Wiki Navigation

CherryPy Tutorial

This document is for CherryPy 3.

(Old 2.x) translations: Italian, Spanish, French

What is CherryPy?

CherryPy is a pythonic, object-oriented HTTP framework. It provides the foundation over which complex web-based applications can be written, with little or no knowledge of the underlying protocols. CherryPy allows developers to build web applications in much the same way they would build any other object-oriented Python program. This usually results in smaller source code developed in less time.

CherryPy does its best to stay out of the way between the programmer and the problem. CherryPy applications are usually very simple. It works out of the box; default behavior is sensible enough to allow use without extensive setup or customization. The embedded web server allows one to deploy web applications anywhere Python is installed. In short, CherryPy is as pythonic as it gets.

What CherryPy is NOT?

As an HTTP framework, CherryPy does all that is necessary to allow Python code to be executed when some resource (or URL) is requested by the user. However, it is not a templating language, such as PHP. CherryPy can work with several templating packages, including Cheetah, CherryTemplate, and several others. But please note that, while useful to some extent, templating packages are not strictly necessary, and that pure Python code can be used to generate the Web pages.

CherryPy includes a light embedded web server, capable enough to handle sizeable loads. It is estimated that CherryPy can handle about 530 requests/second on a reasonable configuration (as of January, 2007). However, CherryPy is not Apache. If for any reason you need a enterprise-class web server, you can run CherryPy and Apache together.

What is this tutorial about?

This tutorial cover the basic steps for a newcomer to get to grips with CherryPy unique approach to web application development. After following this tutorial, the programmer will be able to understand how CherryPy applications work, and also to implement simple but yet powerful applications on her own. Some knowledge of the Python programming language is assumed. One does not need to be an expert to work with CherryPy, but a good understanding of object-oriented basics is strongly recommended.

Knowledge required

It is assumed that the user has:

Some knowledge of the Python programming language;
Some experience with basic object oriented programming;
Some knowledge of HTML, which is necessary to build the Web pages.

Learning Python

As said above, this is not a guide to the Python language. There are plenty of good resources for those learning Python (just to name a few among the best: Python course in Bioinformatics, A Byte Of Python and Dive into Python). The official Python website lists some good resources, including an excellent tutorial.

Your first CherryPy application

The standard 'Hello world!' application takes less than 10 lines of code, when written using CherryPy.

import cherrypy

class HelloWorld:
    def index(self):
        return "Hello world!"
    index.exposed = True

cherrypy.quickstart(HelloWorld())

We assumed that you already have installed CherryPy. Copy this file and save it locally as hello.py, then start the application at the command prompt:

python hello.py

Direct your favorite web browser to http://localhost:8080 and you should see Hello world! printed there.

How does it work?

Let's take a look at the hello.py:

The import cherrypy statement imports the main CherryPy module. This is all that is required to have CherryPy working. Feel free to "import cherrypy" in an interactive session and see what's available!
We declare a class named HelloWorld. An instance of this class is the object that will be published by CherryPy. It contains a single method, named index, which will get called when the root URL for the site is asked for (for example, http://localhost/). This method returns the contents of the Web page; in this case, the 'Hello World!' string.
The index.exposed = True is a necessary step to tell CherryPy that the index() method will be exposed. Only exposed methods can be called to answer a request. This feature allows the user to select which methods of an object will be accessible via the Web; non-exposed methods can't be accessed.
cherrypy.quickstart(HelloWorld()) publishes an instance of the HelloWorld class, and starts the embedded webserver. In runs until explicitly interrupted, with Ctrl-C or via a suitable signal (a simple kill on Unix will do it).

When the application is executed, the CherryPy server is started with the default configuration. It will listen on localhost at port 8080. These defaults can be overridden by using a configuration file (more on this later).

Finally, the web server receives the request for the URL http://localhost:8080. It searches for the best method to handle the request, starting from the HelloWorld instance. In this particular case, the root of the site is automatically mapped to the index() method (similar to the index.html that is the standard page for conventional Web servers). The HelloWorld class defines an index() method and exposes it. CherryPy calls HelloWorld().index(), and the result of the call is sent back to the browser as the contents of the index page for the website. All work is done automatically; the application programmer only needs to provide the desired content as the return value of the index method.

Concepts

Publishing objects

Any object that is attached to the root object is said to be published. This means that the object is accessible via the internal URL-to-object mapping routine. However, it does not mean that the object itself is directly accessible via the Web. For this to happen, the object has to be exposed.

Exposing objects

CherryPy maps URL requests to objects and calls the suitable method automatically. The methods that can be called as a result of external requests are said to be exposed.

Objects are exposed in CherryPy by setting the exposed attribute. This can be done directly on the object itself: <object>.exposed = True. Methods can also be exposed using a special decorator:

    @cherrypy.expose
    def index(self):
        ...

Why this distinction?

