Файл: pyramid.pdf

COPYRIGHT, TRADEMARKS, AND ATTRIBUTIONS

The Pyramid Web Application Development Framework, Version 1.1

by Chris McDonough

Copyright © 2008-2011, Agendaless Consulting.

ISBN-10: 0615445675

ISBN-13: 978-0615445670

First print publishing: February, 2011

All rights reserved. This material may be copied or distributed only subject to the terms and conditions set forth in the Creative Commons Attribution-Noncommercial-Share Alike 3.0 United States License. You must give the original author credit. You may not use this work for commercial purposes. If you alter, transform, or build upon this work, you may distribute the resulting work only under the same or similar license to this one.

latex-note.png

While the Pyramid documentation is offered under the Creative Commons Attribution-Nonconmmercial-Share Alike 3.0 United States License, the Pyramid software is offered under a less restrictive (BSD-like) license .

All terms mentioned in this book that are known to be trademarks or service marks have been appropriately capitalized. However, use of a term in this book should not be regarded as affecting the validity of any trademark or service mark.

Every effort has been made to make this book as complete and as accurate as possible, but no warranty or fitness is implied. The information provided is on as “as-is” basis. The author and the publisher shall have neither liability nor responsibility to any person or entity with respect to any loss or damages arising from the information contained in this book. No patent liability is assumed with respect to the use of the information contained herein.

iii

Attributions

Editor: Casey Duncan

Contributors: Ben Bangert, Blaise Laflamme, Rob Miller, Mike Orr, Carlos de la Guardia, Paul Everitt, Tres Seaver, John Shipman, Marius Gedminas, Chris Rossi, Joachim Krebs, Xavier Spriet, Reed O’Brien, William Chambers, Charlie Choiniere, Jamaludin Ahmad, Graham Higgins, Patricio Paez, Michael Merickel, Eric Ongerth, Niall O’Higgins, Christoph Zwerschke, John Anderson, Atsushi Odagiri, Kirk Strauser, JD Navarro, Joe Dallago, Savoir-Faire Linux, Łukasz Fidosz, Christopher Lambacher, Claus Conrad, Chris Beelby, Phil Jenvey and a number of people with only pseudonyms on GitHub.

Cover Designer: Hugues Laflamme of Kemeneur.

Used with permission:

The Request and Response Objects chapter is adapted, with permission, from documentation originally written by Ian Bicking.

The Much Ado About Traversal chapter is adapted, with permission, from an article written by Rob Miller.

The Logging is adapted, with permission, from the Pylons documentation logging chapter, originally written by Phil Jenvey.

Print Production

The print version of this book was produced using the Sphinx documentation generation system and the LaTeX typesetting system.

Contacting The Publisher

Please send documentation licensing inquiries, translation inquiries, and other business communications to Agendaless Consulting. Please send software and other technical queries to the Pylons-devel maillist.

HTML Version and Source Code

An HTML version of this book is freely available via http://docs.pylonsproject.org

The source code for the examples used in this book are available within the Pyramid software distribution, always available via https://github.com/Pylons/pyramid

iv

TYPOGRAPHICAL CONVENTIONS

Literals, filenames and function arguments are presented using the following style:

argument1

Warnings, which represent limitations and need-to-know information related to a topic or concept are presented in the following style:

latex-warning.png

This is a warning.

Notes, which represent additional information related to a topic or concept are presented in the following style:

latex-note.png

This is a note.

We present Python method names using the following style:

pyramid.config.Configurator.add_view()

We present Python class names, module names, attributes and global variables using the following style:

pyramid.config.Configurator.registry

References to glossary terms are presented using the following style:

Pylons

v

URLs are presented using the following style:

Pylons

References to sections and chapters are presented using the following style:

Traversal

Code and configuration file blocks are presented in the following style:

Example blocks representing UNIX shell commands are prefixed with a $ character, e.g.:

$ ../bin/nosetests

Example blocks representing Windows cmd.exe commands are prefixed with a drive letter and/or a directory name, e.g.:

c:\examples> ..\Scripts\nosetests

Sometimes, when it’s unknown which directory is current, Windows cmd.exe example block commands are prefixed only with a > character, e.g.:

> ..\Scripts\nosetests

When a command that should be typed on one line is too long to fit on a page, the backslash \ is used to indicate that the following printed line should actually be part of the command:

c:\bigfntut\tutorial> ..\Scripts\nosetests --cover-package=tutorial \ --cover-erase --with-coverage

A sidebar, which presents a concept tangentially related to content discussed on a page, is rendered like so:

This is a sidebar

Sidebar information.

vi

AUTHOR INTRODUCTION

Welcome to “The Pyramid Web Application Framework”. In this introduction, I’ll describe the audience for this book, I’ll describe the book content, I’ll provide some context regarding the genesis of Pyramid, and I’ll thank some important people.

