2009-10-24 05:01:49 +08:00
< HTML xmlns:rdf = "http://www.w3.org/1999/02/22-rdf-syntax-ns#" xmlns:dc = "http://purl.org/dc/elements/1.1/" xmlns:dcq = "http://purl.org/dc/qualifiers/1.0/" xmlns:fo = "http://www.w3.org/1999/XSL/Format" xmlns:fn = "http://www.w3.org/2005/xpath-functions" >
2009-07-24 04:09:56 +08:00
< HEAD >
< meta http-equiv = "Content-Type" content = "text/html; charset=UTF-8" >
< TITLE > Google Python Style Guide< / TITLE >
< LINK HREF = "http://www.google.com/favicon.ico" type = "image/x-icon" rel = "shortcut icon" >
< LINK HREF = "styleguide.css" type = "text/css" rel = "stylesheet" >
< SCRIPT language = "javascript" type = "text/javascript" >
function ShowHideByName(bodyName, buttonName) {
var bodyElements;
if (document.getElementsByName) {
bodyElements = document.getElementsByName(bodyName);
} else {
bodyElements = [document.getElementById(bodyName)];
}
if (bodyElements.length != 1) {
alert("ShowHideByName() got the wrong number of bodyElements: " + bodyElements.length);
} else {
var bodyElement = bodyElements[0];
var buttonElement;
if (document.getElementsByName) {
var buttonElements = document.getElementsByName(buttonName);
buttonElement = buttonElements[0];
} else {
buttonElement = document.getElementById(buttonName);
}
if (bodyElement.style.display == "none" || bodyElement.style.display == "") {
bodyElement.style.display = "inline";
buttonElement.innerHTML = '▽';
} else {
bodyElement.style.display = "none";
buttonElement.innerHTML = '▶';
}
}
}
function ShowHideAll() {
var allButton;
if (document.getElementsByName) {
var allButtons = document.getElementsByName("show_hide_all_button");
allButton = allButtons[0];
} else {
allButton = document.getElementById("show_hide_all_button");
}
if (allButton.innerHTML == '▽') {
allButton.innerHTML = '▶';
SetHiddenState(document.getElementsByTagName("body")[0].childNodes, "none", '▶');
} else {
allButton.innerHTML = '▽';
SetHiddenState(document.getElementsByTagName("body")[0].childNodes, "inline", '▽');
}
}
// Recursively sets state of all children
// of a particular node.
function SetHiddenState(root, newState, newButton) {
for (var i = 0; i != root.length; i++) {
SetHiddenState(root[i].childNodes, newState, newButton);
if (root[i].className == 'showhide_button') {
root[i].innerHTML = newButton;
}
if (root[i].className == 'stylepoint_body') {
root[i].style.display = newState;
}
}
}
window.onload = function() {
// if the URL contains "?showall=y", expand the details of all children
{
var showHideAllRegex = new RegExp("[\\?&](showall)=([^&#]*)");
var showHideAllValue = showHideAllRegex.exec(window.location.href);
if (showHideAllValue != null) {
if (showHideAllValue[2] == "y") {
SetHiddenState(document.getElementsByTagName("body")[0].childNodes, "inline", '▽');
} else {
SetHiddenState(document.getElementsByTagName("body")[0].childNodes, "none", '▶');
}
}
var showOneRegex = new RegExp("[\\?&](showone)=([^&#]*)");
var showOneValue = showOneRegex.exec(window.location.href);
if (showOneValue != null) {
var body_name = showOneValue[2] + '__body';
var button_name = showOneValue[2] + '__button';
ShowHideByName(body_name, button_name);
}
}
}
< / SCRIPT >
< / HEAD >
< BODY >
< H1 > Google Python Style Guide< / H1 >
< p align = "right" >
2009-10-24 05:01:49 +08:00
Revision 2.14
2009-07-24 04:09:56 +08:00
< / p >
< address >
Amit Patel< br >
Antoine Picard< br >
Eugene Jhong< br >
Jeremy Hylton< br >
Matt Smart< br >
Mike Shields< br >
< / address >
< DIV style = "margin-left: 50%; font-size: 75%;" >
< P >
Each style point has a summary for which additional information is available
by toggling the accompanying arrow button that looks this way:
< SPAN class = "showhide_button" style = "margin-left: 0;" > ▶< / SPAN > .
