Cookiecutter: Better Project Templates¶
Cookiecutter creates projects from cookiecutters (project templates), e.g. Python package projects from Python package temnplates.
Basics¶
Cookiecutter¶
A command-line utility that creates projects from cookiecutters (project templates), e.g. creating a Python package project from a Python package project template.
Documentation: https://cookiecutter.readthedocs.io
Free and open source software: BSD license
Features¶
Cross-platform: Windows, Mac, and Linux are officially supported.
You don’t have to know/write Python code to use Cookiecutter.
Works with Python 3.7, 3.8, 3.9., 3.10
Project templates can be in any programming language or markup format: Python, JavaScript, Ruby, CoffeeScript, RST, Markdown, CSS, HTML, you name it. You can use multiple languages in the same project template.
For users of existing templates¶
Simple command line usage:
# Create project from the cookiecutter-pypackage.git repo template # You'll be prompted to enter values. # Then it'll create your Python package in the current working directory, # based on those values. $ cookiecutter https://github.com/audreyfeldroy/cookiecutter-pypackage # For the sake of brevity, repos on GitHub can just use the 'gh' prefix $ cookiecutter gh:audreyfeldroy/cookiecutter-pypackage
Use it at the command line with a local template:
# Create project in the current working directory, from the local # cookiecutter-pypackage/ template $ cookiecutter cookiecutter-pypackage/
Or use it from Python:
from cookiecutter.main import cookiecutter # Create project from the cookiecutter-pypackage/ template cookiecutter('cookiecutter-pypackage/') # Create project from the cookiecutter-pypackage.git repo template cookiecutter('https://github.com/audreyfeldroy/cookiecutter-pypackage.git')
Unless you suppress it with
--no-input, you are prompted for input:Prompts are the keys in
cookiecutter.json.Default responses are the values in
cookiecutter.json.Prompts are shown in order.
Cross-platform support for
~/.cookiecutterrcfiles:default_context: full_name: "Audrey Roy Greenfeld" email: "audreyr@gmail.com" github_username: "audreyfeldroy" cookiecutters_dir: "~/.cookiecutters/"
Cookiecutters (cloned Cookiecutter project templates) are put into
~/.cookiecutters/by default, or cookiecutters_dir if specified.If you have already cloned a cookiecutter into
~/.cookiecutters/, you can reference it by directory name:# Clone cookiecutter-pypackage $ cookiecutter gh:audreyfeldroy/cookiecutter-pypackage # Now you can use the already cloned cookiecutter by name $ cookiecutter cookiecutter-pypackage
You can use local cookiecutters, or remote cookiecutters directly from Git repos or from Mercurial repos on Bitbucket.
Default context: specify key/value pairs that you want used as defaults whenever you generate a project.
Inject extra context with command-line arguments:
cookiecutter --no-input gh:msabramo/cookiecutter-supervisor program_name=foobar startsecs=10
Direct access to the Cookiecutter API allows for injection of extra context.
Paths to local projects can be specified as absolute or relative.
Projects generated to your current directory or to target directory if specified with
-ooption.
For template creators¶
Supports unlimited levels of directory nesting.
100% of templating is done with Jinja2.
Both, directory names and filenames can be templated. For example:
{{cookiecutter.repo_name}}/{{cookiecutter.repo_name}}/{{cookiecutter.repo_name}}.py
Simply define your template variables in a
cookiecutter.jsonfile. For example:{ "full_name": "Audrey Roy Greenfeld", "email": "audreyr@gmail.com", "project_name": "Complexity", "repo_name": "complexity", "project_short_description": "Refreshingly simple static site generator.", "release_date": "2013-07-10", "year": "2013", "version": "0.1.1" }
Pre- and post-generate hooks: Python or shell scripts to run before or after generating a project.
Community¶
The core committer team can be found in authors section. We are always welcome and invite you to participate.
Stuck? Try one of the following:
See the Troubleshooting page.
Ask for help on Stack Overflow.
You are strongly encouraged to file an issue about the problem. Do it even if it’s just “I can’t get it to work on this cookiecutter” with a link to your cookiecutter. Don’t worry about naming/pinpointing the issue properly.
Ask for help on Discord if you must (but please try one of the other options first, so that others can benefit from the discussion).
Development on Cookiecutter is community-driven:
Huge thanks to all the contributors who have pitched in to help make Cookiecutter an even better tool.
Everyone is invited to contribute. Read the contributing instructions, then get started.
Connect with other Cookiecutter contributors and users on Discord (note: due to work and other commitments, a core committer might not always be available)
Encouragement is unbelievably motivating. If you want more work done on Cookiecutter, show support:
Thank a core committer for their efforts.
Star Cookiecutter on GitHub.
Got criticism or complaints?
File an issue so that Cookiecutter can be improved. Be friendly and constructive about what could be better. Make detailed suggestions.
Keep us in the loop so that we can help. For example, if you are discussing problems with Cookiecutter on a mailing list, file an issue where you link to the discussion thread and/or cc at least 1 core committer on the email.
Be encouraging. A comment like “This function ought to be rewritten like this” is much more likely to result in action than a comment like “Eww, look how bad this function is.”
Waiting for a response to an issue/question?
Be patient and persistent. All issues are on the core committer team’s radar and will be considered thoughtfully, but we have a lot of issues to work through. If urgent, it’s fine to ping a core committer in the issue with a reminder.
Ask others to comment, discuss, review, etc.
Search the Cookiecutter repo for issues related to yours.
Need a fix/feature/release/help urgently, and can’t wait? @audreyfeldroy is available for hire for consultation or custom development.
Support This Project¶
This project is run by volunteers. Shortly we will be providing means for organizations and individuals to support the project.
Code of Conduct¶
Everyone interacting in the Cookiecutter project’s codebases and documentation is expected to follow the PyPA Code of Conduct. This includes, but is not limited to, issue trackers, chat rooms, mailing lists, and other virtual or in real life communication.
Creator / Leader¶
This project was created and is led by Audrey Roy Greenfeld.
She is supported by a team of maintainers.
Overview¶
Cookiecutter takes a template provided as a directory structure with template-files. Templates can be in located in the filesystem, as a ZIP-file or on a VCS-Server (Git/Hg) like GitHub.
It reads a settings file and prompts the user interactivly wether to change the settings.
Then it takes both and generates an output directory structure from it.
Additional the template can provide code (Python or shell-script) to be executed before and after generation (pre-gen- and post-gen-hooks).
Input¶
This is a directory structure for a simple cookiecutter:
cookiecutter-something/
├── {{ cookiecutter.project_name }}/ <--------- Project template
│ └── ...
├── blah.txt <--------- Non-templated files/dirs
│ go outside
│
└── cookiecutter.json <--------- Prompts & default values
You must have:
A
cookiecutter.jsonfile.A
{{ cookiecutter.project_name }}/directory, whereproject_nameis defined in yourcookiecutter.json.
Beyond that, you can have whatever files/directories you want.
See https://github.com/audreyr/cookiecutter-pypackage for a real-world example of this.
Output¶
This is what will be generated locally, in your current directory:
mysomething/ <---------- Value corresponding to what you enter at the
│ project_name prompt
│
└── ... <-------- Files corresponding to those in your
cookiecutter's `{{ cookiecutter.project_name }}/` dir
Installation¶
Prerequisites¶
Python interpreter
Adjust your path
Packaging tools
Python interpreter¶
Install Python for your operating system. On Windows and macOS this is usually necessary. Most Linux distributions are coming with Python pre-installed. Consult the official Python documentation for details.
You can install the Python binaries from python.org. Alternatively on macOS, you can use the homebrew package manager.
brew install python3
Adjust your path¶
Ensure that your bin folder is on your path for your platform. Typically ~/.local/ for UNIX and macOS, or %APPDATA%\Python on Windows. (See the Python documentation for site.USER_BASE for full details.)
UNIX and macOS¶
For bash shells, add the following to your .bash_profile (adjust for other shells):
# Add ~/.local/ to PATH
export PATH=$HOME/.local/bin:$PATH
Remember to load changes with source ~/.bash_profile or open a new shell session.
Windows¶
Ensure the directory where cookiecutter will be installed is in your environment’s Path in order to make it possible to invoke it from a command prompt. To do so, search for “Environment Variables” on your computer (on Windows 10, it is under System Properties –> Advanced) and add that directory to the Path environment variable, using the GUI to edit path segments.
Example segments should look like %APPDATA%\Python\Python3x\Scripts, where you have your version of Python instead of Python3x.
You may need to restart your command prompt session to load the environment variables.
See also
See Configuring Python (on Windows) for full details.
Unix on Windows
You may also install Windows Subsystem for Linux or GNU utilities for Win32 to use Unix commands on Windows.
Packaging tools¶
See the Python Packaging Authority’s (PyPA) documentation Requirements for Installing Packages for full details.
Alternate installations¶
Homebrew (Mac OS X only):
brew install cookiecutter
Pipx (Linux, OSX and Windows):
pipx install cookiecutter
Upgrading¶
from 0.6.4 to 0.7.0 or greater¶
First, read History in detail. There are a lot of major changes. The big ones are:
Cookiecutter no longer deletes the cloned repo after generating a project.
Cloned repos are saved into ~/.cookiecutters/.
You can optionally create a ~/.cookiecutterrc config file.
Or with pip:
python3 -m pip install --upgrade cookiecutter
Upgrade Cookiecutter either with easy_install (deprecated):
easy_install --upgrade cookiecutter
Then you should be good to go.
Usage¶
Make your changes¶
Modify the variables defined in cookiecutter.json.
Open up the skeleton project. If you need to change it around a bit, do so.
You probably also want to create a repo, name it differently, and push it as your own new Cookiecutter project template, for handy future use.
Generate your project¶
Then generate your project from the project template:
$ cookiecutter cookiecutter-pypackage/
The only argument is the input directory. (The output directory is generated by rendering that, and it can’t be the same as the input directory.)
Note
see Command Line Options for extra command line arguments
Try it out!
Works directly with git and hg (mercurial) repos too¶
To create a project from the cookiecutter-pypackage.git repo template:
$ cookiecutter gh:audreyr/cookiecutter-pypackage
Cookiecutter knows abbreviations for Github (gh), Bitbucket (bb), and
GitLab (gl) projects, but you can also give it the full URL to any
repository:
$ cookiecutter https://github.com/audreyr/cookiecutter-pypackage.git
$ cookiecutter git+ssh://git@github.com/audreyr/cookiecutter-pypackage.git
$ cookiecutter hg+ssh://hg@bitbucket.org/audreyr/cookiecutter-pypackage
You will be prompted to enter a bunch of project config values. (These are defined in the project’s cookiecutter.json.)
Then, Cookiecutter will generate a project from the template, using the values that you entered. It will be placed in your current directory.
And if you want to specify a branch you can do that with:
$ cookiecutter https://github.com/audreyr/cookiecutter-pypackage.git --checkout develop
Works with private repos¶
If you want to work with repos that are not hosted in github or bitbucket you can indicate explicitly the type of repo that you want to use prepending hg+ or git+ to repo url:
$ cookiecutter hg+https://example.com/repo
In addition, one can provide a path to the cookiecutter stored on a local server:
$ cookiecutter file://server/folder/project.git
Works with Zip files¶
You can also distribute cookiecutter templates as Zip files. To use a Zip file template, point cookiecutter at a Zip file on your local machine:
$ cookiecutter /path/to/template.zip
Or, if the Zip file is online:
$ cookiecutter https://example.com/path/to/template.zip
If the template has already been downloaded, or a template with the same name has already been downloaded, you will be prompted to delete the existing template before proceeding.
The Zip file contents should be the same as a git/hg repository for a template - that is, the zipfile should unpack into a top level directory that contains the name of the template. The name of the zipfile doesn’t have to match the name of the template - for example, you can label a zipfile with a version number, but omit the version number from the directory inside the Zip file.
If you want to see an example Zipfile, find any Cookiecutter repository on Github and download that repository as a zip file - Github repository downloads are in a valid format for Cookiecutter.