I hope you enjoy both this book and the software it documents. I’ve had a blast writing both.

Audience

This book is aimed primarily at a reader that has the following attributes:

•At least a moderate amount of Python experience.

•A familiarity with web protocols such as HTTP and CGI.

If you fit into both of these categories, you’re in the direct target audience for this book. But don’t worry, even if you have no experience with Python or the web, both are easy to pick up “on the fly”.

Python is an excellent language in which to write applications; becoming productive in Python is almost mind-blowingly easy. If you already have experience in another language such as Java, Visual Basic, Perl, Ruby, or even C/C++, learning Python will be a snap; it should take you no longer than a couple of days to become modestly productive. If you don’t have previous programming experience, it will be slightly harder, and it will take a little longer, but you’d be hard-pressed to find a better “first language.”

Web technology familiarity is assumed in various places within the book. For example, the book doesn’t try to define common web-related concepts like “URL” or “query string.” Likewise, the book describes various interactions in terms of the HTTP protocol, but it does not describe how the HTTP protocol works in detail. Like any good web framework, though, Pyramid shields you from needing to know most of the gory details of web protocols and low-level data structures. As a result, you can usually avoid becoming “blocked” while you read this book even if you don’t yet deeply understand web technologies.

vii

Book Content

This book is divided into three major parts:

Narrative Documentation

This is documentation which describes Pyramid concepts in narrative form, written in a largely conversational tone. Each narrative documentation chapter describes an isolated Pyramid concept. You should be able to get useful information out of the narrative chapters if you read them out-of-order, or when you need only a reminder about a particular topic while you’re developing an application.

Tutorials

Each tutorial builds a sample application or implements a set of concepts with a sample; it then describes the application or concepts in terms of the sample. You should read the tutorials if you want a guided tour of Pyramid.

API Reference

Comprehensive reference material for every public API exposed by Pyramid. The API documentation is organized alphabetically by module name.

The Genesis of repoze.bfg

Before the end of 2010, Pyramid was known as repoze.bfg.

I wrote repoze.bfg after many years of writing applications using Zope. Zope provided me with a lot of mileage: it wasn’t until almost a decade of successfully creating applications using it that I decided to write a different web framework. Although repoze.bfg takes inspiration from a variety of web frameworks, it owes more of its core design to Zope than any other.

The Repoze “brand” existed before repoze.bfg was created. One of the first packages developed as part of the Repoze brand was a package named repoze.zope2. This was a package that allowed Zope 2 applications to run under a WSGI server without modification. Zope 2 did not have reasonable WSGI support at the time.

During the development of the repoze.zope2 package, I found that replicating the Zope 2 “publisher”

– the machinery that maps URLs to code – was time-consuming and fiddly. Zope 2 had evolved over many years, and emulating all of its edge cases was extremely difficult. I finished the repoze.zope2

viii

package, and it emulates the normal Zope 2 publisher pretty well. But during its development, it became clear that Zope 2 had simply begun to exceed my tolerance for complexity, and I began to look around for simpler options.

I considered using the Zope 3 application server machinery, but it turned out that it had become more indirect than the Zope 2 machinery it aimed to replace, which didn’t fulfill the goal of simplification. I also considered using Django and Pylons, but neither of those frameworks offer much along the axes of traversal, contextual declarative security, or application extensibility; these were features I had become accustomed to as a Zope developer.

I decided that in the long term, creating a simpler framework that retained features I had become accustomed to when developing Zope applications was a more reasonable idea than continuing to use any Zope publisher or living with the limitations and unfamiliarities of a different framework. The result is what is now Pyramid.

The Genesis of Pyramid

What was repoze.bfg has become Pyramid as the result of a coalition built between the Repoze and Pylons community throughout the year 2010. By merging technology, we’re able to reduce duplication of effort, and take advantage of more of each others’ technology.

Thanks

This book is dedicated to my grandmother, who gave me my first typewriter (a Royal), and my mother, who bought me my first computer (a VIC-20).

Thanks to the following people for providing expertise, resources, and software. Without the help of these folks, neither this book nor the software which it details would exist: Paul Everitt, Tres Seaver, Andrew Sawyers, Malthe Borch, Carlos de la Guardia, Chris Rossi, Shane Hathaway, Daniel Holth, Wichert Akkerman, Georg Brandl, Blaise Laflamme, Ben Bangert, Casey Duncan, Hugues Laflamme, Mike Orr, John Shipman, Chris Beelby, Patricio Paez, Simon Oram, Nat Hardwick, Ian Bicking, Jim Fulton, Michael Merickel, Tom Moroz of the Open Society Institute, and Todd Koym of Environmental Health Sciences.

Thanks to Guido van Rossum and Tim Peters for Python.

Special thanks to Tricia for putting up with me.

ix