You may toggle all summaries with the big arrow button:
< / P >
< DIV style = " font-size: larger; margin-left: +2em;" >
< SPAN class = "showhide_button" style = "font-size: 180%;" onclick = "javascript:ShowHideAll()" name = "show_hide_all_button" id = "show_hide_all_button" > ▶< / SPAN >
Toggle all summaries
< / DIV >
< / DIV >
< DIV class = "toc" >
< DIV class = "toc_title" > Table of Contents< / DIV >
< TABLE >
< TR valign = "top" class = "" >
< TD > < DIV class = "toc_category" > < A href = "#Python_Language_Rules" > Python Language Rules< / A > < / DIV > < / TD >
< TD > < DIV class = "toc_stylepoint" >
2009-10-24 05:01:49 +08:00
< SPAN style = "padding-right: 1em; white-space:nowrap;" class = "" > < A href = "#pychecker" > pychecker< / A > < / SPAN > < SPAN style = "padding-right: 1em; white-space:nowrap;" class = "" > < A href = "#Imports" > Imports< / A > < / SPAN > < SPAN style = "padding-right: 1em; white-space:nowrap;" class = "" > < A href = "#Packages" > Packages< / A > < / SPAN > < SPAN style = "padding-right: 1em; white-space:nowrap;" class = "" > < A href = "#Exceptions" > Exceptions< / A > < / SPAN > < SPAN style = "padding-right: 1em; white-space:nowrap;" class = "" > < A href = "#Global_variables" > Global variables< / A > < / SPAN > < SPAN style = "padding-right: 1em; white-space:nowrap;" class = "" > < A href = "#Nested/Local/Inner_Classes_and_Functions" > Nested/Local/Inner Classes and Functions< / A > < / SPAN > < SPAN style = "padding-right: 1em; white-space:nowrap;" class = "" > < A href = "#List_Comprehensions" > List Comprehensions< / A > < / SPAN > < SPAN style = "padding-right: 1em; white-space:nowrap;" class = "" > < A href = "#Default_Iterators_and_Operators" > Default Iterators and Operators< / A > < / SPAN > < SPAN style = "padding-right: 1em; white-space:nowrap;" class = "" > < A href = "#Generators" > Generators< / A > < / SPAN > < SPAN style = "padding-right: 1em; white-space:nowrap;" class = "" > < A href = "#Lambda_Functions" > Lambda Functions< / A > < / SPAN > < SPAN style = "padding-right: 1em; white-space:nowrap;" class = "" > < A href = "#Default_Argument_Values" > Default Argument Values< / A > < / SPAN > < SPAN style = "padding-right: 1em; white-space:nowrap;" class = "" > < A href = "#Properties" > Properties< / A > < / SPAN > < SPAN style = "padding-right: 1em; white-space:nowrap;" class = "" > < A href = "#True/False_evaluations" > True/False evaluations< / A > < / SPAN > < SPAN style = "padding-right: 1em; white-space:nowrap;" class = "" > < A href = "#Deprecated_Language_Features" > Deprecated Language Features< / A > < / SPAN > < SPAN style = "padding-right: 1em; white-space:nowrap;" class = "" > < A href = "#Lexical_Scoping" > Lexical Scoping< / A > < / SPAN > < SPAN style = "padding-right: 1em; white-space:nowrap;" class = "" > < A href = "#Function_and_Method_Decorators" > Function and Method Decorators< / A > < / SPAN > < SPAN style = "padding-right: 1em; white-space:nowrap;" class = "" > < A href = "#Threading" > Threading< / A > < / SPAN > < SPAN style = "padding-right: 1em; white-space:nowrap;" class = "" > < A href = "#Power_Features" > Power Features< / A > < / SPAN > < / DIV > < / TD >
2009-07-24 04:09:56 +08:00
< / TR >
< TR valign = "top" class = "" >
< TD > < DIV class = "toc_category" > < A href = "#Python_Style_Rules" > Python Style Rules< / A > < / DIV > < / TD >
< TD > < DIV class = "toc_stylepoint" >