Password-protected Zip files¶
If your repository Zip file is password protected, Cookiecutter will prompt you for that password whenever the template is used.
Alternatively, if you want to use a password-protected Zip file in an automated environment, you can export the COOKIECUTTER_REPO_PASSWORD environment variable; the value of that environment variable will be used whenever a password is required.
Command Line Options¶
Tutorials¶
Tutorials by @audreyr
External Links¶
Learn the Basics of Cookiecutter by Creating a Cookiecutter - first steps tutorial with example template by @BruceEckel
Cookiedozer Tutorials by @hackebrot
Advanced Usage¶
Various advanced topics regarding cookiecutter usage.
Using Pre/Post-Generate Hooks¶
New in cookiecutter 0.7
You can have Python or Shell scripts that run before and/or after your project is generated.
Put them in hooks/ like this:
cookiecutter-something/
├── {{cookiecutter.project_slug}}/
├── hooks
│ ├── pre_gen_project.py
│ └── post_gen_project.py
└── cookiecutter.json
Shell scripts work similarly:
cookiecutter-something/
├── {{cookiecutter.project_slug}}/
├── hooks
│ ├── pre_gen_project.sh
│ └── post_gen_project.sh
└── cookiecutter.json
It shouldn’t be too hard to extend Cookiecutter to work with other types of scripts too. Pull requests are welcome.
For portability, you should use Python scripts (with extension .py) for your hooks, as these can be run on any platform. However, if you intend for your template to only be run on a single platform, a shell script (or .bat file on Windows) can be a quicker alternative.
Writing hooks¶
Here are some details on how to write pre/post-generate hook scripts.
Exit with an appropriate status¶
Make sure your hook scripts work in a robust manner. If a hook script fails (that is, if it finishes with a nonzero exit status), the project generation will stop and the generated directory will be cleaned up.
Current working directory¶
When the hook scripts script are run, their current working directory is the root of the generated project. This makes it easy for a post-generate hook to find generated files using relative paths.
Template variables are rendered in the script¶
Just like your project template, Cookiecutter also renders Jinja template syntax in your scripts.
This lets you incorporate Jinja template variables in your scripts.
For example, this line of Python sets module_name to the value of the cookiecutter.module_name template variable:
module_name = '{{ cookiecutter.module_name }}'
Example: Validating template variables¶
Here is an example of a pre-generate hook script, defined at hooks/pre_gen_project.py, that validates a template variable before generating the project:
import re
import sys
MODULE_REGEX = r'^[_a-zA-Z][_a-zA-Z0-9]+$'
module_name = '{{ cookiecutter.module_name }}'
if not re.match(MODULE_REGEX, module_name):
print('ERROR: %s is not a valid Python module name!' % module_name)
# exits with status 1 to indicate failure
sys.exit(1)
Example: Conditional files / directories¶
Here is an example of a post-generate hook script.
The file hooks/post_gen_project.py shows how to achieve conditional control of files and directories after generating the project.
The script ensures that the directory structure is as expected by removing unwanted files and directories:
import os
import sys
REMOVE_PATHS = [
'{% if cookiecutter.packaging != "pip" %} requirements.txt {% endif %}',
'{% if cookiecutter.packaging != "poetry" %} poetry.lock {% endif %}',
]
for path in REMOVE_PATHS:
path = path.strip()
if path and os.path.exists(path):
if os.path.isdir(path):
os.rmdir(path)
else:
os.unlink(path)
User Config¶
New in Cookiecutter 0.7
If you use Cookiecutter a lot, you’ll find it useful to have a user config file. By default Cookiecutter tries to retrieve settings from a .cookiecutterrc file in your home directory.
New in Cookiecutter 1.3
You can also specify a config file on the command line via --config-file.
cookiecutter --config-file /home/audreyr/my-custom-config.yaml cookiecutter-pypackage
Or you can set the COOKIECUTTER_CONFIG environment variable:
export COOKIECUTTER_CONFIG=/home/audreyr/my-custom-config.yaml
If you wish to stick to the built-in config and not load any user config file at all, use the CLI option --default-config instead.
Preventing Cookiecutter from loading user settings is crucial for writing integration tests in an isolated environment.
Example user config:
default_context:
full_name: "Audrey Roy"
email: "audreyr@example.com"
github_username: "audreyr"
cookiecutters_dir: "/home/audreyr/my-custom-cookiecutters-dir/"
replay_dir: "/home/audreyr/my-custom-replay-dir/"
abbreviations:
pp: https://github.com/audreyr/cookiecutter-pypackage.git
gh: https://github.com/{0}.git
bb: https://bitbucket.org/{0}
Possible settings are:
default_context:A list of key/value pairs that you want injected as context whenever you generate a project with Cookiecutter. These values are treated like the defaults in
cookiecutter.json, upon generation of any project.cookiecutters_dirDirectory where your cookiecutters are cloned to when you use Cookiecutter with a repo argument.
replay_dirDirectory where Cookiecutter dumps context data to, which you can fetch later on when using the replay feature.
abbreviationsA list of abbreviations for cookiecutters. Abbreviations can be simple aliases for a repo name, or can be used as a prefix, in the form
abbr:suffix. Any suffix will be inserted into the expansion in place of the text{0}, using standard Python string formatting. With the above aliases, you could use thecookiecutter-pypackagetemplate simply by sayingcookiecutter pp, orcookiecutter gh:audreyr/cookiecutter-pypackage. Thegh(GitHub),bb(Bitbucket), andgl(Gitlab) abbreviations shown above are actually built in, and can be used without defining them yourself.
Read also: Injecting Extra Context
Injecting Extra Context¶
You can specify an extra_context dictionary that will override values from cookiecutter.json or .cookiecutterrc:
cookiecutter(
'cookiecutter-pypackage/',
extra_context={'project_name': 'TheGreatest'},
)
This works as command-line parameters as well:
cookiecutter --no-input cookiecutter-pypackage/ project_name=TheGreatest
You will also need to add these keys to the cookiecutter.json or .cookiecutterrc.
Example: Injecting a Timestamp¶
If you have cookiecutter.json that has the following keys:
{
"timestamp": "{{ cookiecutter.timestamp }}"
}
This Python script will dynamically inject a timestamp value as the project is generated:
from cookiecutter.main import cookiecutter
from datetime import datetime
cookiecutter(
'cookiecutter-django',
extra_context={'timestamp': datetime.utcnow().isoformat()}
)
How this works:
The script uses
datetimeto get the current UTC time in ISO format.To generate the project,
cookiecutter()is called, passing the timestamp in as context via theextra_context`dict.
Suppressing Command-Line Prompts¶
To suppress the prompts asking for input, use no_input.
Basic Example: Using the Defaults¶
Cookiecutter will pick a default value if used with no_input:
from cookiecutter.main import cookiecutter
cookiecutter(
'cookiecutter-django',
no_input=True,
)
In this case it will be using the default defined in cookiecutter.json or .cookiecutterrc.
Note
values from cookiecutter.json will be overridden by values from .cookiecutterrc
Advanced Example: Defaults + Extra Context¶
If you combine an extra_context dict with the no_input argument, you can programmatically create the project with a set list of context parameters and without any command line prompts:
cookiecutter('cookiecutter-pypackage/',
no_input=True,
extra_context={'project_name': 'TheGreatest'})
See also Injecting Extra Context and the API Reference for more details.
Templates in Context Values¶
The values (but not the keys!) of cookiecutter.json are also Jinja2 templates. Values from user prompts are added to the context immediately, such that one context value can be derived from previous values. This approach can potentially save your user a lot of keystrokes by providing more sensible defaults.
Basic Example: Templates in Context¶
Python packages show some patterns for their naming conventions:
a human-readable project name
a lowercase, dashed repository name
an importable, dash-less package name
Here is a cookiecutter.json with templated values for this pattern:
{
"project_name": "My New Project",
"project_slug": "{{ cookiecutter.project_name|lower|replace(' ', '-') }}",
"pkg_name": "{{ cookiecutter.project_slug|replace('-', '') }}"
}
If the user takes the defaults, or uses no_input, the templated values will be:
my-new-project
mynewproject
Or, if the user gives Yet Another New Project, the values will be:
yet-another-new-projectyetanothernewproject
Private Variables¶
Cookiecutter allows the definition private variables by prepending an underscore to the variable name.
The user will not be required to fill those variables in.
These can either be not rendered, by using a prepending underscore, or rendered, prepending a double underscore.
For example, the cookiecutter.json:
{
"project_name": "Really cool project",
"_not_rendered": "{{ cookiecutter.project_name|lower }}",
"__rendered": "{{ cookiecutter.project_name|lower }}"
}
Will be rendered as:
{
"project_name": "Really cool project",
"_not_rendered": "{{ cookiecutter.project_name|lower }}",
"__rendered": "really cool project"
}
The user will only be asked for project_name.
Non-rendered private variables can be used for defining constants.
An example of where you may wish to use private rendered variables is creating a Python package repository and want to enforce naming consistency.
To ensure the repository and package name are based on the project name, you could create a cookiecutter.json such as:
{
"project_name": "Project Name",
"__project_slug": "{{ cookiecutter.project_name|lower|replace(' ', '-') }}",
"__package_name": "{{ cookiecutter.project_name|lower|replace(' ', '_') }}",
}
Which could create a structure like this:
project-name
├── Makefile
├── README.md
├── requirements.txt
└── src
├── project_name
│ └── __init__.py
├── setup.py
└── tests
└── __init__.py
The README.md can then have a plain English project title.
Copy without Render¶
New in Cookiecutter 1.1
To avoid rendering directories and files of a cookiecutter, the _copy_without_render key can be used in the cookiecutter.json.
The value of this key accepts a list of Unix shell-style wildcards:
{
"project_slug": "sample",
"_copy_without_render": [
"*.html",
"*not_rendered_dir",
"rendered_dir/not_rendered_file.ini"
]
}
Note: Only the content of the files will be copied without being rendered. The paths are subject to rendering. This allows you to write:
{
"project_slug": "sample",
"_copy_without_render": [
"{{cookiecutter.repo_name}}/templates/*.html",
]
}
In this example, {{cookiecutter.repo_name}} will be rendered as expected but the html file content will be copied without rendering.
Replay Project Generation¶
New in Cookiecutter 1.1
On invocation Cookiecutter dumps a json file to ~/.cookiecutter_replay/ which enables you to replay later on.
In other words, it persists your input for a template and fetches it when you run the same template again.
Example for a replay file (which was created via cookiecutter gh:hackebrot/cookiedozer):
{
"cookiecutter": {
"app_class_name": "FooBarApp",
"app_title": "Foo Bar",
"email": "raphael@example.com",
"full_name": "Raphael Pierzina",
"github_username": "hackebrot",
"kivy_version": "1.8.0",
"project_slug": "foobar",
"short_description": "A sleek slideshow app that supports swipe gestures.",
"version": "0.1.0",
"year": "2015"
}
}
To fetch this context data without being prompted on the command line you can use either of the following methods.
Pass the according option on the CLI:
cookiecutter --replay gh:hackebrot/cookiedozer
Or use the Python API:
.. code-block:: python
from cookiecutter.main import cookiecutter cookiecutter(‘gh:hackebrot/cookiedozer’, replay=True)
This feature comes in handy if, for instance, you want to create a new project from an updated template.
Custom replay file¶
New in Cookiecutter 2.0
To specify a custom filename, you can use the --replay-file option:
cookiecutter --replay-file ./cookiedozer.json gh:hackebrot/cookiedozer
This may be useful to run the same replay file over several machines, in tests or when a user of the template reports a problem.
Choice Variables¶
New in Cookiecutter 1.1
Choice variables provide different choices when creating a project. Depending on a user’s choice the template renders things differently.
Basic Usage¶
Choice variables are regular key / value pairs, but with the value being a list of strings.
For example, if you provide the following choice variable in your cookiecutter.json:
{
"license": ["MIT", "BSD-3", "GNU GPL v3.0", "Apache Software License 2.0"]
}
you’d get the following choices when running Cookiecutter:
Select license:
1 - MIT
2 - BSD-3
3 - GNU GPL v3.0
4 - Apache Software License 2.0
Choose from 1, 2, 3, 4 [1]:
Depending on an user’s choice, a different license is rendered by Cookiecutter.