The distinction between published and exposed objects may seem silly or unnecessary at first. By definition, all exposed objects are also published, so why do we need to care about non-exposed objects?

The answer is that, although only exposed objects can be accessed as a result of a external query, there are some situations where requests can be remapped to any published object, including non-exposed ones. This is an advanced feature that will come into play later. For now, all that is necessary is to understand the basic concepts.

Finding the correct object

For the user, a web application is just like a website with static files. The user types (or clicks) a URL, and gets to the desired webpage. A conventional webserver uses the URL to retrieve a static file from the filesystem. On the other hand, a web application server not only serves the content from static files; it can also map the URL it receives into some object and call it. The result is then sent back to the user's browser, where it is rendered into a viewable page. The result is a dynamic web application; for each URL, a unique object can be called into action.

The key to understand how to write a new web application is to understand how this mapping occurs. CherryPy uses a fairly straightforward mapping procedure. The root of the site is the Application.root object. When it receives a URL, it breaks it into its path components, and proceeds looking down into the site until it finds an object that is the 'best match' for that particular URL. For each path component it tries to find an object with the same name, starting from root, and going down for each component it finds, until it can't find a match. An example shows it better:

root = HelloWorld()
root.onepage = OnePage()
root.otherpage = OtherPage()

In the example above, the URL http://localhost/onepage will point at the first object and the URL http://localhost/otherpage will point at the second one. As usual, this search is done automatically. But it goes even further:

root.some = Page()
root.some.page = Page()

In this example, the URL http://localhost/some/page will be mapped to the root.some.page object. If this object is exposed (or alternatively, its index method is), it will be called for that URL.

In our HelloWorld example, adding the http://.../onepage to OnePage() mapping could be done like this:

class OnePage(object):
    def index(self):
        return "one page!"
    index.exposed = True
 
class HelloWorld(object):
    onepage = OnePage()
 
    def index(self):
        return "hello world"
    index.exposed = True
 
cherrypy.quickstart(HelloWorld())

The index method

The index() method has a special role in CherryPy. Like the index.html file, it's the default page for any internal node in the object tree. The index() method can take additional keyword arguments, which are automatically mapped to the form variables as sent via its GET or POST methods.

The index() method is only called for a full match on the URL. If the URL is a partial match (in other words, if CherryPy could not find a match for one component in the path), it will not be called.

Calling other methods

CherryPy can also directly call methods in the published objects, if it receives a URL that is directly mapped to them. For example:

def foo(self):
    return 'Foo!'
foo.exposed = True

root.foo = foo

In the example, root.foo contains a function object, named foo. When CherryPy receives a request for the /foo URL, it will automatically call the foo() function. Note that it can be a plain function, or a method of any object; any callable will do it.

In some advanced cases, there can be a conflict as CherryPy tries to decide which method it will call to handle a request. The index() method has the precedence. But if CherryPy finds a full match, and the last object in the match is a callable (which means a method, function, or any other Python object that supports the __call__ method); and finally, if the callable itself does not contain a valid index() method, then the object itself will be called. These rules are necessary because classes in Python actually are callables; calling them produces a new instance. It may look confusing, but the rules are very simple use in practice.

Receiving data from HTML forms

Any method that is called by CherryPy - index, or any other suitable method - can receive additional data from HTML forms using keyword arguments. For example, the following login form sends the username and the password as form arguments using the POST method:

<form action="doLogin" method="post">
    <p>Username</p>
    <input type="text" name="username" value="" 
        size="15" maxlength="40"/>
    <p>Password</p>
    <input type="password" name="password" value="" 
        size="10" maxlength="40"/>
    <p><input type="submit" value="Login"/></p>
    <p><input type="reset" value="Clear"/></p>
</form>

The following code can be used to handle this URL:

class Root:
    def doLogin(self, username=None, password=None):
        # check the username & password
        ...
    doLogin.exposed = True

Both arguments have to be declared as keyword arguments. The default value can be used either to provide a suitable default value for optional arguments, or to provide means for the application to detect if some values were missing from the request.

CherryPy supports both the GET and POST method for forms. Arguments are passed the same way, regardless of the original method used by the browser to send data to the web server.

Partial matches and the default method

Partial matches can happen when a URL contains components that do not map to the object tree. This can happen for a number of reasons. For example, it may be an error; the user just typed the wrong URL. But it also can mean that the URL contains extra arguments.

When a partial match happens, CherryPy calls a default method. The default method is similar to the index method; however, it is only called as a last resort method, and it's recommended for two applications:

Error handling, to be called when the user types the wrong URL;
Support for positional arguments (since CherryPy 2.2, positional arguments can be used with all methods except index).