< SPAN style = "padding-right: 1em; white-space:nowrap;" class = "" > < A href = "#Semicolons" > Semicolons< / A > < / SPAN > < SPAN style = "padding-right: 1em; white-space:nowrap;" class = "" > < A href = "#Line_length" > Line length< / A > < / SPAN > < SPAN style = "padding-right: 1em; white-space:nowrap;" class = "" > < A href = "#Parentheses" > Parentheses< / A > < / SPAN > < SPAN style = "padding-right: 1em; white-space:nowrap;" class = "" > < A href = "#Indentation" > Indentation< / A > < / SPAN > < SPAN style = "padding-right: 1em; white-space:nowrap;" class = "" > < A href = "#Blank_Lines" > Blank Lines< / A > < / SPAN > < SPAN style = "padding-right: 1em; white-space:nowrap;" class = "" > < A href = "#Whitespace" > Whitespace< / A > < / SPAN > < SPAN style = "padding-right: 1em; white-space:nowrap;" class = "" > < A href = "#Python_Interpreter" > Python Interpreter< / A > < / SPAN > < SPAN style = "padding-right: 1em; white-space:nowrap;" class = "" > < A href = "#Comments" > Comments< / A > < / SPAN > < SPAN style = "padding-right: 1em; white-space:nowrap;" class = "" > < A href = "#Classes" > Classes< / A > < / SPAN > < SPAN style = "padding-right: 1em; white-space:nowrap;" class = "" > < A href = "#Strings" > Strings< / A > < / SPAN > < SPAN style = "padding-right: 1em; white-space:nowrap;" class = "" > < A href = "#TODO_Comments" > TODO Comments< / A > < / SPAN > < SPAN style = "padding-right: 1em; white-space:nowrap;" class = "" > < A href = "#Imports_formatting" > Imports formatting< / A > < / SPAN > < SPAN style = "padding-right: 1em; white-space:nowrap;" class = "" > < A href = "#Statements" > Statements< / A > < / SPAN > < SPAN style = "padding-right: 1em; white-space:nowrap;" class = "" > < A href = "#Access_Control" > Access Control< / A > < / SPAN > < SPAN style = "padding-right: 1em; white-space:nowrap;" class = "" > < A href = "#Naming" > Naming< / A > < / SPAN > < SPAN style = "padding-right: 1em; white-space:nowrap;" class = "" > < A href = "#Main" > Main< / A > < / SPAN > < / DIV > < / TD >
< / TR >
< / TABLE >
< / DIV >
< H2 > Overview< / H2 >
< SPAN class = "" > < H2 name = "Important_Note" id = "Important_Note" > Important Note< / H2 >
< SPAN class = "" > < H3 > < A name = "Displaying_Hidden_Details_in_this_Guide" id = "Displaying_Hidden_Details_in_this_Guide" > Displaying Hidden Details in this Guide< / A > < / H3 >
< SPAN class = "showhide_button" onclick = "javascript:ShowHideByName('Displaying_Hidden_Details_in_this_Guide__body','Displaying_Hidden_Details_in_this_Guide__button')" name = "Displaying_Hidden_Details_in_this_Guide__button" id = "Displaying_Hidden_Details_in_this_Guide__button" > ▶< / SPAN >
< SPAN class = "" >
This style guide contains many details that are initially
hidden from view. They are marked by the triangle icon, which you
see here on your left. Click it now.
You should see "Hooray" appear below.
< / SPAN >
< SPAN class = "" > < BR > < SPAN class = "stylepoint_body" name = "Displaying_Hidden_Details_in_this_Guide__body" id = "Displaying_Hidden_Details_in_this_Guide__body" style = "display: none" > < SPAN class = "link_button" > < A href = "?showone=Displaying_Hidden_Details_in_this_Guide#Displaying_Hidden_Details_in_this_Guide" >
link
< / A > < / SPAN >
< p >
Hooray! Now you know you can expand points to get more
details. Alternatively, there's a "toggle all" at the
top of this document.
< / p >
< / SPAN > < / SPAN >
< / SPAN >
< / SPAN >
< SPAN class = "" > < H2 name = "Background" id = "Background" > Background< / H2 >
< p >
Python is the main scripting language used at Google. This
style guide is a list of < em > do< / em > s and < em > don't< / em > s for Python
programs.
< / p >
< / SPAN >
< SPAN class = "" > < H2 name = "Python_Language_Rules" id = "Python_Language_Rules" > Python Language Rules< / H2 >
< SPAN class = "" > < H3 > < A name = "pychecker" id = "pychecker" > pychecker< / A > < / H3 >
< SPAN class = "showhide_button" onclick = "javascript:ShowHideByName('pychecker__body','pychecker__button')" name = "pychecker__button" id = "pychecker__button" > ▶< / SPAN >
< SPAN class = "" >
Run < code > pychecker< / code > over your code.