The above license choice variable creates cookiecutter.license, which can be used like this:
{%- if cookiecutter.license == "MIT" -%}
# Possible license content here
{%- elif cookiecutter.license == "BSD-3" -%}
# More possible license content here
{% endif %}
Cookiecutter is using Jinja2’s if conditional expression to determine the correct license.
The created choice variable is still a regular Cookiecutter variable and can be used like this:
License
-------
Distributed under the terms of the `{{cookiecutter.license}}`_ license,
Overwriting Default Choice Values¶
Choice Variables are overwritable using a User Config file.
For example, a choice variable can be created in cookiecutter.json by using a list as value:
{
"license": ["MIT", "BSD-3", "GNU GPL v3.0", "Apache Software License 2.0"]
}
By default, the first entry in the values list serves as default value in the prompt.
Setting the default license agreement to Apache Software License 2.0 can be done using:
default_context:
license: "Apache Software License 2.0"
in the User Config file.
The resulting prompt changes and looks like:
Select license:
1 - Apache Software License 2.0
2 - MIT
3 - BSD-3
4 - GNU GPL v3.0
Choose from 1, 2, 3, 4 [1]:
Note
As you can see the order of the options changed from 1 - MIT to 1 - Apache Software License 2.0. Cookiecutter takes the first value in the list as the default.
Dictionary Variables¶
New in Cookiecutter 1.5
Dictionary variables provide a way to define deep structured information when rendering a template.
Basic Usage¶
Dictionary variables are, as the name suggests, dictionaries of key-value pairs. The dictionary values can, themselves, be other dictionaries and lists - the data structure can be as deep as you need.
For example, you could provide the following dictionary variable in your cookiecutter.json:
{
"project_slug": "new_project",
"file_types": {
"png": {
"name": "Portable Network Graphic",
"library": "libpng",
"apps": [
"GIMP"
]
},
"bmp": {
"name": "Bitmap",
"library": "libbmp",
"apps": [
"Paint",
"GIMP"
]
}
}
}
The above file_types dictionary variable creates cookiecutter.file_types, which can be used like this:
{% for extension, details in cookiecutter.file_types|dictsort %}
<dl>
<dt>Format name:</dt>
<dd>{{ details.name }}</dd>
<dt>Extension:</dt>
<dd>{{ extension }}</dd>
<dt>Applications:</dt>
<dd>
<ul>
{% for app in details.apps -%}
<li>{{ app }}</li>
{% endfor -%}
</ul>
</dd>
</dl>
{% endfor %}
Cookiecutter is using Jinja2’s for expression to iterate over the items in the dictionary.
Template Extensions¶
New in Cookiecutter 1.4
A template may extend the Cookiecutter environment with custom Jinja2 extensions. It can add extra filters, tests, globals or even extend the parser.
To do so, a template author must specify the required extensions in cookiecutter.json as follows:
{
"project_slug": "Foobar",
"year": "{% now 'utc', '%Y' %}",
"_extensions": ["jinja2_time.TimeExtension"]
}
On invocation Cookiecutter tries to import the extensions and add them to its environment respectively.
In the above example, Cookiecutter provides the additional tag now, after installing the jinja2_time.TimeExtension and enabling it in cookiecutter.json.
Please note that Cookiecutter will not install any dependencies on its own! As a user you need to make sure you have all the extensions installed, before running Cookiecutter on a template that requires custom Jinja2 extensions.
By default Cookiecutter includes the following extensions:
cookiecutter.extensions.JsonifyExtensioncookiecutter.extensions.RandomStringExtensioncookiecutter.extensions.SlugifyExtensioncookiecutter.extensions.UUIDExtensionjinja2_time.TimeExtension
Jsonify extension¶
The cookiecutter.extensions.JsonifyExtension extension provides a jsonify filter in templates that converts a Python object to JSON:
{% {'a': True} | jsonify %}
Would output:
{"a": true}
Random string extension¶
New in Cookiecutter 1.7
The cookiecutter.extensions.RandomStringExtension extension provides a random_ascii_string method in templates that generates a random fixed-length string, optionally with punctuation.
Generate a random n-size character string. Example for n=12:
{{ random_ascii_string(12) }}
Outputs:
bIIUczoNvswh
The second argument controls if punctuation and special characters !"#$%&\'()*+,-./:;<=>?@[\\]^_`{|}~ should be present in the result:
{{ random_ascii_string(12, punctuation=True) }}
Outputs:
fQupUkY}W!)!
Slugify extension¶
The cookiecutter.extensions.SlugifyExtension extension provides a slugify filter in templates that converts string into its dashed (“slugified”) version:
{% "It's a random version" | slugify %}
Would output:
it-s-a-random-version
It is different from a mere replace of spaces since it also trates some special characters such as ' in the example above.
The function accepts all arguments that can be passed to the slugify function of python-slugify.
For example to change the output from it-s-a-random-version` to it_s_a_random_version, the separator parameter would be passed: slugify(separator='_').
UUID4 extension¶
New in Cookiecutter 1.x
The cookiecutter.extensions.UUIDExtension extension provides a uuid4()
method in templates that generates a uuid4.
Generate a uuid4 string:
{{ uuid4() }}
Outputs:
83b5de62-31b4-4a1e-83fa-8c548de65a11
Working with line-ends special symbols LF/CRLF¶
New in Cookiecutter 2.0
Note
Before version 2.0 Cookiecutter silently used system line end character. LF for POSIX and CRLF for Windows. Since version 2.0 this behaviour changed and now can be forced at template level.
By default Cookiecutter checks every file at render stage and uses the same line end as in source.
This allow template developers to have both types of files in the same template.
Developers should correctly configure their .gitattributes file to avoid line-end character overwrite by git.
The special template variable _new_lines enforces a specific line ending.
Acceptable variables: '\n\r' for CRLF and '\n' for POSIX.
Here is example how to force line endings to CRLF on any deployment:
{
"project_slug": "sample",
"_new_lines": "\n\r"
}
Local Extensions¶
New in Cookiecutter 2.1
A template may extend the Cookiecutter environment with local extensions. These can be part of the template itself, providing it with more sophisticated custom tags and filters.
To do so, a template author must specify the required extensions in cookiecutter.json as follows:
{
"project_slug": "Foobar",
"year": "{% now 'utc', '%Y' %}",
"_extensions": ["local_extensions.FoobarExtension"]
}
This example uses a simple module local_extensions.py which exists in the template root, containing the following (for instance):
from jinja2.ext import Extension
class FoobarExtension(Extension):
def __init__(self, environment):
super(FoobarExtension, self).__init__(environment)
environment.filters['foobar'] = lambda v: v * 2
This will register the foobar filter for the template.
For many cases, this will be unneccessarily complicated.
It’s likely that we’d only want to register a single function as a filter. For this, we can use the simple_filter decorator:
{
"project_slug": "Foobar",
"year": "{% now 'utc', '%Y' %}",
"_extensions": ["local_extensions.simplefilterextension"]
}
from cookiecutter.utils import simple_filter
@simple_filter
def simplefilterextension(v):
return v * 2
This snippet will achieve the exact same result as the previous one.
For complex use cases, a python module local_extensions (a folder with an __init__.py) can also be created in the template root.
Here, for example, a module main.py would have to export all extensions with from .main import FoobarExtension, simplefilterextension or from .main import * in the __init__.py.
Troubleshooting¶
I’m having trouble generating Jinja templates from Jinja templates¶
Make sure you escape things properly, like this:
{{ "{{" }}
Or this:
{% raw %}
<p>Go <a href="{{ url_for('home') }}">Home</a></p>
{% endraw %}
Or this:
{{ {{ url_for('home') }} }}
See http://jinja.pocoo.org/docs/templates/#escaping for more info.
You can also use the _copy_without_render key in your cookiecutter.json file to escape entire files and directories.
Other common issues¶
TODO: add a bunch of common new user issues here.
This document is incomplete. If you have knowledge that could help other users, adding a section or filing an issue with details would be greatly appreciated.
API Reference¶
API¶
This is the Cookiecutter modules API documentation.
Project Info¶
Contributing¶
Contributions are welcome, and they are greatly appreciated! Every little bit helps, and credit will always be given.
Types of Contributions¶
You can contribute in many ways:
Report Bugs¶
Report bugs at https://github.com/cookiecutter/cookiecutter/issues.
If you are reporting a bug, please include:
Your operating system name and version.
Any details about your local setup that might be helpful in troubleshooting.
If you can, provide detailed steps to reproduce the bug.
If you don’t have steps to reproduce the bug, just note your observations in as much detail as you can. Questions to start a discussion about the issue are welcome.
Fix Bugs¶
Look through the GitHub issues for bugs. Anything tagged with “bug” is open to whoever wants to implement it.
Implement Features¶
Look through the GitHub issues for features. Anything tagged with “enhancement” and “please-help” is open to whoever wants to implement it.
Please do not combine multiple feature enhancements into a single pull request.
Note: this project is very conservative, so new features that aren’t tagged with “please-help” might not get into core. We’re trying to keep the code base small, extensible, and streamlined. Whenever possible, it’s best to try and implement feature ideas as separate projects outside of the core codebase.
Write Documentation¶
Cookiecutter could always use more documentation, whether as part of the official Cookiecutter docs, in docstrings, or even on the web in blog posts, articles, and such.
If you want to review your changes on the documentation locally, you can do:
pip install -r docs/requirements.txt
make servedocs
This will compile the documentation, open it in your browser and start watching the files for changes, recompiling as you save.
Submit Feedback¶
The best way to send feedback is to file an issue at https://github.com/cookiecutter/cookiecutter/issues.
If you are proposing a feature:
Explain in detail how it would work.
Keep the scope as narrow as possible, to make it easier to implement.
Remember that this is a volunteer-driven project, and that contributions are welcome :)
Setting Up the Code for Local Development¶
Here’s how to set up cookiecutter for local development.
Fork the
cookiecutterrepo on GitHub.Clone your fork locally:
git clone git@github.com:your_name_here/cookiecutter.git
Install your local copy into a virtualenv. Assuming you have virtualenvwrapper installed, this is how you set up your fork for local development:
cd cookiecutter/ pip install -e .Create a branch for local development:
git checkout -b name-of-your-bugfix-or-feature
Now you can make your changes locally.
When you’re done making changes, check that your changes pass the tests and lint check:
pip install tox tox
Please note that tox runs lint check automatically, since we have a test environment for it.
If you feel like running only the lint environment, please use the following command:
make lint
Ensure that your feature or commit is fully covered by tests. Check report after regular tox run. You can also run coverage only report and get html report with statement by statement highlighting:
make coverage
You report will be placed to
htmlcovdirectory. Please do not include this directory to your commits. By default this directory in our.gitignorefile.Commit your changes and push your branch to GitHub:
git add . git commit -m "Your detailed description of your changes." git push origin name-of-your-bugfix-or-featureSubmit a pull request through the GitHub website.
Contributor Guidelines¶
Pull Request Guidelines¶
Before you submit a pull request, check that it meets these guidelines:
The pull request should include tests.
The pull request should be contained: if it’s too big consider splitting it into smaller pull requests.
If the pull request adds functionality, the docs should be updated. Put your new functionality into a function with a docstring, and add the feature to the list in README.md.
The pull request must pass all CI/CD jobs before being ready for review.
If one CI/CD job is failing for unrelated reasons you may want to create another PR to fix that first.
Coding Standards¶
PEP8
Functions over classes except in tests
Quotes via http://stackoverflow.com/a/56190/5549
Use double quotes around strings that are used for interpolation or that are natural language messages
Use single quotes for small symbol-like strings (but break the rules if the strings contain quotes)
Use triple double quotes for docstrings and raw string literals for regular expressions even if they aren’t needed.
