* Copy over and update content from the Wiki to CTFd Sphinx docs
* Closes #947
* Closes #742
selenium-screenshot-testing
Kevin Chung 2019-05-19 00:39:53 -04:00 committed by GitHub
parent 49b48a9e53
commit ac236e26ea
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
9 changed files with 778 additions and 7 deletions

View File

@ -13,3 +13,4 @@ bandit==1.5.1
flask_profiler==1.8.1 flask_profiler==1.8.1
pytest-xdist==1.28.0 pytest-xdist==1.28.0
pytest-cov==2.6.1 pytest-cov==2.6.1
sphinx_rtd_theme==0.4.3

2
docs/api.rst Normal file
View File

@ -0,0 +1,2 @@
API
===

View File

@ -20,13 +20,13 @@
# -- Project information ----------------------------------------------------- # -- Project information -----------------------------------------------------
project = u'CTFd' project = u'CTFd'
copyright = u'2019, Kevin Chung' copyright = u'2019, CTFd LLC'
author = u'Kevin Chung' author = u'Kevin Chung'
# The short X.Y version # The short X.Y version
version = u'' version = u''
# The full version, including alpha/beta/rc tags # The full version, including alpha/beta/rc tags
release = u'2.1.0' release = u'2.1.2'
# -- General configuration --------------------------------------------------- # -- General configuration ---------------------------------------------------
@ -78,7 +78,7 @@ pygments_style = None
# The theme to use for HTML and HTML Help pages. See the documentation for # The theme to use for HTML and HTML Help pages. See the documentation for
# a list of builtin themes. # a list of builtin themes.
# #
html_theme = 'alabaster' html_theme = 'sphinx_rtd_theme'
# Theme options are theme-specific and customize the look and feel of a theme # Theme options are theme-specific and customize the look and feel of a theme
# further. For a list of options available for each theme, see the # further. For a list of options available for each theme, see the
@ -154,7 +154,7 @@ man_pages = [
# dir menu entry, description, category) # dir menu entry, description, category)
texinfo_documents = [ texinfo_documents = [
(master_doc, 'CTFd', u'CTFd Documentation', (master_doc, 'CTFd', u'CTFd Documentation',
author, 'CTFd', 'One line description of project.', author, 'CTFd', 'A Capture The Flag framework focusing on ease of use and customizability.',
'Miscellaneous'), 'Miscellaneous'),
] ]

147
docs/configuration.rst Normal file
View File

@ -0,0 +1,147 @@
Configuration
=============
CTFd provides a number of configuration options which are used to configure server behavior. CTFd makes a distinction between configuration values which can be configured only with server-level access and values which can be configured by those with administrative priveleges on CTFd.
Server Level Configuration
--------------------------
Server level configuration can be modified from the ``config.py`` file in CTFd.
SECRET_KEY
~~~~~~~~~~
The secret value used to creation sessions and sign strings. This should be set to a random string. In the
interest of ease, CTFd will automatically create a secret key file for you. If you wish to add this secret key
to your instance you should hard code this value to a random static value.
You can also remove .ctfd_secret_key from the .gitignore file and commit this file into whatever repository
you are using.
http://flask.pocoo.org/docs/latest/quickstart/#sessions
DATABASE_URL
~~~~~~~~~~~~
The URI that specifies the username, password, hostname, port, and database of the server
used to hold the CTFd database.
e.g. ``mysql+pymysql://root:<YOUR_PASSWORD_HERE>@localhost/ctfd``
REDIS_URL
~~~~~~~~~
The URI to connect to a Redis server.
e.g. ``redis://user:password@localhost:6379``
http://pythonhosted.org/Flask-Caching/#configuring-flask-caching
MAILFROM_ADDR
~~~~~~~~~~~~~
The email address that emails are sent from if not overridden in the configuration panel.
MAIL_SERVER
~~~~~~~~~~~
The mail server that emails are sent from if not overriden in the configuration panel.
