

With most common dependencies

Everything, except ML libraries which are required for data augmentation only

pip install faker-file'[common]'

With DOCX support

pip install faker-file'[docx]'

With EPUB support

pip install faker-file'[epub]'

With images support

pip install faker-file'[images]'

With PDF support

pip install faker-file'[pdf]'

With MP3 support

pip install faker-file'[mp3]'

With XLSX support

pip install faker-file'[xlsx]'

With ODS support

pip install faker-file'[ods]'

With ODT support

pip install faker-file'[odt]'

With data augmentation support

pip install faker-file'[data-augmentation]'

With GoogleCloudStorage support

pip install faker-file'[gcs]'

With AzureCloudStorage support

pip install faker-file'[azure]'

With AWSS3Storage support

pip install faker-file'[s3]'

Or development version from GitHub

pip install https://github.com/barseghyanartur/faker-file/archive/main.tar.gz

Features

Supported file types

• BIN
• BMP
• CSV
• DOCX
• EML
• EPUB
• ICO
• GIF
• JPEG
• JSON
• MP3
• ODS
• ODT
• ODP
• PDF
• PNG
• RTF
• PPTX
• SVG
• TAR
• TIFF
• TXT
• WEBP
• XLSX
• XML
• ZIP

For all image formats (BMP, ICO, GIF, JPEG, PNG, SVG, TIFF and WEBP) and PDF, there are both graphic-only and mixed-content file providers (that also have text-to-image capabilities).

Additional providers

• AugmentFileFromDirProvider: Make an augmented copy of randomly picked file from given directory. The following types are supported : DOCX, EML, EPUB, ODT, PDF, RTF and TXT.
• AugmentRandomImageFromDirProvider: Augment a random image file from given directory. The following types are supported : BMP, GIF, JPEG, PNG, TIFF and WEBP.
• AugmentImageFromPathProvider: Augment an image file from given path. Supported file types are the same as for AugmentRandomImageFromDirProvider provider.
• GenericFileProvider: Create files in any format from raw bytes or a predefined template.
• RandomFileFromDirProvider: Pick a random file from given directory.
• FileFromPathProvider: File from given path.

Supported file storages

• Native file system storage
• AWS S3 storage
• Azure Cloud Storage
• Google Cloud Storage
• SFTP storage

Usage examples

With Faker

Recommended way

from faker import Faker
# Import the file provider we want to use
from faker_file.providers.txt_file import TxtFileProvider

FAKER = Faker()  # Initialise Faker instance
FAKER.add_provider(TxtFileProvider)  # Register the TXT file provider

file = FAKER.txt_file()  # Generate a TXT file

# Meta-data is stored inside a ``data`` attribute (``dict``).
# The following line would produce something like /tmp/tmp/tmphzzb8mot.txt
print(file.data["filename"])
# The following line would produce a text generated by Faker, used as
# the content of the generated file.
print(file.data["content"])

Note

Note, that in this case file value is a StringValue instance, which inherits from str but contains meta-data such as absolute path to the generated file, and text used to generate the file, stored in filename and content keys of the data attribute respectively. See Meta-data for more information.

If you just need bytes back (instead of creating the file), provide the raw=True argument (works with all provider classes and inner functions):

from faker import Faker
from faker_file.providers.txt_file import TxtFileProvider

FAKER = Faker()
FAKER.add_provider(TxtFileProvider)

raw = FAKER.txt_file(raw=True)

Note

Note, that in this case file value is a BytesValue instance, which inherits from bytes but contains meta-data such as absolute path to the generated file, and text used to generate the file, stored in filename and content keys of the data attribute respectively. See Meta-data for more information.

But this works too

from faker import Faker
from faker_file.providers.txt_file import TxtFileProvider

FAKER = Faker()

file = TxtFileProvider(FAKER).txt_file()

If you just need bytes back:

from faker import Faker
from faker_file.providers.txt_file import TxtFileProvider

FAKER = Faker()

raw = TxtFileProvider(FAKER).txt_file(raw=True)

With factory_boy

Filename: upload/models.py

from django.db import models

class Upload(models.Model):

    # ...
    file = models.FileField()

    class Meta:
        app_label = "upload"

Filename: upload/factories.py

Note, that when using faker-file with Django and native file system storages, you need to pass your MEDIA_ROOT setting as root_path value to the chosen file storage as show below.

import factory
from django.conf import settings
from factory import Faker
from factory.django import DjangoModelFactory
from faker_file.providers.docx_file import DocxFileProvider
from faker_file.storages.filesystem import FileSystemStorage

from upload.models import Upload

FS_STORAGE = FileSystemStorage(
    root_path=settings.MEDIA_ROOT,
    rel_path="tmp"
)
factory.Faker.add_provider(DocxFileProvider)

class UploadFactory(DjangoModelFactory):

    # ...
    file = Faker("docx_file", storage=FS_STORAGE)

    class Meta:
        model = Upload

Meta-data

The return value of any file provider file generator function is either StringValue or BytesValue, which inherit from str and bytes respectively.