Example:
LIGHT_MESSAGES = { 'English': "There are %(number_of_lights)s lights.", 'Pirate': "Arr! Thar be %(number_of_lights)s lights." } def lights_message(language, number_of_lights): """Return a language-appropriate string reporting the light count.""" return LIGHT_MESSAGES[language] % locals() def is_pirate(message): """Return True if the given message sounds piratical.""" return re.search(r"(?i)(arr|avast|yohoho)!", message) is not None
Testing with tox¶
Tox uses py.test under the hood, hence it supports the same syntax for selecting tests.
For further information please consult the pytest usage docs.
To run a particular test class with tox:
tox -e py '-k TestFindHooks'
To run some tests with names matching a string expression:
tox -e py '-k generate'
Will run all tests matching “generate”, test_generate_files for example.
To run just one method:
tox -e py '-k "TestFindHooks and test_find_hook"'
To run all tests using various versions of python in virtualenvs defined in tox.ini, just run tox:
tox
This configuration file setup the pytest-cov plugin and it is an additional dependency. It generate a coverage report after the tests.
It is possible to test with specific versions of Python. To do this, the command is:
tox -e py37,py38
This will run py.test with the python3.7 and python3.8 interpreters.
Core Committer Guide¶
Vision and Scope¶
Core committers, use this section to:
Guide your instinct and decisions as a core committer
Limit the codebase from growing infinitely
Command-Line Accessible¶
Provides a command-line utility that creates projects from cookiecutters
Extremely easy to use without having to think too hard
Flexible for more complex use via optional arguments
API Accessible¶
Entirely function-based and stateless (Class-free by intentional design)
Usable in pieces for developers of template generation tools
Being Jinja2-specific¶
Sets a standard baseline for project template creators, facilitating reuse
Minimizes the learning curve for those who already use Flask or Django
Minimizes scope of Cookiecutter codebase
Extensible¶
Being extendable by people with different ideas for Jinja2-based project template tools.
Entirely function-based
Aim for statelessness
Lets anyone write more opinionated tools
Freedom for Cookiecutter users to build and extend.
No officially-maintained cookiecutter templates, only ones by individuals
Commercial project-friendly licensing, allowing for private cookiecutters and private Cookiecutter-based tools
Fast and Focused¶
Cookiecutter is designed to do one thing, and do that one thing very well.
Cover the use cases that the core committers need, and as little as possible beyond that :)
Generates project templates from the command-line or API, nothing more
Minimize internal line of code (LOC) count
Ultra-fast project generation for high performance downstream tools
Inclusive¶
Cross-platform and cross-version support are more important than features/functionality
Fixing Windows bugs even if it’s a pain, to allow for use by more beginner coders
Stable¶
Aim for 100% test coverage and covering corner cases
No pull requests will be accepted that drop test coverage on any platform, including Windows
Conservative decisions patterned after CPython’s conservative decisions with stability in mind
Stable APIs that tool builders can rely on
New features require a +1 from 3 core committers
VCS-Hosted Templates¶
Cookiecutter project templates are intentionally hosted VCS repos as-is.
They are easily forkable
It’s easy for users to browse forks and files
They are searchable via standard Github/Bitbucket/other search interface
Minimizes the need for packaging-related cruft files
Easy to create a public project template and host it for free
Easy to collaborate
Process: Pull Requests¶
If a pull request is untriaged:
Look at the roadmap
Set it for the milestone where it makes the most sense
Add it to the roadmap
How to prioritize pull requests, from most to least important:
Fixes for broken tests. Broken means broken on any supported platform or Python version.
Extra tests to cover corner cases.
Minor edits to docs.
Bug fixes.
Major edits to docs.
Features.
Pull Requests Review Guidelines¶
Think carefully about the long-term implications of the change. How will it affect existing projects that are dependent on this? If this is complicated, do we really want to maintain it forever?
Take the time to get things right, PRs almost always require additional improvements to meet the bar for quality. Be very strict about quality.
When you merge a pull request take care of closing/updating every related issue explaining how they were affected by those changes. Also, remember to add the author to
AUTHORS.md.
Process: Issues¶
If an issue is a bug that needs an urgent fix, mark it for the next patch release. Then either fix it or mark as please-help.
For other issues: encourage friendly discussion, moderate debate, offer your thoughts.
New features require a +1 from 2 other core committers (besides yourself).
Process: Roadmap¶
The roadmap located here
Due dates are flexible. Core committers can change them as needed. Note that GitHub sort on them is buggy.
How to number milestones:
Follow semantic versioning. Look at: http://semver.org
Milestone size:
If a milestone contains too much, move some to the next milestone.
Err on the side of more frequent patch releases.
Process: Your own code changes¶
All code changes, regardless of who does them, need to be reviewed and merged by someone else. This rule applies to all the core committers.
Exceptions:
Minor corrections and fixes to pull requests submitted by others.
While making a formal release, the release manager can make necessary, appropriate changes.
Small documentation changes that reinforce existing subject matter. Most commonly being, but not limited to spelling and grammar corrections.
Responsibilities¶
Ensure cross-platform compatibility for every change that’s accepted. Windows, macOS and Linux.
Create issues for any major changes and enhancements that you wish to make. Discuss things transparently and get community feedback.
Don’t add any classes to the codebase unless absolutely needed. Err on the side of using functions.
Keep feature versions as small as possible, preferably one new feature per version.
Be welcoming to newcomers and encourage diverse new contributors from all backgrounds. Look at Code of Conduct.
Becoming a Core Committer¶
Contributors may be given core commit privileges. Preference will be given to those with:
Past contributions to Cookiecutter and other open-source projects. Contributions to Cookiecutter include both code (both accepted and pending) and friendly participation in the issue tracker. Quantity and quality are considered.
A coding style that the other core committers find simple, minimal, and clean.
Access to resources for cross-platform development and testing.
Time to devote to the project regularly.
Credits¶
Development Leads¶
Audrey Roy Greenfeld (@audreyfeldroy)
Daniel Roy Greenfeld (@pydanny)
Raphael Pierzina (@hackebrot)
Core Committers¶
Michael Joseph (@michaeljoseph)
Paul Moore (@pfmoore)
Andrey Shpak (@insspb)
Sorin Sbarnea (@ssbarnea)
Fábio C. Barrionuevo da Luz (@luzfcb)
Simone Basso (@simobasso)
Jens Klein (@jensens)
Érico Andrei (@ericof)
Contributors¶
Steven Loria (@sloria)
Goran Peretin (@gperetin)
Hamish Downer (@foobacca)
Thomas Orozco (@krallin)
Jindrich Smitka (@s-m-i-t-a)
Benjamin Schwarze (@benjixx)
Raphi (@raphigaziano)
Thomas Chiroux (@ThomasChiroux)
Sergi Almacellas Abellana (@pokoli)
Alex Gaynor (@alex)
Rolo (@rolo)
Pablo (@oubiga)
Bruno Rocha (@rochacbruno)
Alexander Artemenko (@svetlyak40wt)
Mahmoud Abdelkader (@mahmoudimus)
Leonardo Borges Avelino (@lborgav)
Chris Trotman (@solarnz)
Rolf (@relekang)
Noah Kantrowitz (@coderanger)
Vincent Bernat (@vincentbernat)
Germán Moya (@pbacterio)
Ned Batchelder (@nedbat)
Dave Dash (@davedash)
Johan Charpentier (@cyberj)
Éric Araujo (@merwok)
saxix (@saxix)
Tzu-ping Chung (@uranusjr)
Caleb Hattingh (@cjrh)
Flavio Curella (@fcurella)
Adam Venturella (@aventurella)
Monty Taylor (@emonty)
schacki (@schacki)
Ryan Olson (@ryanolson)
Trey Hunner (@treyhunner)
Russell Keith-Magee (@freakboy3742)
Mishbah Razzaque (@mishbahr)
Robin Andeer (@robinandeer)
Rachel Sanders (@trustrachel)
Rémy Hubscher (@Natim)
Dino Petron3 (@dinopetrone)
Peter Inglesby (@inglesp)
Ramiro Batista da Luz (@ramiroluz)
Omer Katz (@thedrow)
lord63 (@lord63)
Randy Syring (@rsyring)
Mark Jones (@mark0978)
Marc Abramowitz (@msabramo)
Lucian Ursu (@LucianU)
Osvaldo Santana Neto (@osantana)
Matthias84 (@Matthias84)
Simeon Visser (@svisser)
Guruprasad (@lgp171188)
Charles-Axel Dein (@charlax)
Diego Garcia (@drgarcia1986)
maiksensi (@maiksensi)
Andrew Conti (@agconti)
Valentin Lab (@vaab)
Ilja Bauer (@iljabauer)
Elias Dorneles (@eliasdorneles)
Matias Saguir (@mativs)
Johannes (@johtso)
macrotim (@macrotim)
Will McGinnis (@wdm0006)
Cédric Krier (@cedk)
Tim Osborn (@ptim)
Aaron Gallagher (@habnabit)
mozillazg (@mozillazg)
Joachim Jablon (@ewjoachim)
Andrew Ittner (@tephyr)
Diane DeMers Chen (@purplediane)
zzzirk (@zzzirk)
Carol Willing (@willingc)
phoebebauer (@phoebebauer)
Adam Chainz (@adamchainz)
Sulé (@suledev)
Evan Palmer (@palmerev)
Bruce Eckel (@BruceEckel)
Robert Lyon (@ivanlyon)
Terry Bates (@terryjbates)
Brett Cannon (@brettcannon)
Michael Warkentin (@mwarkentin)
Bartłomiej Kurzeja (@B3QL)
Thomas O’Donnell (@andytom)
Jeremy Carbaugh (@jcarbaugh)
Nathan Cheung (@cheungnj)
Abdó Roig-Maranges (@aroig)
Steve Piercy (@stevepiercy)
Corey (@coreysnyder04)
Dmitry Evstratov (@devstrat)
Eyal Levin (@eyalev)
mathagician (@mathagician)
Guillaume Gelin (@ramnes)
@delirious-lettuce (@delirious-lettuce)
Gasper Vozel (@karantan)
Joshua Carp (@jmcarp)
@meahow (@meahow)
Andrea Grandi (@andreagrandi)
Issa Jubril (@jubrilissa)
Nytiennzo Madooray (@Nythiennzo)
Erik Bachorski (@dornheimer)
cclauss (@cclauss)
Andy Craze (@accraze)
Anthony Sottile (@asottile)
Jonathan Sick (@jonathansick)
Hugo (@hugovk)
Min ho Kim (@minho42)
Ryan Ly (@rly)
Akintola Rahmat (@mihrab34)
Jai Ram Rideout (@jairideout)
Diego Carrasco Gubernatis (@dacog)
Wagner Negrão (@wagnernegrao)
Josh Barnes (@jcb91)
Nikita Sobolev (@sobolevn)
Matt Stibbs (@mattstibbs)
MinchinWeb (@MinchinWeb)
kishan (@kishan)
tonytheleg (@tonytheleg)
Roman Hartmann (@RomHartmann)
DSEnvel (@DSEnvel)
kishan (@kishan)
Bruno Alla (@browniebroke)
nicain (@nicain)
Carsten Rösnick-Neugebauer (@croesnick)
igorbasko01 (@igorbasko01)
Dan Booth Dev (@DanBoothDev)
Pablo Panero (@ppanero)
Chuan-Heng Hsiao (@chhsiao1981)
Mohammad Hossein Sekhavat (@mhsekhavat)
Amey Joshi (@amey589)
Paul Harrison (@smoothml)
Fabio Todaro (@SharpEdgeMarshall)
Nicholas Bollweg (@bollwyvl)
Jace Browning (@jacebrowning)
Ionel Cristian Mărieș (@ionelmc)
Kishan Mehta (@kishan3)
Wieland Hoffmann (@mineo)
Antony Lee (@anntzer)
Aurélien Gâteau (@agateau)
Axel H. (@noirbizarre)
Chris (@chrisbrake)
Chris Streeter (@streeter)
Gábor Lipták (@gliptak)
Javier Sánchez Portero (@javiersanp)
Nimrod Milo (@milonimrod)
Philipp Kats (@Casyfill)
Reinout van Rees (@reinout)
Rémy Greinhofer (@rgreinho)
Sebastian (@sebix)
Stuart Mumford (@Cadair)
Tom Forbes (@orf)
Xie Yanbo (@xyb)
Maxim Ivanov (@ivanovmg)
Backers¶
We would like to thank the following people for supporting us in our efforts to maintain and improve Cookiecutter:
Alex DeBrie
Alexandre Y. Harano
Bruno Alla
Carol Willing
Russell Keith-Magee
Sprint Contributors¶
PyCon 2016 Sprint¶
The following people made contributions to the cookiecutter project at the PyCon sprints in Portland, OR from June 2-5 2016. Contributions include user testing, debugging, improving documentation, reviewing issues, writing tutorials, creating and updating project templates, and teaching each other.