MAIL_PORT
~~~~~~~~~
The mail port that emails are sent from if not overriden in the configuration panel.
MAIL_USEAUTH
~~~~~~~~~~~~
Whether or not to use username and password to authenticate to the SMTP server
MAIL_USERNAME
~~~~~~~~~~~~~
The username used to authenticate to the SMTP server if MAIL_USEAUTH is defined
MAIL_PASSWORD
~~~~~~~~~~~~~
The password used to authenticate to the SMTP server if MAIL_USEAUTH is defined
MAIL_TLS
~~~~~~~~
Whether to connect to the SMTP server over TLS
MAIL_SSL
~~~~~~~~
Whether to connect to the SMTP server over SSL
MAILGUN_API_KEY
~~~~~~~~~~~~~~~
Mailgun API key to send email over Mailgun
MAILGUN_BASE_URL
~~~~~~~~~~~~~~~~
Mailgun base url to send email over Mailgun
LOG_FOLDER
~~~~~~~~~~
The location where logs are written. These are the logs for CTFd key submissions, registrations, and logins.
The default location is the CTFd/logs folder.
UPLOAD_PROVIDER
~~~~~~~~~~~~~~~
Specifies the service that CTFd should use to store files.
UPLOAD_FOLDER
~~~~~~~~~~~~~
The location where files are uploaded. The default destination is the CTFd/uploads folder.
AWS_ACCESS_KEY_ID
~~~~~~~~~~~~~~~~~
AWS access token used to authenticate to the S3 bucket.
AWS_SECRET_ACCESS_KEY
~~~~~~~~~~~~~~~~~~~~~
AWS secret token used to authenticate to the S3 bucket.
AWS_S3_BUCKET
~~~~~~~~~~~~~
The unique identifier for your S3 bucket.
AWS_S3_ENDPOINT_URL
~~~~~~~~~~~~~~~~~~~
A URL pointing to a custom S3 implementation.
REVERSE_PROXY
~~~~~~~~~~~~~
Specifies whether CTFd is behind a reverse proxy or not. Set to ``True`` if using a reverse proxy like nginx.
See `Flask documentation <https://werkzeug.palletsprojects.com/en/0.15.x/middleware/proxy_fix/#werkzeug.middleware.proxy_fix.ProxyFix.>`_ for full details.
.. Tip::
You can also specify a comma seperated set of numbers specifying the reverse proxy configuration settings. For example to configure `x_for=1, x_proto=1, x_host=1, x_port=1, x_prefix=1` specify `1,1,1,1,1`. By setting the value to ``True``, CTFd will default to the above behavior with all proxy settings set to 1.
TEMPLATES_AUTO_RELOAD
~~~~~~~~~~~~~~~~~~~~~
Specifies whether Flask should check for modifications to templates and reload them automatically.
SQLALCHEMY_TRACK_MODIFICATIONS
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Automatically disabled to suppress warnings and save memory. You should only enable this if you need it.
SWAGGER_UI
~~~~~~~~~~
Enable the Swagger UI endpoint at ``/api/v1/``
UPDATE_CHECK
~~~~~~~~~~~~
Specifies whether or not CTFd will check whether or not there is a new version of CTFd
APPLICATION_ROOT
~~~~~~~~~~~~~~~~
Specifies what path CTFd is mounted under. It can be used to run CTFd in a subdirectory.
Example: /ctfd
OAUTH_CLIENT_ID
~~~~~~~~~~~~~~~
OAUTH_CLIENT_SECRET
~~~~~~~~~~~~~~~~~~~
Application Level Configuration
-------------------------------

86
docs/contributing.rst Normal file
View File

@ -0,0 +1,86 @@
Contributing
============
Developing on CTFd
~~~~~~~~~~~~~~~~~~
Developing new code for CTFd is easy and fun! CTFd is organized into components which each serve a distinct purpose.