< / SPAN >
< SPAN class = "" > < BR > < SPAN class = "stylepoint_body" name = "pychecker__body" id = "pychecker__body" style = "display: none" > < SPAN class = "link_button" > < A href = "?showone=pychecker#pychecker" >
link
< / A > < / SPAN >
< P class = "" >
< SPAN class = "stylepoint_section" > Definition: < / SPAN >
PyChecker is a tool for finding bugs in Python source code. It finds
problems that are typically caught by a compiler for less dynamic
languages like C and C++. It is similar to lint. Because of the
dynamic nature of Python, some warnings may be incorrect; however,
spurious warnings should be fairly infrequent.
< / P >
< P class = "" >
< SPAN class = "stylepoint_section" > Pros: < / SPAN >
Catches easy-to-miss errors like typos, use-vars-before-assignment, etc.
< / P >
< P class = "" >
< SPAN class = "stylepoint_section" > Cons: < / SPAN >
< code > pychecker< / code > isn't perfect. To take
advantage of it, we'll need to sometimes: a) Write around it b)
Suppress its warnings c) Improve it or d) Ignore it.
< / P >
< P class = "" >
< SPAN class = "stylepoint_section" > Decision: < / SPAN >
Make sure you run < code > pychecker< / code > on your code.
< / P >
< p >
For information on how to run < code > pychecker< / code > , see the
< a HREF = "http://pychecker.sourceforge.net" > pychecker
homepage< / a >
< / p >
< p >
To suppress warnings, you can set a module-level variable named
< code > __pychecker__< / code > to suppress appropriate warnings.
For example:
< / p >
< SPAN class = "" > < PRE >
< span class = "external" > < / span > __pychecker__ = 'no-callinit no-classattr'< / PRE > < / SPAN >
< p >
Suppressing in this way has the advantage that we can easily search
for suppressions and revisit them.
< / p >
< p >
You can get a list of pychecker warnings by doing
< code > pychecker --help< / code > .
< / p >
< p >
Unused argument warnings can be suppressed by using `_' as the
identifier for the unused argument or prefixing the argument name with
`unused_'. In situations where changing the argument names is
infeasible, you can mention them at the beginning of the function.
For example:
< / p >
< SPAN class = "" > < PRE >
< span class = "external" > < / span > def foo(a, unused_b, unused_c, d=None, e=None):
< span class = "external" > < / span > (d, e) = (d, e) # Silence pychecker
< span class = "external" > < / span > return a
< span class = "external" > < / span >
< / PRE > < / SPAN >
< p >
Ideally, pychecker would be extended to ensure that such `unused
declarations' were true.
< / p >
< / SPAN > < / SPAN >
< / SPAN >
< SPAN class = "" > < H3 > < A name = "Imports" id = "Imports" > Imports< / A > < / H3 >
< SPAN class = "showhide_button" onclick = "javascript:ShowHideByName('Imports__body','Imports__button')" name = "Imports__button" id = "Imports__button" > ▶< / SPAN >
< SPAN class = "" >
Use < code > import< / code > s for packages and modules only.
< / SPAN >
< SPAN class = "" > < BR > < SPAN class = "stylepoint_body" name = "Imports__body" id = "Imports__body" style = "display: none" > < SPAN class = "link_button" > < A href = "?showone=Imports#Imports" >
link
< / A > < / SPAN >
< P class = "" >
< SPAN class = "stylepoint_section" > Definition: < / SPAN >
Reusability mechanism for sharing code from one module to another.
< / P >
< P class = "" >
< SPAN class = "stylepoint_section" > Pros: < / SPAN >
Simplest and most commonly used way of sharing things.
< / P >
< P class = "" >
< SPAN class = "stylepoint_section" > Cons: < / SPAN > < code > from foo import *< / code > or
< code > from foo import Bar< / code > is
very nasty and can lead to serious maintenance issues because
it makes it hard to find module dependencies.
< / P >
< P class = "" >
< SPAN class = "stylepoint_section" > Decision: < / SPAN >
Use < code > import x< / code > for importing packages and modules.