Adam Chainz (@adamchainz)
Andrew Ittner (@tephyr)
Audrey Roy Greenfeld (@audreyr)
Carol Willing (@willingc)
Christopher Clarke (@chrisdev)
Citlalli Murillo (@citmusa)
Daniel Roy Greenfeld (@pydanny)
Diane DeMers Chen (@purplediane)
Elaine Wong (@elainewong)
Elias Dorneles (@eliasdorneles)
Emily Cain (@emcain)
John Roa (@jhonjairoroa87)
Jonan Scheffler (@1337807)
Phoebe Bauer (@phoebebauer)
Kartik Sundararajan (@skarbot)
Katia Lira (@katialira)
Leonardo Jimenez (@xpostudio4)
Lindsay Slazakowski (@lslaz1)
Meghan Heintz (@dot2dotseurat)
Raphael Pierzina (@hackebrot)
Umair Ashraf (@umrashrf)
Valdir Stumm Junior (@stummjr)
Vivian Guillen (@viviangb)
Zaro (@zaro0508)
History¶
History is important, but our current roadmap can be found here
2.1.1 (2022-06-01)¶
Documentation updates¶
Fix local extensions documentation (#1686) @alkatar21
Bugfixes¶
Sanitize Mercurial branch information before checkout. (#1689) @ericof
This release is made by wonderfull contributors:¶
@alkatar21, @ericof and @jensens
2.1.0 (2022-05-30)¶
Changes¶
Move contributors and backers to credits section (#1599) @doobrie
test_generate_file_verbose_template_syntax_error fixed (#1671) @MaciejPatro
Removed changes related to setuptools_scm (#1629) @ozer550
Feature/local extensions (#1240) @mwesterhof
CI/CD and QA changes¶
Check manifest: pre-commit, fixes, cleaning (#1683) @jensens
Follow PyPA guide to release package using GitHub Actions. (#1682) @ericof
Documentation updates¶
Fix typo in dict_variables.rst (#1680) @ericof
Documentation overhaul (#1677) @jensens
Fixed incorrect link on docs. (#1649) @luzfcb
Bugfixes¶
Restore accidentally deleted support for click 8.x (#1643) @jaklan
This release was made possible by our wonderful contributors:¶
@doobrie, @jensens, @ericof, @luzfcb
2.0.2 (2021-12-27)¶
Remark: This release never made it to official PyPI
Fix Python version number in cookiecutter –version and test on Python 3.10 (#1621) @ozer550
Removed changes related to setuptools_scm (#1629) @audreyfeldroy @ozer550
2.0.1 (2021-12-11)¶
Remark: This release never made it to official PyPI
Breaking Changes¶
Release preparation for 2.0.1rc1 (#1608) @audreyfeldroy
Replace poyo with pyyaml. (#1489) @dHannasch
Added: Path templates will be rendered when copy_without_render used (#839) @noirbizarre
Added: End of line detection and configuration. (#1407) @insspb
Remove support for python2.7 (#1386) @ssbarnea
Minor Changes¶
Adopt setuptools-scm packaging (#1577) @ssbarnea
Log the error message when git clone fails, not just the return code (#1505) @logworthy
allow jinja 3.0.0 (#1548) @wouterdb
Added uuid extension to be able to generate uuids (#1493) @jonaswre
Alert user if choice is invalid (#1496) @dHannasch
Replace poyo with pyyaml. (#1489) @dHannasch
update AUTHOR lead (#1532) @HosamAlmoghraby
Add Python 3.9 (#1478) @gliptak
Added: –list-installed cli option, listing already downloaded cookiecutter packages (#1096) @chrisbrake
Added: Jinja2 Environment extension on files generation stage (#1419) @insspb
Added: –replay-file cli option, for replay file distributing (#906) @Cadair
Added: _output_dir to cookiecutter context (#1034) @Casyfill
Added: CLI option to ignore hooks (#992) @rgreinho
Changed: Generated projects can use multiple type hooks at same time. (sh + py) (#974) @milonimrod
Added: Path templates will be rendered when copy_without_render used (#839) @noirbizarre
Added: End of line detection and configuration. (#1407) @insspb
Making code python 3 only: Remove python2 u’ sign, fix some strings (#1402) @insspb
py3: remove futures, six and encoding (#1401) @insspb
Render variables starting with an underscore. (#1339) @smoothml
Tests refactoring: test_utils write issues fixed #1405 (#1406) @insspb
CI/CD and QA changes¶
enable branch coverage (#1542) @simobasso
Make release-drafter diff only between master releases (#1568) @SharpEdgeMarshall
ensure filesystem isolation during tests execution (#1564) @simobasso
add safety ci step (#1560) @simobasso
pre-commit: add bandit hook (#1559) @simobasso
Replace tmpdir in favour of tmp_path (#1545) @SharpEdgeMarshall
Fix linting in CI (#1546) @SharpEdgeMarshall
Coverage 100% (#1526) @SharpEdgeMarshall
Run coverage with matrix (#1521) @SharpEdgeMarshall
Lint rst files (#1443) @ssbarnea
Python3: Changed io.open to build-in open (PEP3116) (#1408) @insspb
Making code python 3 only: Remove python2 u’ sign, fix some strings (#1402) @insspb
py3: remove futures, six and encoding (#1401) @insspb
Removed: Bumpversion, setup.py arguments. (#1404) @insspb
Tests refactoring: test_utils write issues fixed #1405 (#1406) @insspb
Added: Automatic PyPI deploy on tag creation (#1400) @insspb
Changed: Restored coverage reporter (#1399) @insspb
Documentation updates¶
Fix pull requests checklist reference (#1537) @glumia
Fix author name (#1544) @HosamAlmoghraby
Add missing contributors (#1535) @glumia
Update CONTRIBUTING.md (#1529) @glumia
Update LICENSE (#1519) @simobasso
docs: rewrite the conditional files / directories example description. (#1437) @lyz-code
Fix incorrect years in release history (#1473) @graue70
Add slugify in the default extensions list (#1470) @oncleben31
Renamed cookiecutter.package to API (#1442) @grrlic
Fixed wording detail (#1427) @steltenpower
Changed: CLI Commands documentation engine (#1418) @insspb
Added: Example for conditional files / directories in hooks (#1397) @xyb
Changed: README.md PyPI URLs changed to the modern PyPI last version (#1391) @brettcannon
Fixed: Comma in README.md (#1390) @Cy-dev-tex
Fixed: Replaced no longer maintained pipsi by pipx (#1395) @ndclt
Bugfixes¶
Add support for click 8.x (#1569) @cjolowicz
Force click<8.0.0 (#1562) @SharpEdgeMarshall
Remove direct dependency on markupsafe (#1549) @ssbarnea
fixes prompting private rendered dicts (#1504) @juhuebner
User’s JSON parse error causes ugly Python exception #809 (#1468) @noone234
config: set default on missing default_context key (#1516) @simobasso
Fixed: Values encoding on Windows (#1414) @agateau
Fixed: Fail with gitolite repositories (#1144) @javiersanp
MANIFEST: Fix file name extensions (#1387) @sebix
Deprecations¶
Removed: Bumpversion, setup.py arguments. (#1404) @insspb
Removed support for Python 3.6 and PyPy (#1608) @audreyfeldroy
This release was made possible by our wonderful contributors:¶
@Cadair, @Casyfill, @Cy-dev-tex, @HosamAlmoghraby, @SharpEdgeMarshall, @agateau, @audreyfeldroy, @brettcannon, @chrisbrake, @cjolowicz, @dHannasch, @gliptak, @glumia, @graue70, @grrlic, @insspb, @javiersanp, @jonaswre, @jsoref, @Jthevos, @juhuebner, @logworthy, @lyz-code, @milonimrod, @ndclt, @noirbizarre, @noone234, @oncleben31, @ozer550, @rgreinho, @sebix, @Sahil-101, @simobasso, @smoothml, @ssbarnea, @steltenpower, @wouterdb, @xyb, Christopher Wolfe and Hosam Almoghraby ( RIAG Digital )
1.7.2 (2020-04-21)¶
Fixed: Jinja2&Six version limits causing build errors with ansible project @insspb (#1385)
1.7.1 (2020-04-21)¶
This release was focused on internal code and CI/CD changes. During this release all code was verified to match pep8, pep257 and other code-styling guides. Project CI/CD was significantly changed, Windows platform checks based on Appveyor engine was replaced by GitHub actions tests. Appveyor was removed. Also our CI/CD was extended with Mac builds, to verify project builds on Apple devices.