Core Routes/Controllers
-----------------------
The core routes are identified in blueprints in the main CTFd folder for user facing routes and in the CTFd/admin folder for the admin panel. These core routes are used to display theme content and render the main server response that the user will receive.
API Routes/Controllers
----------------------
The API routes are implemented in the ``CTFd/api`` folder as seperate blueprints for each type of resource used in CTFd. Most behavior that manipulates data should be implemented at the API level and seperated by method and resource level. The most common API methods are ``GET``, ``POST``, ``PATCH``, ``DELETE``.
Models
------
CTFd makes heavy usage of the SQLAlchemy ORM to access database contents. The core CTFd models are defined here. Plugins are however capable of adding their own models.
CTFd makes use of ``alembic`` and ``Flask-Migrate`` to perform database migrations between versions.
Schemas
-------
Schemas provide an abstraction layer on top of the database models for permissioning and filtering of data. Schemas and the API work together to make distinctions about what data to show users and what data a user can edit.
Jinja2
------
Jinja2 is used by the CTFd server to render HTML content with data. In some cases (i.e. JavaScript challenge rendering), Mozilla Nunjucks templates are used as a mostly compatible alternative. Any templates written in Nunjucks must still be compatible with Jinja2.
JavaScript & CSS
----------------
JavaScript & CSS are used to style the front end of CTFd.
Linting
~~~~~~~
Python
------
Python code in CTFd is linted with `flake8` and `black`.
Javascript & CSS
----------------
JavaScript and CSS are linted with `prettier`.
.. Tip::
The recommendation is to integrate all linters into your editor as your changes will fail to pass if the lint checks fail. See the ``Makefile`` for ``make lint``
Testing
~~~~~~~
Python
------
Python tests are run using ``pytest`` on Travis. To run the test suite you can run `make test`. By default tests run against sqlite but this can be configured by setting the `TESTING_DATABASE_URL` environment variable.
CTFd will support both Python 2 and 3 until Python 2's EOL in 2020.
Tests are run in parallel with ``pytest-xdist`` and each test is run in its own database.
Documentation
~~~~~~~~~~~~~
CTFd's documentation is written using Sphinx and hosted by `Read the Docs <https://readthedocs.org/>`_.
To build the documentation, you should go into the ``docs`` folder and run `make html`. The content output into the `docs/_build` folder will be the resulting hosted output.
Tips & Tricks
~~~~~~~~~~~~~
Typically while developing CTFd, developers use the provided ``serve.py`` script or its ``make serve`` wrapper and access CTFd at ``http://localhost:4000``.
Very often you will need to generate testing data so that you can exercise CTFd's behavior. The included ``populate.py`` script will insert randomized testing data into the CTFd database.
The ``export.py`` script can be used to create a CTFd export on the command line.
The ``import.py`` script can be used to load in a CTFd export on the command line.
If you need to wipe CTFd completely, you should:
* delete the database (CTFd/ctfd.db by default)
* empty the cache. By default it will be stored in the ``.data`` folder if Redis is unavailable
* (optional) remove the contents of the `CTFd/uploads` folder

108
docs/deployment.rst Normal file
View File

@ -0,0 +1,108 @@
Deployment
==========
CTFd is a standard WSGI application so most if not all `Flask documentation`_ on deploying a Flask based application should apply. This page will focus on the recommended ways to deploy CTFd.
.. Important::
Fully managed and maintained CTFd deployments are available at https://ctfd.io.
Docker
------
CTFd provides automatically generated `Docker images`_ and one of the simplest means of deploying a CTFd instance is to use Docker Compose.
.. Caution:: While Docker can be a very simple means of deploying CTFd, it can make debugging and receiving support more complicated. Before deploying using Docker, be sure you have a cursory understanding of how Docker works.