Use < code > from x import y< / code > only when < code > x< / code > is a
package and < code > y< / code > is a module. This allows the
importer to refer to the module without specifying the full
package prefix. For example the module
< code > sound.effects.echo< / code > may be imported as follows:
< / P >
< SPAN class = "" > < PRE >
< span class = "external" > < / span > from sound.effects import echo
< span class = "external" > < / span > ...
< span class = "external" > < / span > echo.echofilter(input, output, delay=0.7, atten=4)
< span class = "external" > < / span >
< / PRE > < / SPAN >
< p >
Even if the module is in the same package, do not directly import
the module without the full package name. This might cause the
package to be imported twice (with unintended side effects) when the
"main" module that is used to start an application lives inside a
package (and uses modules from that same package).
< / p >
< / SPAN > < / SPAN >
< / SPAN >
< SPAN class = "" > < H3 > < A name = "Packages" id = "Packages" > Packages< / A > < / H3 >
< SPAN class = "showhide_button" onclick = "javascript:ShowHideByName('Packages__body','Packages__button')" name = "Packages__button" id = "Packages__button" > ▶< / SPAN >
< SPAN class = "" >
Import and refer to each module using the full pathname location of
that module.
< / SPAN >
< SPAN class = "" > < BR > < SPAN class = "stylepoint_body" name = "Packages__body" id = "Packages__body" style = "display: none" > < SPAN class = "link_button" > < A href = "?showone=Packages#Packages" >
link
< / A > < / SPAN >
< P class = "" >
< SPAN class = "stylepoint_section" > Pros: < / SPAN >
Avoids conflicts in module names. Makes it easier to find modules.
< / P >
< P class = "" >
< SPAN class = "stylepoint_section" > Cons: < / SPAN >
Makes it harder to deploy code because you have to replicate the
package hierarchy.
< / P >
< P class = "" >
< SPAN class = "stylepoint_section" > Decision: < / SPAN >
All new code should refer to modules based on their package
name.
< / P >
< p >
Imports should be as follows:
< / p >
< SPAN class = "" > < PRE > # Reference in code with complete name.
import sound.effects.echo
# Reference in code with just module name.
from sound.effects import echo
< / PRE > < / SPAN >
< / SPAN > < / SPAN >
< / SPAN >
< SPAN class = "" > < H3 > < A name = "Exceptions" id = "Exceptions" > Exceptions< / A > < / H3 >
< SPAN class = "showhide_button" onclick = "javascript:ShowHideByName('Exceptions__body','Exceptions__button')" name = "Exceptions__button" id = "Exceptions__button" > ▶< / SPAN >
< SPAN class = "" >
Exceptions are allowed but must be used carefully.
< / SPAN >
< SPAN class = "" > < BR > < SPAN class = "stylepoint_body" name = "Exceptions__body" id = "Exceptions__body" style = "display: none" > < SPAN class = "link_button" > < A href = "?showone=Exceptions#Exceptions" >
link
< / A > < / SPAN >
< P class = "" >
< SPAN class = "stylepoint_section" > Definition: < / SPAN >
Exceptions are a means of breaking out of the normal flow of control
of a code block to handle errors or other exceptional conditions.
< / P >
< P class = "" >
< SPAN class = "stylepoint_section" > Pros: < / SPAN >
The control flow of normal operation code is not cluttered by
error-handling code. It also allows the control flow to skip multiple
frames when a certain condition occurs, e.g., returning from N
nested functions in one step instead of having to carry-through
error codes.
< / P >
< P class = "" >
< SPAN class = "stylepoint_section" > Cons: < / SPAN >
May cause the control flow to be confusing. Easy to miss error
cases when making library calls.
< / P >
< P class = "" >
< SPAN class = "stylepoint_section" > Decision: < / SPAN >
Exceptions must follow certain conditions:
< ul >
< li > Raise exceptions like this: < code > raise MyException("Error
message")< / code > or < code > raise MyException< / code > . Do not
use the two-argument form (< code > raise MyException, "Error
message"< / code > ) or deprecated string-based exceptions
(< code > raise "Error message"< / code > ).< / li >
< li > Modules or packages should define their own domain-specific
base exception class, which should inherit from the built-in
Exception class. The base exception for a module should be called
< code > Error< / code > .