Important Changes:
Added: Added debug messages for get_user_config @ssbarnea (#1357)
Multiple templates per one repository feature added. @RomHartmann (#1224, #1063)
Update replay.py json.dump indent for easy viewing @nicain (#1293)
‘future’ library replaced with ‘six’ as a more lightweight python porting library @asottile (#941)
Added extension: Slugify template filter @ppanero (#1336)
Added command line option:
--skip-if-file-exists, allow to skip the existing files when doingoverwrite_if_exists. @chhsiao1981 (#1076)Some packages versions limited to be compatible with python2.7 and python 3.5 @insspb (#1349)
Internal CI/CD and tests changes:
Coverage comment in future merge requests disabled @ssbarnea (#1279)
Fixed Python 3.8 travis tests and setup.py message @insspb (#1295, #1297)
Travis builds extended with Windows setup for all supported python versions @insspb (#1300, #1301)
Update .travis.yml to be compatible with latest travis cfg specs @luzfcb (#1346)
Added new test to improve tests coverage @amey589 (#1023)
Added missed coverage lines highlight to pytest-coverage report @insspb (#1352)
pytest-catchlog package removed from test_requirements, as now it is included in pytest @insspb (#1347)
Fixed
cov-reporttox invocation environment @insspb (#1350)Added: Release drafter support and configuration to exclude changelog update work and focus on development @ssbarnea @insspb (#1356, #1362)
Added: CI/CD steps for Github actions to speedup CI/CD @insspb (#1360)
Removed: Appveyor CI/CD completely removed @insspb @ssbarnea @insspb (#1363, #1367)
Code style and docs changes:
Added black formatting verification on lint stage + project files reformatting @ssbarnea @insspb (#1368)
Added pep257 docstring for tests/* files @insspb (#1369, #1370, #1371, #1372, #1373, #1374, #1375, #1376, #1377, #1378, #1380, #1381)
Added pep257 docstring for tests/conftests.py @kishan (#1272, #1263)
Added pep257 docstring for tests/replay/conftest.py @kishan (#1270, #1268)
Added pep257 docstring for docs/init.py @kishan (#1273, #1265)
Added missing docstring headers to all files @croesnick (#1269, #1283)
Gitter links replaced by Slack in README @browniebroke (#1282)
flake8-docstrings tests added to CI/CD @ssbarnea (#1284)
Activated pydocstyle rule: D401 - First line should be in imperative mood @ssbarnea (#1285)
Activated pydocstyle rule: D200 - One-line docstring should fit on one line with quotes @ssbarnea (#1288)
Activated pydocstyle rule: D202 - No blank lines allowed after function docstring @ssbarnea (#1288)
Activated pydocstyle rule: D205 - 1 blank line required between summary line and description @ssbarnea (#1286, #1287)
Activated pydocstyle rule: ABS101 @ssbarnea (#1288)
Replaced click documentation links to point to version 7 @igorbasko01 (#1303)
Updated submodule link to latest version with documentation links fix @DanBoothDev (#1388)
Fixed links in main README file. @insspb (#1342)
Fix indentation of .cookiecutterrc in README.md @mhsekhavat (#1322)
Changed format of loggers invocation @insspb (#1307)
1.7.0 (2019-12-22) Old friend¶
Important Changes:
Drop support for EOL Python 3.4, thanks to @jamescurtin and @insspb (#1024)
Drop support for EOL Python 3.3, thanks to @hugovk (#1024)
Increase the minimum click version to 7.0, thanks to @rly and @luzfcb (#1168)
Other Changes:
PEP257 fixing docstrings in exceptions.py. Thanks to @MinchinWeb (#1237)
PEP257 fixing docstrings in replay.py. Thanks to @kishan (#1234)
PEP257 fixing docstrings in test_unzip.py. Thanks to @tonytheleg and @insspb (#1236, #1262)
Fixed tests sequence for appveyor, to exclude file not found bug. Thanks to @insspb (#1257)
Updates REAMDE.md with svg badge for appveyor. Thanks to @sobolevn (#1254)
Add missing {% endif %} to Choice Variables example. Thanks to @mattstibbs (#1249)
Core documentation converted to Markdown format thanks to @wagnernegrao, @insspb (#1216)
Tests update: use sys.executable when invoking python in python 3 only environment thanks to @vincentbernat (#1221)
Prevent
clickAPI v7.0 from showing choices when already shown, thanks to @rly and @luzfcb (#1168)Test the codebase with python3.8 beta on tox and travis-ci (#1206), thanks to @mihrab34
Add a CODE_OF_CONDUCT.md file to the project, thanks to @andreagrandi (#1009)
Update docstrings in
cookiecutter/main.py,cookiecutter/__init__.py, andcookiecutter/log.pyto follow the PEP 257 style guide, thanks to @meahow (#998, #999, #1000)Update docstrings in
cookiecutter/utils.pyto follow the PEP 257 style guide, thanks to @dornheimer(#1026)Fix grammar in Choice Variables documentation, thanks to @jubrilissa (#1011)
Update installation docs with links to the Windows Subsystem and GNU utilities, thanks to @Nythiennzo for the PR and @BruceEckel for the review (#1016)
Upgrade flake8 to version 3.5.0, thanks to @cclauss (#1038)
Update tutorial with explanation for how cookiecutter finds the template file, thanks to @accraze(#1025)
Update CI config files to use
TOXENVenvironment variable, thanks to @asottile (#1019)Improve user documentation for writing hooks, thanks to @jonathansick (#1057)
Make sure to preserve the order of items in the generated cookiecutter context, thanks to @hackebrot (#1074)
Fixed DeprecationWarning for a regular expression on python 3.6, thanks to @reinout (#1124)
Document use of cookiecutter-template topic on GitHub, thanks to @ssbarnea (#1189)
Update README badge links, thanks to @luzfcb (#1207)
Update prompt.py to match pep257 guidelines, thanks to @jairideout (#1105)
Update link to Jinja2 extensions documentation, thanks to @dacog (#1193)
Require pip 9.0.0 or newer for tox environments, thanks to @hackebrot (#1215)
Use io.open contextmanager when reading hook files, thanks to @jcb91 (#1147)
Add more cookiecutter templates to the mix:
cookiecutter-python-cli by @xuanluong (#1003)
cookiecutter-docker-science by @takahi-i (#1040)
cookiecutter-flask-skeleton by @mjhea0 (#1052)
cookiecutter-awesome by @Pawamoy (#1051)
cookiecutter-flask-ask by @machinekoder (#1056)
cookiecutter-data-driven-journalism by @JAStark (#1020)
cookiecutter-tox-plugin by @obestwalter (#1103)
cookiecutter-django-dokku by @mashrikt (#1093)
1.6.0 (2017-10-15) Tim Tam¶
New Features:
Include template path or template URL in cookiecutter context under
_template, thanks to @aroig (#774)Add a URL abbreviation for GitLab template projects, thanks to @hackebrot (#963)
Add option to use templates from Zip files or Zip URLs, thanks to @freakboy3742 (#961)
Bug Fixes:
Fix an issue with missing default template abbreviations for when a user defined custom abbreviations, thanks to @noirbizarre for the issue report and @hackebrot for the fix (#966, #967)
Preserve existing output directory on project generation failure, thanks to @ionelmc for the report and @michaeljoseph for the fix (#629, #964)
Fix Python 3.x error handling for
gitoperation failures, thanks to @jmcarp (#905)
Other Changes:
Fix broken link to Copy without Render docs, thanks to @coreysnyder04 (#912)
Improve debug log message for when a hook is not found, thanks to @raphigaziano (#160)
Fix module summary and
expand_abbreviations()doc string as per pep257, thanks to @terryjbates (#772)Update doc strings in
cookiecutter/cli.pyandcookiecutter/config.pyaccording to pep257, thanks to @terryjbates (#922, #931)Update doc string for
is_copy_only_path()according to pep257, thanks to @mathagician and @terryjbates (#935, #949)Update doc strings in
cookiecutter/extensions.pyaccording to pep257, thanks to @meahow (#996)Fix miscellaneous issues with building docs, thanks to @stevepiercy (#889)
Re-implement Makefile and update several make rules, thanks to @hackebrot (#930)
Fix broken link to pytest docs, thanks to @eyalev for the issue report and @devstrat for the fix (#939, #940)
Add
test_requirements.txtfile for easier testing outside of tox, thanks to @ramnes (#945)Improve wording in copy without render docs, thanks to @eyalev (#938)
Fix a number of typos, thanks to @delirious-lettuce (#968)
Improved extra context docs by noting that extra context keys must be present in the template’s
cookiecutter.json, thanks to @karantan for the report and fix (#863, #864)Added more cookiecutter templates to the mix:
cookiecutter-kata-cpputest by @13coders (#901)
cookiecutter-kata-gtest by @13coders (#901)
cookiecutter-pyramid-talk-python-starter by @mikeckennedy (#915)
cookiecutter-android by @alexfu (#890)
cookiecutter-lux-python by @alexkey (#895)
cookiecutter-git by @tuxredux (#921)
cookiecutter-ansible-role-ci by @ferrarimarco (#903)
cookiecutter_dotfile by @bdcaf (#925)
cookiecutter-molecule by @retr0h (#954)
sublime-snippet-package-template by @agenoria (#956)
cookiecutter-conda-python by @conda (#969)
cookiecutter-flask-minimal by @candidtim (#977)
cookiecutter-pypackage-rust-cross-platform-publish by @mckaymatt (#957)
cookie-cookie by @tuxredux (#951)
cookiecutter-telegram-bot by @Ars2014 (#984)
python-project-template by @Kwpolska (#986)
wemake-django-template by @wemake-services (#990)
cookiecutter-raml by @genzj (#994)
cookiecutter-anyblok-project by @AnyBlok (#988)
cookiecutter-devenv by @greenguavalabs (#991)
1.5.1 (2017-02-04) Alfajor¶
New Features:
Major update to installation documentation, thanks to @stevepiercy (#880)
Bug Fixes:
Resolve an issue around default values for dict variables, thanks to @e-kolpakov for raising the issue and @hackebrot for the PR (#882, #884)
Other Changes:
Contributor documentation reST fixes, thanks to @stevepiercy (#878)
Added more cookiecutter templates to the mix:
widget-cookiecutter by @willingc (#781)
cookiecutter-django-foundation by @Parbhat (#804)
cookiecutter-tornado by @hkage (#807)
CICADA by @elenimijalis (#840)
cookiecutter-tf-module by @VDuda (#843)
cookiecutter-pyqt4 by @aeroaks (#847)
cookiecutter-golang by @mjhea0 and @lacion (#872, #873)
cookiecutter-elm, cookiecutter-java and cookiecutter-spring-boot by @m-x-k (#879)
1.5.0 (2016-12-18) Alfajor¶
The primary goal of this release was to add command-line support for passing extra context, address minor bugs and make a number of improvements.
New Features:
Inject extra context with command-line arguments, thanks to @msabramo and @michaeljoseph (#666).
Updated conda installation instructions to work with the new conda-forge distribution of Cookiecutter, thanks to @pydanny and especially @bollwyvl (#232, #705).
Refactor code responsible for interaction with version control systems and raise better error messages, thanks to @michaeljoseph (#778).
Add support for executing cookiecutter using
python -m cookiecutteror from a checkout/zip file, thanks to @brettcannon (#788).New CLI option
--debug-file PATHto store a log file on disk. By default no log file is written. Entries forDEBUGlevel and higher. Thanks to @hackebrot(#792).Existing templates in a user’s
cookiecutters_dir(default is~/.cookiecutters/) can now be referenced by directory name, thanks to @michaeljoseph (#825).Add support for dict values in
cookiecutter.json, thanks to @freakboy3742 and @hackebrot (#815, #858).Add a
jsonifyfilter to default jinja2 extensions that json.dumps a Python object into a string, thanks to @aroig (#791).
Bug Fixes:
Fix typo in the error logging text for when a hook did not exit successfully, thanks to @luzfcb (#656)
Fix an issue around replay file names when cookiecutter is used with a relative path to a template, thanks to @eliasdorneles for raising the issue and @hackebrot for the PR (#752, #753)
Ignore hook files with tilde-suffixes, thanks to @hackebrot (#768)
Fix a minor issue with the code that generates a name for a template, thanks to @hackebrot(#798)
Handle empty hook file or other OS errors, thanks to @christianmlong for raising this bug and @jcarbaugh and @hackebrot for the fix (#632, #729, #862)
Resolve an issue with custom extensions not being loaded for
pre_gen_projectandpost_gen_projecthooks, thanks to @cheungnj (#860)
Other Changes:
Remove external dependencies from tests, so that tests can be run w/o network connection, thanks to @hackebrot (#603)
Remove execute permissions on Python files, thanks to @mozillazg (#650)
Report code coverage info from AppVeyor build to codecov, thanks to @ewjoachim (#670)
Documented functions and methods lacking documentation, thanks to @pydanny (#673)