Both StringValue and BytesValue instances have a meta data attribute named data (type dict). Various file providers use data to store meta-data, such as filename (absolute path to the generated file; valid for all file providers), or content (text used when generating the file; valid for most file providers, except FileFromPathProvider, RandomFileFromDirProvider, TarFileProvider and ZipFileProvider).

All file providers store an absolute path to the generated file in filename key of the data attribute and instance of the storage used in storage key. See the table below.

Key name	File provider
filename	all
storage	all
content	all except FileFromPathProvider, RandomFileFromDirProvider, TarFileProvider, ZipFileProvider and all graphic file providers such as GraphicBmpFileProvider, GraphicGifFileProvider, GraphicIcoFileProvider, GraphicJpegFileProvider, GraphicPdfFileProvider, GraphicPngFileProvider, GraphicTiffFileProvider and GraphicWebpFileProvider
inner	only EmlFileProvider, TarFileProvider and ZipFileProvider

File storages

All file operations are delegated to a separate abstraction layer of storages.

The following storages are implemented:

• FileSystemStorage: Does not have additional requirements.
• PathyFileSystemStorage: Requires pathy.
• AzureCloudStorage: Requires pathy and Azure related dependencies.
• GoogleCloudStorage: Requires pathy and Google Cloud related dependencies.
• AWSS3Storage: Requires pathy and AWS S3 related dependencies.
• SFTPStorage: Requires paramiko and related dependencies.

Usage example with storages

FileSystemStorage example

Native file system storage. Does not have dependencies.

• root_path: Path to the root directory. Given the example of Django, this would be the path to the MEDIA_ROOT directory. It's important to know, that root_path will not be embedded into the string representation of the file. Only rel_path will.
• rel_path: Relative path from the root directory. Given the example of Django, this would be the rest of the path to the file.

import tempfile
from faker import Faker
from faker_file.providers.txt_file import TxtFileProvider
from faker_file.storages.filesystem import FileSystemStorage

FS_STORAGE = FileSystemStorage(
    root_path=tempfile.gettempdir(),  # Use settings.MEDIA_ROOT for Django
    rel_path="tmp",
)

FAKER = Faker()

file = TxtFileProvider(FAKER).txt_file(storage=FS_STORAGE)

FS_STORAGE.exists(file)

PathyFileSystemStorage example

Native file system storage. Requires pathy.

import tempfile
from pathy import use_fs
from faker import Faker
from faker_file.providers.txt_file import TxtFileProvider
from faker_file.storages.cloud import PathyFileSystemStorage

use_fs(tempfile.gettempdir())
PATHY_FS_STORAGE = PathyFileSystemStorage(
    bucket_name="bucket_name",
    root_path="tmp",
    rel_path="sub-tmp",
)

FAKER = Faker()

file = TxtFileProvider(FAKER).txt_file(storage=PATHY_FS_STORAGE)

PATHY_FS_STORAGE.exists(file)

AWSS3Storage example

AWS S3 storage. Requires pathy and boto3.

from faker import Faker
from faker_file.providers.txt_file import TxtFileProvider
from faker_file.storages.aws_s3 import AWSS3Storage

S3_STORAGE = AWSS3Storage(
    bucket_name="bucket_name",
    root_path="tmp",  # Optional
    rel_path="sub-tmp",  # Optional
    # Credentials are optional too. If your AWS credentials are properly
    # set in the ~/.aws/credentials, you don't need to send them
    # explicitly.
    credentials={
        "key_id": "YOUR KEY ID",
        "key_secret": "YOUR KEY SECRET"
    },
)

FAKER = Faker()

file = TxtFileProvider(FAKER).txt_file(storage=S3_STORAGE)

assert S3_STORAGE.exists(file)

Testing

Simply type:

pytest -vrx

Or use tox:

tox

Or use tox to check specific env:

tox -e py310-django41

Writing documentation

Keep the following hierarchy.

=====
title
=====

header
======

sub-header
----------

sub-sub-header
~~~~~~~~~~~~~~

sub-sub-sub-header
^^^^^^^^^^^^^^^^^^

sub-sub-sub-sub-header
++++++++++++++++++++++

sub-sub-sub-sub-sub-header
**************************

License

MIT

Support

For security issues contact me at the e-mail given in the Author section.

For overall issues, go to GitHub.

Author

Artur Barseghyan <artur.barseghyan@gmail.com>

Citation

Please, use the following entry when citing faker-file in your research:

@software{faker-file,
  author = {Artur Barseghyan},
  title = {faker-file: Create files with fake data. In many formats. With no efforts.},
  year = {2022-2025},
  publisher = {GitHub},
  journal = {GitHub repository},
  howpublished = {https://github.com/barseghyanartur/faker-file},
}