For example, assume that you have a blog-like application written in CherryPy that takes the year, month and day as part of the URL http://localhost/blog/2005/01/17. This URL can be handled by the following code:

class Blog:
    def default(self, year, month, day):
        ...
    default.exposed = True
...
root.blog = Blog()

So the URL above will be mapped as a call to:

root.blog.default('2005', '1', '17')

In this case, there is a partial match up to the blog component. The rest of the URL can't be found in the published object tree. In this case, the default() method will be called, and the positional parameters will receive the remaining path components as arguments. The values are passed as strings; in the above mentioned example, the arguments would still need to be converted back into numbers, but the idea is correctly presented.

The CherryPy configuration file

CherryPy uses a simple configuration file format to customize some aspects of its behavior. There are actually two (or more) files, one for the global "site" and one for each "application"; but if you only have one app you can put them both in the same file. The configuration files can be edited with any conventional text editor (even Notepad will do it), and can be used even by non-technical users for some simple customization tasks. For example:

[global]
server.socket_port = 8000
server.thread_pool = 10
tools.sessions.on = True
tools.staticdir.root = "/home/site"

[/static]
tools.staticdir.on = True
tools.staticdir.dir = "static"

Many of the values are self explanatory (for example, server.socket_port, which allows changing the default port at which CherryPy listens); others need a better understanding of CherryPy internals.

The server.thread_pool option determines how many threads CherryPy starts to serve requests.
The tools.sessions.on statement enables the session functionality. Sessions are necessary to implement complex Web applications, with user identification, for example.
The [/static] statement specifies that static content from /home/site/static/* is served as /static/*
The tools.staticdir.root statement specifies the directory from which the static files are served. See StaticContent.

If you're using quickstart, you can pass a single configuration filename (or dict) containing both site and app config to cherrypy.quickstart(Root(), '/', filename_or_dict). Otherwise, you need to register global site config as cherrypy.config.update(filename_or_dict) and app config in cherrypy.tree.mount(Root(), '/', filename_or_dict). See the config docs for more information.

The cherrypy structure

Most of the features of CherryPy are available through the cherrypy module. It contains several members:

cherrypy.engine contains the API to control the CherryPy engine.
cherrypy.server contains the API to control the HTTP server.
cherrypy.request contains the all the information that comes with the HTTP request, after it is parsed and analyzed by CherryPy.
cherrypy.request.headers contains a mapping with the header options that were sent as part of the request.
cherrypy.session is a special mapping that is automatically generated and encoded by CherryPy; it can be used to store session-data in a persistent cookie. For it to work you have to enable the session functionality by setting 'tools.session.on' to True in your config.
cherrypy.response contains the data that is used to build the HTTP response.
cherrypy.response.headers contains a mapping with the header options that will be returned by the server, before the contents get sent.
cherrypy.response.body contains the actual contents of the webpage that will be sent as a response.

Tools

CherryPy core is extremely light and clean. It contains only the necessary features to support the HTTP protocol and to call the correct object for each request. Additional features can be added to it using modular tools.

A tool is an object that has a chance to work on a request as it goes through the usual CherryPy processing chain. Several tools are provided as part of the standard CherryPy library, available in cherrypy.tools. Some examples are:

tools.decode: automatically handles Unicode data on the request, converting the raw strings that are sent by the browser into native Python strings.
tools.encode: automatically converts the response from the native Python Unicode string format to some suitable encoding (Latin-1 or UTF-8, for example).
tools.gzip: Compresses the contents on the fly, using the gzip format. Saves bandwidth.
tools.xmlrpc: Implements a special XML-RPC adaptation layer over the standard CherryPy. It takes care of translating the data on request and response (a process called 'marshalling').

Tools provide a lot of flexibility. Different tools can be applied to different parts of the site, and the order of tools can be changed. The user can write custom tools for special applications, changing the behavior of CherryPy without the need to change its internals.

The tools for any part of the site are usually enabled in the configuration file:

[/]
tools.encode.on = True
tools.gzip.on = True

In this case, the application can use Unicode strings for the contents it generates; translation to utf8 will be done automatically via the encoding tool. Also, all the content will be automatically compressed with gzip, saving bandwidth.

Conclusion

This tutorial only covers the basic features of CherryPy, but it tries to present them in a way that makes it easier for the user to discover how to use them. The CherryPy distribution comes with several good tutorials; however, the best way to master CherryPy is to use it to write your own Web applications. The embedded web server makes it easy for anyone not only to try, but also to deploy local applications, or even small Internet-enabled web sites. Try it, and let us know what you did with it!

Older versions

	replace this	with this
2.2	cherrypy.quickstart(HelloWorld())	cherrypy.root = HelloWorld() cherrypy.server.start()
	tools.sessions	session_filter
	tools.staticdir	static_filter
2.1	simple_cookie	simpleCookie
	socket_port	socketPort
	thread_pool	threadPool
	session_filter	sessionFilter
	static_filter	staticFilter
	headers	headerMap
2.0	import cherrypy	from cherrypy import cpg as cherrypy

Download in other formats:

Plain Text