1. Install `Docker`_
2. Install `Docker Compose`_
3. Clone the CTFd repository with ``git clone https://github.com/CTFd/CTFd.git``
4. Modify the ``docker-compose.yml`` file from the repository to specify a ``SECRET_KEY`` environment for the CTFd service. ::
environment:
- SECRET_KEY=<SPECIFY_RANDOM_VALUE>
- UPLOAD_FOLDER=/var/uploads
- LOG_FOLDER=/var/log/CTFd
- DATABASE_URL=mysql+pymysql://root:ctfd@db/ctfd
- REDIS_URL=redis://cache:6379
- WORKERS=4
.. Tip::
You can also run ``python -c "import os; f=open('.ctfd_secret_key', 'a+'); f.write(os.urandom(64)); f.close()"`` within the CTFd repo to generate a .ctfd_secret_key file.
5. Run ``docker-compose up``
6. You should now be able to access CTFd at http://localhost:8000
.. Note::
The default Docker Compose configuration files do not provide a reverse proxy server or configure SSL/TLS. This is left as an exercise to the reader by design.
Standard WSGI Deployment
------------------------
As a web application, CTFd has a few dependencies which require installation.
1. WSGI server
2. Database server
3. Caching server
WSGI Server
~~~~~~~~~~~
While CTFd is a standard WSGI application and most WSGI servers (e.g. gunicorn, UWSGI, waitress) will likely suffice, CTFd is most commonly deployed with gunicorn. This is because of its simplicity and reasonable performance. It is installed by default and used by other deployment methods discussed on this page. By default it is configured to use gevent as a worker class.
Database Server
~~~~~~~~~~~~~~~
CTFd makes use of SQLAlchemy and as such supports a number of SQL databases. As of CTFd 2.0, the recommended database type is MySQL. CTFd is tested and has been installed against SQLite, Postgres, and MariaDB but this could change in the future.
By default CTFd will create a SQLite database if no database server has been configured.
.. Note::
CTFd makes use of the JSON data type. MySQL >= 5.7.8 implements a proper JSON type while MariaDB does not. Small differences like these could eventually result in CTFd only supporting a few database servers.
Caching Server
~~~~~~~~~~~~~~
CTFd makes heavy use of caching servers to store configuration values, user sessions, and page content. It is important to deploy CTFd with a caching server. The preferred caching server option is Redis.
By default if no cache server is configured, CTFd will attempt to use the filesystem as a cache and store values in the ``.data`` folder. This type of caching is not very performant thus it is highly recommended that you configure a Redis server.
Vagrant
-------
CTFd provides a basic Vagrantfile for use with Vagrant. To run using Vagrant run the following commands:
::
vagrant up
Visit http://localhost:8000 where CTFd will be running.
To access the internal gunicorn terminal session inside Vagrant run:
::
vagrant ssh
tmux attach ctfd
.. Note::
CTFd's Vagrantfile is not commonly used and is only community supported
Debug Server
------------
The absolute simplest way to deploy CTFd merely involves running `python serve.py` to start Flask's built-in debugging server. This isn't recommended for anything but debugging and should not be used for any kind of load. It is discussed here because the debugging server can make identifying bugs and misconfigurations easier. In addition, development mostly occurs using the debug server.
.. Important::
CTFd makes every effort to be an easy to setup application.
However, deploying CTFd for large amounts of users can be difficult.
Fully managed and maintained CTFd deployments are available at https://ctfd.io. If you're interested in a specialized CTFd deployment with custom features please `contact us <https://ctfd.io/contact/>`_.
.. _Flask documentation: http://flask.pocoo.org/docs/latest/deploying/
.. _Docker images: https://hub.docker.com/r/ctfd/ctfd/
.. _Docker: https://docs.docker.com/install/
.. _Docker Compose: https://docs.docker.com/compose/install/
.. _contact us: https://ctfd.io/contact/

View File

@ -3,14 +3,29 @@
You can adapt this file completely to your liking, but it should at least You can adapt this file completely to your liking, but it should at least
contain the root `toctree` directive. contain the root `toctree` directive.