Documented
__init__methods for Environment objects, thanks to @pydanny (#677)Updated whichcraft to 0.4.0, thanks to @pydanny.
Updated documentation link to Read the Docs, thanks to @natim (#687)
Moved cookiecutter templates and added category links, thanks to @willingc (#674)
Added Github Issue Template, thanks to @luzfcb (#700)
Added
sshrepository examples, thanks to @pokoli (#702)Fix links to the cookiecutter-data-science template and its documentation, thanks to @tephyr for the PR and @willingc for the review (#711, #714)
Update link to docs for Django’s
--templatecommand line option, thanks to @purplediane (#754)Create hook backup files during the tests as opposed to having them as static files in the repository, thanks to @hackebrot (#789)
Applied PEP 257 docstring conventions to:
environment.py, thanks to @terryjbates (#759)find.py, thanks to @terryjbates (#761)generate.py, thanks to @terryjbates (#764)hooks.py, thanks to @terryjbates (#766)repository.py, thanks to @terryjbates (#833)vcs.py, thanks to @terryjbates (#831)
Fix link to the Tryton cookiecutter, thanks to @cedk and @nicoe (#697, #698)
Added PyCon US 2016 sponsorship to README, thanks to @purplediane (#720)
Added a sprint contributor doc, thanks to @phoebebauer (#727)
Converted readthedocs links (.org -> .io), thanks to @adamchainz (#718)
Added Python 3.6 support, thanks to @suledev (#728)
Update occurrences of
repo_namein documentation, thanks to @palmerev (#734)Added case studies document, thanks to @pydanny (#735)
Added first steps cookiecutter creation tutorial, thanks to @BruceEckel (#736)
Reorganised tutorials and setup git submodule to external tutorial, thanks to @dot2dotseurat (#740)
Debian installation instructions, thanks to @ivanlyon (#738)
Usage documentation typo fix., thanks to @terryjbates (#739)
Updated documentation copyright date, thanks to @zzzirk (#747)
Add a make rule to update git submodules, thanks to @hackebrot (#746)
Split up advanced usage docs, thanks to @zzzirk (#749)
Documentation for the
no_inputoption, thanks to @pokoli (#701)Remove unnecessary shebangs from python files, thanks to @michaeljoseph (#763)
Refactor cookiecutter template identification, thanks to @michaeljoseph (#777)
Add a
cli_runnertest fixture to simplify CLI tests, thanks to @hackebrot (#790)Add a check to ensure cookiecutter repositories have JSON context, thanks to @michaeljoseph(#782)
Rename the internal function that determines whether a file should be rendered, thanks to @audreyr for raising the issue and @hackebrotfor the PR (#741, #802)
Fix typo in docs, thanks to @mwarkentin (#828)
Fix broken link to Invoke docs, thanks to @B3QL (#820)
Add documentation to
render_variablefunction inprompt.py, thanks to @pydanny (#678)Fix python3.6 travis-ci and tox configuration, thanks to @luzfcb (#844)
Add missing encoding declarations to python files, thanks to @andytom (#852)
Disable poyo logging for tests, thanks to @hackebrot (#855)
Remove pycache directories in make clean-pyc, thanks to @hackebrot (#849)
Refactor hook system to only find the requested hook, thanks to @michaeljoseph (#834)
Add tests for custom extensions in
pre_gen_projectandpost_gen_projecthooks, thanks to @hackebrot (#856)Make the build reproducible by avoiding nondeterministic keyword arguments, thanks to @lamby and @hackebrot (#800, #861)
Extend CLI help message and point users to the github project to engage with the community, thanks to @hackebrot (#859)
Added more cookiecutter templates to the mix:
cookiecutter-funkload-friendly by @tokibito (#657)
cookiecutter-reveal.js by @keimlink (#660)
cookiecutter-python-app by @mdklatt (#659)
morepath-cookiecutter by @href (#672)
hovercraft-slides by @jhermann (#665)
cookiecutter-es6-package by @ratson (#667)
cookiecutter-webpack by @hzdg (#668)
cookiecutter-django-herokuapp by @dulaccc (#374)
cookiecutter-django-aws-eb by @peterlauri (#626)
wagtail-starter-kit by @tkjone (#658)
cookiecutter-dpf-effect by @SpotlightKid (#663)
cookiecutter-dpf-audiotk by @SpotlightKid (#663)
cookiecutter-template by @eviweb (#664)
cookiecutter-angular2 by @matheuspoleza (#675)
cookiecutter-data-science by @pjbull (#680)
cc_django_ember_app by @nanuxbe (#686)
cc_project_app_drf by @nanuxbe (#686)
cc_project_app_full_with_hooks by @nanuxbe (#686)
beat-generator by @ruflin (#695)
cookiecutter-scala by @Plippe (#751)
cookiecutter-py3tkinter by @ivanlyon (#730)
pyramid-cookiecutter-alchemy by @stevepiercy (#745)
pyramid-cookiecutter-starter by @stevepiercy (#745)
pyramid-cookiecutter-zodb by @stevepiercy (#745)
substanced-cookiecutter by @stevepiercy (#745)
cookiecutter-simple-django-cn by @shenyushun (#765)
cookiecutter-pyqt5 by @mandeepbhutani (#797)
cookiecutter-xontrib by @laerus (#817)
cookiecutter-reproducible-science by @mkrapp (#826)
cc-automated-drf-template by @elenimijalis (#832)
1.4.0 (2016-03-20) Shortbread¶
The goal of this release is changing to a strict Jinja2 environment, paving the way to more awesome in the future, as well as adding support for Jinja2 extensions.
New Features:
Added support for Jinja2 extension support, thanks to @hackebrot (#617).
Now raises an error if Cookiecutter tries to render a template that contains an undefined variable. Makes generation more robust and secure (#586). Work done by @hackebrot (#111, #586, #592)
Uses strict Jinja2 env in prompt, thanks to @hackebrot (#598, #613)
Switched from pyyaml/ruamel.yaml libraries that were problematic across platforms to the pure Python poyo library, thanks to @hackebrot (#557, #569, #621)
User config values for
cookiecutters_dirandreplay_dirnow support environment variable and user home expansion, thanks to @nfarrar for the suggestion and @hackebrot for the PR (#640,#642)Add jinja2-time as default extension for dates and times in templates via
{% now 'utc' %},thanks to @hackebrot (#653)
Bug Fixes:
Provided way to define options that have no defaults, thanks to @johtso (#587, #588)
Make sure that
replay.dump()andreplay.load()use the correct user config, thanks to @hackebrot (#590, #594)Added correct CA bundle for Git on Appveyor, thanks to @maiksensi (#599, #602)
Open
HISTORY.rstwithutf-8encoding when reading the changelog, thanks to @0-wiz-0 for submitting the issue and @hackebrot for the fix (#638, #639)Fix repository indicators for privaterepository urls, thanks to @habnabit for the fix (#595) and @hackebrot for the tests (#655)
Other Changes:
Set path before running tox, thanks to @maiksensi (#615, #620)
Removed xfail in test_cookiecutters, thanks to @hackebrot (#618)
Removed django-cms-plugin on account of 404 error, thanks to @mativs and @pydanny (#593)
Fixed docs/usage.rst, thanks to @macrotim (#604)
Update .gitignore to latest Python.gitignore and ignore PyCharm files, thanks to @audreyr
Use open context manager to read context_file in generate() function, thanks to @hackebrot (#607, #608)
Added documentation for choice variables, thanks to @maiksensi (#611)
Set up Scrutinizer to check code quality, thanks to @audreyr
Drop distutils support in setup.py, thanks to @hackebrot (#606, #609)
Change cookiecutter-pypackage-minimal link, thanks to @kragniz (#614)
Fix typo in one of the template’s description, thanks to @ryanfreckleton (#643)
Fix broken link to _copy_without_render in troubleshooting.rst, thanks to @ptim (#647)
Added more cookiecutter templates to the mix:
cookiecutter-pipproject by @wdm0006 (#624)
cookiecutter-flask-2 by @wdm0006 (#624)
cookiecutter-kotlin-gradle by @thomaslee (#622)
cookiecutter-tryton-fulfilio by @cedk (#631)
django-starter by @tkjone (#635)
django-docker-bootstrap by @legios89 (#636)
cookiecutter-django-gulp by @valerymelou (#648)
1.3.0 (2015-11-10) Pumpkin Spice¶
The goal of this release is to extend the user config feature and to make hook execution more robust.
New Features:
Abort project generation if
pre_gen_projectorpost_gen_projecthook scripts fail, thanks to @eliasdorneles (#464, #549)Extend user config capabilities with additional cli options
--config-fileand--default-configand environment variableCOOKIECUTTER_CONFIG, thanks to @jhermann, @pfmoore, and @hackebrot (#258, #424, #565)
Bug Fixes:
Fixed conditional dependencies for wheels in setup.py, thanks to @hackebrot (#557, #568)
Reverted skipif markers to use correct reasons (bug fixed in pytest), thanks to @hackebrot (#574)
Other Changes:
Improved path and documentation for rendering the Sphinx documentation, thanks to @eliasdorneles and @hackebrot (#562, #583)
Added additional help entrypoints, thanks to @michaeljoseph (#563, #492)
Added Two Scoops Academy to the README, thanks to @hackebrot (#576)
Now handling trailing slash on URL, thanks to @ramiroluz (#573, #546)
Support for testing x86 and x86-64 architectures on appveyor, thanks to @maiksensi (#567)
Made tests work without installing Cookiecutter, thanks to @vincentbernat (#550)
Encoded the result of the hook template to utf8, thanks to @ionelmc (#577. #578)
Added test for _run_hook_from_repo_dir, thanks to @hackebrot (#579, #580)
Implemented bumpversion, thanks to @hackebrot (#582)
Added more cookiecutter templates to the mix:
cookiecutter-octoprint-plugin by @foosel (#560)
wagtail-cookiecutter-foundation by @chrisdev, et al. (#566)
1.2.1 (2015-10-18) Zimtsterne¶
Zimtsterne are cinnamon star cookies.
New Feature:
Returns rendered project dir, thanks to @hackebrot (#553)
Bug Fixes:
Factor in choice variables (as introduced in 1.1.0) when using a user config or extra context, thanks to @ionelmc and @hackebrot (#536, #542).
Other Changes:
Enable py35 support on Travis by using Python 3.5 as base Python (@maiksensi / #540)
If a filename is empty, do not generate. Log instead (@iljabauer / #444)
Fix tests as per last changes in cookiecutter-pypackage, thanks to @eliasdorneles(#555).
Removed deprecated cookiecutter-pylibrary-minimal from the list, thanks to @ionelmc (#556)
Moved to using rualmel.yaml instead of PyYAML, except for Windows users on Python 2.7, thanks to @pydanny (#557)
Why 1.2.1 instead of 1.2.0? There was a problem in the distribution that we pushed to PyPI. Since you can’t replace previous files uploaded to PyPI, we deleted the files on PyPI and released 1.2.1.
1.1.0 (2015-09-26) Snickerdoodle¶
The goals of this release were copy without render and a few additional command-line options such as --overwrite-if-exists, ---replay, and output-dir.
Features:
Added copy without render feature, making it much easier for developers of Ansible, Salt Stack, and other recipe-based tools to work with Cookiecutter. Thanks to @osantana and @LucianU for their innovation, as well as @hackebrot for fixing the Windows problems (#132, #184, #425).
Added specify output directory, thanks to @tony and @hackebrot (#531, #452).
Abort template rendering if the project output directory already exists, thanks to @lgp171188 (#470, #471).
Add a flag to overwrite existing output directory, thanks to @lgp171188 for the implementation (#495) and @schacki, @ionelmc, @pydanny and @hackebrot for submitting issues and code reviews (#475, #493).
Remove test command in favor of tox, thanks to @hackebrot (#480).
Allow cookiecutter invocation, even without installing it, via
python -m cookiecutter.cli, thanks to @vincentbernat and @hackebrot (#449, #487).Improve the type detection handler for online and offline repositories, thanks to @charlax (#490).
Add replay feature, thanks to @hackebrot (#501).
Be more precise when raising an error for an invalid user config file, thanks to @vaab and @hackebrot (#378, #528).
Added official Python 3.5 support, thanks to @pydanny and @hackebrot (#522).
Added support for choice variables and switch to click style prompts, thanks to @hackebrot (#441, #455).
Other Changes:
Updated click requirement to < 6.0, thanks to @pydanny (#473).
Added landscape.io flair, thanks to @michaeljoseph (#439).
Descriptions of PEP8 specifications and milestone management, thanks to @michaeljoseph (#440).
Added alternate installation options in the documentation, thanks to @pydanny (#117, #315).
The test of the which() function now tests against the date command, thanks to @vincentbernat (#446)
Ensure file handles in setup.py are closed using with statement, thanks to @svisser (#280).
Removed deprecated and fully extraneous compat.is_exe() function, thanks to @hackebrot (#485).
Disabled sudo in .travis, thanks to @hackebrot (#482).
Switched to shields.io for problematic badges, thanks to @pydanny (#491).
Added whichcraft and removed
compat.which(), thanks to @pydanny (#511).Changed to export tox environment variables to codecov, thanks to @maiksensi. (#508).
Moved to using click version command, thanks to @hackebrot (#489).
Don’t use unicode_literals to please click, thanks to @vincentbernat (#503).
Remove warning for Python 2.6 from __init__.py, thanks to @hackebrot.
Removed compat.py module, thanks to @hackebrot.
Added future to requirements, thanks to @hackebrot.
Fixed problem where expanduser does not resolve “~” correctly on windows 10 using tox, thanks to @maiksensi. (#527)
Added more cookiecutter templates to the mix:
cookiecutter-beamer by @luismartingil (#307)
cookiecutter-pytest-plugin by @pytest-dev and @hackebrot (#481)
cookiecutter-flask-foundation by @JackStouffer (#457)
cookiecutter-tryton-fulfilio by @fulfilio (#465)
cookiecutter-tapioca by @vintasoftware (#496)
cookiecutter-muffin by @drgarcia1986 (#494)
cookiecutter-django-rest by @agconti (#520)
cookiecutter-es6-boilerplate by @agconti (#521)
cookiecutter-tampermonkey by @christabor (#516)
cookiecutter-wagtail by @torchbox (#533)
1.0.0 (2015-03-13) Chocolate Chip¶
The goals of this release was to formally remove support for Python 2.6 and continue the move to using py.test.