Welcome to CTFd's documentation! CTFd
================================ ====
CTFd is a Capture The Flag framework focusing on ease of use and customizability.
CTFd comes with most of the features needed by an event organizer to run a competition or workshop. In addition, if CTFd's feature set is insufficient, CTFd allows for the usage of plugins and themes to control almost every aspect of how it looks and behaves.
CTFd is used by many different clubs, universities, and companies to host their own Capture The Flags.
While available as open source, CTFd developers also provide a managed hosting service available at https://ctfd.io/.
CTFd is written in Python and makes use of the Flask web framework.
.. toctree:: .. toctree::
:maxdepth: 2 :maxdepth: 2
:caption: Contents: :caption: Contents:
deployment
configuration
themes
plugins
contributing
Indices and tables Indices and tables
================== ==================

402
docs/plugins.rst Normal file
View File

@ -0,0 +1,402 @@
Plugins
=======
Introduction
------------
CTFd features a plugin interface allowing for the modification of CTFd behavior without modifying the core CTFd code. This has a number of benefits over forking and modifying CTFd:
* Your modifications and plugins can be shared more easily
* CTFd can be updated without losing any custom behavior
The CTFd developers will do their best to not introduce breaking changes but keep in mind that the plugin interface is still under development and could change.
.. Tip::
Official CTFd plugins are available at https://ctfd.io/store. `Contact us <https://ctfd.io/contact/>`_ regarding custom plugins and special projects.
.. Tip::
Community themes are available at https://github.com/CTFd/plugins.
Architecture
------------
CTFd plugins are implemented as Python modules with some CTFd specific files.
::
CTFd
└── plugins
   └── CTFd-plugin
      ├── README.md # README file
      ├── __init__.py # Main code file loaded by CTFd
      ├── requirements.txt # Any requirements that need to be installed
      └── config.json # Plugin configuration file
Effectively CTFd will look at every folder in the ``CTFd/plugins`` folder for the ``load()`` function.
If the ``load()`` function is found, CTFd will call that function with itself (as a Flask app) as a parameter (i.e. ``load(app)``). This is done after CTFd has added all of its internal routes but before CTFd has fully instantiated itself. This allows plugins to modify many aspects of CTFd without having to modify CTFd itself.
config.json
~~~~~~~~~~~
``config.json`` exists to give plugin developers a way to define attributes about their plugin. It's primary usage within CTFd is to give users a way to access a Configuration or Settings page for the plugin.
This is an example ``config.json`` file:
.. code-block:: json
{
"name": "CTFd Plugin",
"route": "/admin/custom_plugin_route"
}
This is ultimately rendered to the user with the following template snippet:
.. code-block:: html
{% if plugins %}
<li>
<a href="#" class="dropdown-toggle" data-toggle="dropdown" role="button" aria-haspopup="true" aria-expanded="false">Plugins <span class="caret"></span></a>
<ul class="dropdown-menu">
{% for plugin in plugins %}
<li><a href="{{ request.script_root }}{{ plugin.route }}">{{ plugin.name }}</a></li>
{% endfor %}
</ul>
</li>
{% endif %}
config.html
~~~~~~~~~~~
In the past CTFd used a static file known as ``config.html`` which existed to give plugin developers a page that is loaded by the CTFd admin panel. This has been superceded in favor of `config.json` but is still supported for backwards compatability.
The ``config.html`` file for a plugin is available by CTFd admins at ``/admin/plugins/<plugin-folder-name>``. Thus if ``config.html`` is stored in ``CTFd-S3-plugin``, it would be available at ``/admin/plugins/CTFd-S3-plugin``.
``config.html`` is loaded as a Jinja template so it has access to all of the same functions and abilities that CTFd exposes to Jinja. Jinja templates are technically also capable of running arbitrary Python code but this is ancillary.
Adding New Routes
-----------------
Adding new routes in CTFd is effectively just an exercise in writing new Flask routes. Since the plugin itself is passed the entire app, the plugin can leverage the ``app.route`` decorator to add new routes.
A simple example is as follows:
.. code-block:: python
from flask import render_template
def load(app):
@app.route('/faq', methods=['GET'])
def view_faq():
return render_template('page.html', content="<h1>FAQ Page</h1>")
Modifying Existing Routes
-------------------------
It is slightly more complicated to override existing routes in CTFd/Flask because it is not strictly supported by Flask. The approach currently used is to modify the ``app.view_functions`` dictionary which contains the mapping of routes to the functions used to handle them.
.. code-block:: python
from flask import render_template
from CTFd.models import db
from CTFd.utils import admins_only, is_admin
from CTFd import utils
def load(app):
def view_challenges():
return render_template('page.html', content="<h1>Challenges are currently closed</h1>")
# The format used by the view_functions dictionary is blueprint.view_function_name
app.view_functions['challenges.challenges_view'] = view_challenges
If for some reason you wish to add a new method to an existing route you can modify the ``url_map`` as follows:
.. code-block:: python
from werkzeug.routing import Rule
app.url_map.add(Rule('/challenges', endpoint='challenges.challenges_view', methods=['GET', 'POST']))
Adding Database Tables
----------------------
Sometimes CTFd doesn't have enough database tables or columns to let you do what you need. In this case you can use a plugin to create a new table and then use the information in the previous two sections to create routes or modify existing routes to access your new table.
.. code-block:: python
from CTFd.models import db
class Avatars(db.Model):
id = db.Column(db.Integer, primary_key=True)
team = db.Column(db.Integer, db.ForeignKey('teams.id'))
location = db.Column(db.Text)
def __init__(self, team, location):
self.target = team
self.location = location
def load(app):
app.db.create_all()
@app.route('/profile/avatar', methods=['GET', 'POST'])
def profile_avatars():
raise NotImplementedError
Replacing Templates
-------------------
In some situations it might make sense for your plugin to replace the logic for a single page template instead of creating an entire theme.
The ``utils.override_template()`` function allows a plugin to replace the content of a single template within CTFd such that CTFd will use the new content instead of the content in the original file.
.. code-block:: python
from CTFd.utils import override_template
import os
def load(app):
dir_path = os.path.dirname(os.path.realpath(__file__))
template_path = os.path.join(dir_path, 'new-scoreboard.html')
override_template('scoreboard.html', open(template_path).read())
With this code CTFd will use ``new-scoreboard.html`` instead of the ``scoreboard.html`` file it normally would have used.
Registering Assets
------------------
Very often you will want to provide users with static assets (e.g. JS, CSS). Instead of registering handlers for them on your own, you can use the CTFd built in plugin utilities, ``register_plugin_assets_directory`` and ``register_plugin_asset``.
For example to register an entire assets directory as available to the user:
.. code-block:: python
from CTFd.plugins import register_plugin_assets_directory
def load(app):
# Available at http://ctfd/plugins/test_plugin/assets/
register_plugin_assets_directory(app, base_path='/plugins/test_plugin/assets/')
Or to only provide a single file:
.. code-block:: python
from CTFd.plugins import register_plugin_asset
def load(app):
# Available at http://ctfd/plugins/test_plugin/assets/file.js
register_plugin_asset(app, asset_path='/plugins/test_plugin/assets/file.js')
Challenge Types
---------------
In CTFd, there is a concept of a type of challenge. Most CTFs only ever provide challenges as a snippet of text alongside some files. CTFd expands upon this and allows developers to create new challenge types which diversify what users will see.
Ultimately, users will still read some text, and submit some value but CTFd allows you to style and customize this so users can submit data in new ways.
For example, instead of an input to submit a single flag value, you might require teams to submit multiple flags or you might create some kind of customized UI where teams need to arrange blocks or text in some order.
The approach used by CTFd here is to give each "type" of challenge an ID and a name.