Features:
Convert the unittest suite to py.test for the sake of comprehensibility, thanks to @hackebrot (#322, #332, #334, #336, #337, #338, #340, #341, #343, #345, #347, #351, #412, #413, #414).
Generate pytest coverage, thanks to @michaeljoseph (#326).
Documenting of Pull Request merging and HISTORY.rst maintenance, thanks to @michaeljoseph (#330).
Large expansions to the tutorials thanks to @hackebrot (#384)
Switch to using Click for command-line options, thanks to @michaeljoseph (#391, #393).
Added support for working with private repos, thanks to @marctc (#265).
Wheel configuration thanks to @michaeljoseph (#118).
Other Changes:
Formally removed support for 2.6, thanks to @pydanny (#201).
Moved to codecov for continuous integration test coverage and badges, thanks to @michaeljoseph (#71, #369).
Made JSON parsing errors easier to debug, thanks to @rsyring and @mark0978 (#355, #358, #388).
Updated to Jinja 2.7 or higher in order to control trailing new lines in templates, thanks to @sfermigier (#356).
Tweaked flake8 to ignore e731, thanks to @michaeljoseph (#390).
Fixed failing Windows tests and corrected AppVeyor badge link thanks to @msabramo (#403).
Added more Cookiecutters to the list:
0.9.0 (2015-01-13)¶
The goals of this release were to add the ability to Jinja2ify the cookiecutter.json default values, and formally launch support for Python 3.4.
Features:
Python 3.4 is now a first class citizen, thanks to everyone.
cookiecutter.json values are now rendered Jinja2 templates, thanks to @bollwyvl (#291).
Move to py.test, thanks to @pfmoore (#319) and @ramiroluz (#310).
Add PendingDeprecation warning for users of Python 2.6, as support for it is gone in Python 2.7, thanks to @michaeljoseph (#201).
Bug Fixes:
Corrected typo in Makefile, thanks to @inglesp (#297).
Raise an exception when users don’t have git or hg installed, thanks to @pydanny (#303).
Other changes:
Creation of gitter account for logged chat, thanks to @michaeljoseph.
Added ReadTheDocs badge, thanks to @michaeljoseph.
Added AppVeyor badge, thanks to @pydanny
Documentation and PyPI trove classifier updates, thanks to @thedrow (#323 and #324)
0.8.0 (2014-10-30)¶
The goal of this release was to allow for injection of extra context via the Cookiecutter API, and to fix minor bugs.
Features:
cookiecutter() now takes an optional extra_context parameter, thanks to @michaeljoseph, @fcurella, @aventurella, @emonty, @schacki, @ryanolson, @pfmoore, @pydanny, @audreyr (#260).
Context is now injected into hooks, thanks to @michaeljoseph and @dinopetrone.
Moved all Python 2/3 compatibility code into cookiecutter.compat, making the eventual move to six easier, thanks to @michaeljoseph (#60, #102).
Added cookiecutterrc defined aliases for cookiecutters, thanks to @pfmoore (#246)
Added flake8 to tox to check for pep8 violations, thanks to @natim.
Bug Fixes:
Newlines at the end of files are no longer stripped, thanks to @treyhunner (#183).
Cloning prompt suppressed by respecting the
no\_inputflag, thanks to @trustrachel (#285)With Python 3, input is no longer converted to bytes, thanks to @uranusjr (#98).
Other Changes:
Added more Cookiecutters to the list:
0.7.2 (2014-08-05)¶
The goal of this release was to fix cross-platform compatibility, primarily Windows bugs that had crept in during the addition of new features. As of this release, Windows is a first-class citizen again, now complete with continuous integration.
Bug Fixes:
Fixed the contributing file so it displays nicely in Github, thanks to @pydanny.
Updates 2.6 requirements to include simplejson, thanks to @saxix.
Avoid unwanted extra spaces in string literal, thanks to @merwok.
Fix @unittest.skipIf error on Python 2.6.
Let sphinx parse :param: properly by inserting newlines #213, thanks to @mineo.
Fixed Windows test prompt failure by replacing stdin per @cjrh in #195.
Made rmtree remove readonly files, thanks to @pfmoore.
Now using tox to run tests on Appveyor, thanks to @pfmoore (#241).
Fixed tests that assumed the system encoding was utf-8, thanks to @pfmoore (#242, #244).
Added a tox ini file that uses py.test, thanks to @pfmoore (#245).
Other Changes:
@audreyr formally accepted position as BDFL of cookiecutter.
Elevated @pydanny, @michaeljoseph, and @pfmoore to core committer status.
Added Core Committer guide, by @audreyr.
Generated apidocs from make docs, by @audreyr.
Added contributing command to the makedocs function, by @pydanny.
Refactored contributing documentation, included adding core committer instructions, by @pydanny and @audreyr.
Do not convert input prompt to bytes, thanks to @uranusjr (#192).
Added troubleshooting info about Python 3.3 tests and tox.
Added documentation about command line arguments, thanks to @saxix.
Style cleanups.
Added environment variable to disable network tests for environments without networking, thanks to @vincentbernat.
Added Appveyor support to aid Windows integrations, thanks to @pydanny (#215).
CONTRIBUTING.rst is now generated via make contributing, thanks to @pydanny (#220).
Removed unnecessary endoing argument to json.load, thanks to @pfmoore (#234).
Now generating shell hooks dynamically for Unix/Windows portability, thanks to @pfmoore (#236).
Removed non-portable assumptions about directory structure, thanks to @pfmoore (#238).
Added a note on portability to the hooks documentation, thanks to @pfmoore (#239).
Replaced unicode_open with direct use of io.open, thanks to @pfmoore (#229).
Added more Cookiecutters to the list:
0.7.1 (2014-04-26)¶
Bug fixes:
Use the current Python interpreter to run Python hooks, thanks to @coderanger.
Include tests and documentation in source distribution, thanks to @vincentbernat.
Fix various warnings and missing things in the docs (#129, #130), thanks to @nedbat.
Add command line option to get version (#89), thanks to @davedash and @cyberj.
Other changes:
Add more Cookiecutters to the list:
0.7.0 (2013-11-09)¶
This is a release with significant improvements and changes. Please read through this list before you upgrade.
New features:
Support for --checkout argument, thanks to @foobacca.
Support for pre-generate and post-generate hooks, thanks to @raphigaziano. Hooks are Python or shell scripts that run before and/or after your project is generated.
Support for absolute paths to cookiecutters, thanks to @krallin.
Support for Mercurial version control system, thanks to @pokoli.
When a cookiecutter contains invalid Jinja2 syntax, you get a better message that shows the location of the TemplateSyntaxError. Thanks to @benjixx.
Can now prompt the user to enter values during generation from a local cookiecutter, thanks to @ThomasChiroux. This is now always the default behavior. Prompts can also be suppressed with
--no-input.Your cloned cookiecutters are stored by default in your ~/.cookiecutters/ directory (or Windows equivalent). The location is configurable. (This is a major change from the pre-0.7.0 behavior, where cloned cookiecutters were deleted at the end of project generation.) Thanks @raphigaziano.
User config in a ~/.cookiecutterrc file, thanks to @raphigaziano. Configurable settings are cookiecutters_dir and default_context.
File permissions are now preserved during project generation, thanks to @benjixx.
Bug fixes:
Unicode issues with prompts and answers are fixed, thanks to @s-m-i-t-a.
The test suite now runs on Windows, which was a major effort. Thanks to @pydanny, who collaborated on this with me.
Other changes:
Quite a bit of refactoring and API changes.
Lots of documentation improvements. Thanks @sloria, @alex, @pydanny, @freakboy3742, @es128, @rolo.
Better naming and organization of test suite.
A CookiecutterCleanSystemTestCase to use for unit tests affected by the user’s config and cookiecutters directory.
Improvements to the project’s Makefile.
Improvements to tests. Thanks @gperetin, @s-m-i-t-a.
Removal of subprocess32 dependency. Now using non-context manager version of subprocess.Popen for Python 2 compatibility.
Removal of cookiecutter’s cleanup module.
A bit of setup.py cleanup, thanks to @oubiga.
Now depends on binaryornot 0.2.0.
0.6.4 (2013-08-21)¶
Windows support officially added.
Fix TemplateNotFound Exception on Windows (#37).
0.6.3 (2013-08-20)¶
Fix copying of binary files in nested paths (#41), thanks to @sloria.
0.6.2 (2013-08-19)¶
Depend on Jinja2>=2.4 instead of Jinja2==2.7.
Fix errors on attempt to render binary files. Copy them over from the project template without rendering.
Fix Python 2.6/2.7 UnicodeDecodeError when values containing Unicode chars are in cookiecutter.json.
Set encoding in Python 3 unicode_open() to always be utf-8.
0.6.1 (2013-08-12)¶
Improved project template finding. Now looks for the occurrence of {{,cookiecutter, and }} in a directory name.
Fix help message for input_dir arg at command prompt.
Minor edge cases found and corrected, as a result of improved test coverage.
0.6.0 (2013-08-08)¶
Config is now in a single
cookiecutter.jsoninstead of injson/.When you create a project from a git repo template, Cookiecutter prompts you to enter custom values for the fields defined in
cookiecutter.json.
0.5 (2013-07-28)¶
Friendlier, more simplified command line usage:
# Create project from the cookiecutter-pypackage/ template
$ cookiecutter cookiecutter-pypackage/
# Create project from the cookiecutter-pypackage.git repo template
$ cookiecutter https://github.com/audreyr/cookiecutter-pypackage.git
Can now use Cookiecutter from Python as a package:
from cookiecutter.main import cookiecutter
# Create project from the cookiecutter-pypackage/ template
cookiecutter('cookiecutter-pypackage/')
# Create project from the cookiecutter-pypackage.git repo template
cookiecutter('https://github.com/audreyr/cookiecutter-pypackage.git')
Internal refactor to remove any code that changes the working directory.
0.4 (2013-07-22)¶
Only takes in one argument now: the input directory. The output directory is generated by rendering the name of the input directory.
Output directory cannot be the same as input directory.
0.3 (2013-07-17)¶
Takes in command line args for the input and output directories.
0.2.1 (2013-07-17)¶
Minor cleanup.
0.2 (2013-07-17)¶
Bumped to “Development Status :: 3 - Alpha”.
Works with any type of text file.
Directory names and filenames can be templated.
0.1.0 (2013-07-11)¶
First release on PyPI.
Case Studies¶
This showcase is where organizations can describe how they are using Cookiecutter.
BeeWare¶
Building Python tools for platforms like mobile phones and set top boxes requires a lot of boilerplate code just to get the project running. Cookiecutter has enabled us to very quickly stub out a starter project in which running Python code can be placed, and makes maintaining those templates very easy. With Cookiecutter we’ve been able to deliver support Android devices, iOS devices, tvOS boxes, and we’re planning to add native support for iOS and Windows devices in the future.
BeeWare is an organization building open source libraries for Python support on all platforms.
ChrisDev¶
Anytime we start a new project we begin with a Cookiecutter template that generates a Django/Wagtail project Our developers like it for maintainability and our designers enjoy being able to spin up new sites using our tool chain very quickly. Cookiecutter is very useful for because it supports both Mac OSX and Windows users.
ChrisDev is a Trinidad-based consulting agency.
OpenStack¶
OpenStack uses several Cookiecutter templates to generate:
OpenStack is open source software for creating private and public clouds.
Code of Conduct¶
Everyone interacting in the Cookiecutter project’s codebases and documentation is expected to follow the PyPA Code of Conduct. This includes, but is not limited to, issue trackers, chat rooms, mailing lists, and other virtual or in real life communication.