.. Tip::
You can see how CTFd implements its `default standard challenge here <https://github.com/CTFd/CTFd/blob/master/CTFd/plugins/challenges/__init__.py>`_. You can also see how CTFd implements `dynamic scoring using this feature <https://github.com/CTFd/CTFd/tree/master/CTFd/plugins/dynamic_challenges>`_.
Each challenge is implemented as a child class of the ``BaseChallenge`` and implements static methods named ``create``, ``read``, ``update``, ``delete``, ``attempt``, ``solve``, and ``fail``.
When a user attempts to solve a challenge, CTFd will look up the challenge type and then call the ``solve`` method as shown in the following snippet of code:
.. code-block:: python
chal_class = get_chal_class(chal.type)
status, message = chal_class.attempt(chal, request)
if status: # The challenge plugin says the input is right
if ctftime() or is_admin():
chal_class.solve(team=team, chal=chal, request=request)
return jsonify({'status': 1, 'message': message})
else: # The challenge plugin says the input is wrong
if ctftime() or is_admin():
chal_class.fail(team=team, chal=chal, request=request)
This structure allows each Challenge Type to dictate how they are attempted, solved, and marked incorrect.
The Challenge Type also dictates the database table that it uses to store data. By default this uses the ``type`` column as a ``polymorphic_identity`` to implement `table inheritance <http://docs.sqlalchemy.org/en/latest/orm/inheritance.html#joined-table-inheritance>`_. Effectively each child table will use the Challenges table as a parent. The child table can add whatever columns it wishes but still leverage the existing columns from the parent.
We can see in the following code that the polymorphic_identity is specified to be ``dynamic`` as well as the ``type`` parameter. We can also see the call to ``create_all()`` which will create the table in our database.
.. code-block:: python
class DynamicChallenge(Challenges):
__mapper_args__ = {'polymorphic_identity': 'dynamic'}
id = db.Column(None, db.ForeignKey('challenges.id'), primary_key=True)
initial = db.Column(db.Integer)
minimum = db.Column(db.Integer)
decay = db.Column(db.Integer)
def __init__(self, name, description, value, category, type='dynamic', minimum=1, decay=50):
self.name = name
self.description = description
self.value = value
self.initial = value
self.category = category
self.type = type
self.minimum = minimum
self.decay = decay
def load(app):
app.db.create_all()
CHALLENGE_CLASSES['dynamic'] = DynamicValueChallenge
register_plugin_assets_directory(app, base_path='/plugins/DynamicValueChallenge/assets/')
This code creates the necessary tables for the Challenge Type plugin which should be used in addition to the staticmethods used to define the challenge's behavior.
Every challenge type must be added to the global dictionary that specifies all challenge types:
.. code-block:: python
CHALLENGE_CLASSES = {
"standard": CTFdStandardChallenge
}
def get_chal_class(class_id):
cls = CHALLENGE_CLASSES.get(class_id)
if cls is None:
raise KeyError
return cls
The `Standard Challenge type <https://github.com/CTFd/CTFd/tree/master/CTFd/plugins/challenges>`_ provided within CTFd can be used as a base from which to build additional Challenge Type plugins.
Once new challenges are registered, CTFd will provide a dropdown allowing you to choose from all the challenge types you can create.
Each Challenge Type contains templates and scripts dictionaries which contain the routes for HTML and JS files needed for the operation of the modals used to create and update the challenges.
**These routes are not automatically defined by CTFd.**
Each challenge type plugin specifies the location of their own templates and scripts. An example is the built in `standard challenge type plugin <https://github.com/CTFd/CTFd/blob/master/CTFd/plugins/challenges/__init__.py>`_. It specifies the URLs that the assets are located at for the user's browser to load:
.. code-block:: python
templates = { # Templates used for each aspect of challenge editing & viewing
'create': '/plugins/challenges/assets/create.html',
'update': '/plugins/challenges/assets/update.html',
'view': '/plugins/challenges/assets/view.html',
}
scripts = { # Scripts that are loaded when a template is loaded
'create': '/plugins/challenges/assets/create.js',
'update': '/plugins/challenges/assets/update.js',
'view': '/plugins/challenges/assets/view.js',
}
These files are registered with Flask with the following code:
.. code-block:: python
from CTFd.plugins import register_plugin_assets_directory
def load(app):
register_plugin_assets_directory(app, base_path='/plugins/challenges/assets/')
The aforementioned code handles the Python logic around new challenges but in order to fully integrate with CTFd you will need to create new Nunjucks templates to give admins/teams the ability to modify/update/solve your challenge. The `templates used by the Standard Challenge Type <https://github.com/CTFd/CTFd/tree/master/CTFd/plugins/challenges/assets>`_ should serve as examples.
Flag Types
----------
Flag types conversely are used to give developers a way to allow teams to submit flags which do not conform to a hardcoded string or a regex-able value.
The approach is very similar to Challenges with a base Flag/Key class and a global dictionary specifying all the Flag/Key types:
.. code-block:: python
class BaseFlag(object):
name = None
templates = {}
@staticmethod
def compare(self, saved, provided):
return True
class CTFdStaticFlag(BaseFlag):
name = "static"
templates = { # Nunjucks templates used for key editing & viewing
"create": "/plugins/flags/assets/static/create.html",
"update": "/plugins/flags/assets/static/edit.html",
}
@staticmethod
def compare(chal_key_obj, provided):
saved = chal_key_obj.content
data = chal_key_obj.data
if len(saved) != len(provided):
return False
result = 0
if data == "case_insensitive":
for x, y in zip(saved.lower(), provided.lower()):
result |= ord(x) ^ ord(y)
else:
for x, y in zip(saved, provided):
result |= ord(x) ^ ord(y)
return result == 0
class CTFdRegexFlag(BaseFlag):
name = "regex"
templates = { # Nunjucks templates used for key editing & viewing
"create": "/plugins/flags/assets/regex/create.html",
"update": "/plugins/flags/assets/regex/edit.html",
}
@staticmethod
def compare(chal_key_obj, provided):
saved = chal_key_obj.content
data = chal_key_obj.data
if data == "case_insensitive":
res = re.match(saved, provided, re.IGNORECASE)
else:
res = re.match(saved, provided)
return res and res.group() == provided
FLAG_CLASSES = {"static": CTFdStaticFlag, "regex": CTFdRegexFlag}
def get_flag_class(class_id):
cls = FLAG_CLASSES.get(class_id)
if cls is None:
raise KeyError
return cls
When a challenge solution is submitted, the challenge plugin itself is responsible for:
1. Loading the appropriate Key class using the ``get_flag_class()`` function.
2. Properly calling the static ``compare()`` method defined by each Flag class.
3. Returning the correctness boolean and the message displayed to the user.
This is properly implemented by the following code `copied from the default standard challenge <https://github.com/CTFd/CTFd/blob/master/CTFd/plugins/challenges/__init__.py#L136>`_:
.. code-block:: python
@staticmethod
def attempt(challenge, request):
data = request.form or request.get_json()
submission = data['submission'].strip()
flags = Flags.query.filter_by(challenge_id=challenge.id).all()
for flag in flags:
if get_flag_class(flag.type).compare(flag, submission):
return True, 'Correct'
return False, 'Incorrect'

10
docs/themes.rst Normal file
View File

@ -0,0 +1,10 @@
Themes
======
CTFds allows organizers to customize their CTFd instance with custom themes. The CTFd core routes will load templates from the theme folder specified by the theme configuration value. This value can be configured in the admin panel.
.. Tip::
Official CTFd themes are available at https://ctfd.io/store. `Contact us <https://ctfd.io/contact/>`_ regarding custom themes and special projects.
.. Tip::
Community themes are available at https://github.com/CTFd/themes.