< SPAN class = "" > < PRE >
< span class = "external" > < / span > class Error(Exception):
< span class = "external" > < / span > pass< / PRE > < / SPAN >
< / li >
< li > Never use catch-all < code > except:< / code > statements, or
catch < code > Exception< / code > or < code > StandardError< / code > ,
unless you are re-raising the exception or in the outermost
block in your thread (and printing an error message). Python
is very tolerant in this regard and < code > except:< / code > will
really catch everything including Python syntax errors. It is
easy to hide real bugs using < code > except:< / code > .< / li >
< li > Minimize the amount of code in a
< code > try< / code > /< code > except< / code > block. The larger the
body of the < code > try< / code > , the more likely that an
exception will be raised by a line of code that you didn't
expect to raise an exception. In those cases,
the < code > try< / code > /< code > except< / code > block hides a real
error.< / li >
< li > Use the < code > finally< / code > clause to execute code whether
or not an exception is raised in the < code > try< / code > block.
This is often useful for cleanup, i.e., closing a file.< / li >
< / ul >
< / P >
< / SPAN > < / SPAN >
< / SPAN >
< SPAN class = "" > < H3 > < A name = "Global_variables" id = "Global_variables" > Global variables< / A > < / H3 >
< SPAN class = "showhide_button" onclick = "javascript:ShowHideByName('Global_variables__body','Global_variables__button')" name = "Global_variables__button" id = "Global_variables__button" > ▶< / SPAN >
< SPAN class = "" >
Avoid global variables.
< / SPAN >
< SPAN class = "" > < BR > < SPAN class = "stylepoint_body" name = "Global_variables__body" id = "Global_variables__body" style = "display: none" > < SPAN class = "link_button" > < A href = "?showone=Global_variables#Global_variables" >
link
< / A > < / SPAN >
< P class = "" >
< SPAN class = "stylepoint_section" > Definition: < / SPAN >
Variables that are declared at the module level.
< / P >
< P class = "" >
< SPAN class = "stylepoint_section" > Pros: < / SPAN >
Occasionally useful.
< / P >
< P class = "" >
< SPAN class = "stylepoint_section" > Cons: < / SPAN >
Has the potential to change module behavior during the import,
because assignments to module-level variables are done when the
module is imported.
< / P >
< P class = "" >
< SPAN class = "stylepoint_section" > Decision: < / SPAN >
Avoid global variables in favor of class variables. Some
exceptions are:
< ul >
< li > Default options for scripts.< / li >
< li > Module-level constants. For example: < code > PI = 3.14159< / code > .
Constants should be named using all caps with underscores;
see < a HREF = "#Naming" > Naming< / a > below.< / li >
< li > It is sometimes useful for globals to cache values needed
or returned by functions.< / li >
< li > If needed, globals should be made internal to the module
and accessed through public module level functions;
see < a HREF = "#Naming" > Naming< / a > below.< / li >
< / ul >
< / P >
< / SPAN > < / SPAN >
< / SPAN >
< SPAN class = "" > < H3 > < A name = "Nested/Local/Inner_Classes_and_Functions" id = "Nested/Local/Inner_Classes_and_Functions" > Nested/Local/Inner Classes and Functions< / A > < / H3 >
< SPAN class = "showhide_button" onclick = "javascript:ShowHideByName('Nested/Local/Inner_Classes_and_Functions__body','Nested/Local/Inner_Classes_and_Functions__button')" name = "Nested/Local/Inner_Classes_and_Functions__button" id = "Nested/Local/Inner_Classes_and_Functions__button" > ▶< / SPAN >
< SPAN class = "" >
Nested/local/inner classes and functions are fine.
< / SPAN >
< SPAN class = "" > < BR > < SPAN class = "stylepoint_body" name = "Nested/Local/Inner_Classes_and_Functions__body" id = "Nested/Local/Inner_Classes_and_Functions__body" style = "display: none" > < SPAN class = "link_button" > < A href = "?showone=Nested/Local/Inner_Classes_and_Functions#Nested/Local/Inner_Classes_and_Functions" >
link
< / A > < / SPAN >
< P class = "" >
< SPAN class = "stylepoint_section" > Definition: < / SPAN >
2009-10-24 05:01:49 +08:00
A class can be defined inside of a method, function, or class. A
function can be defined inside a method or function. Nested functions
have read-only access to variables defined in enclosing scopes.
2009-07-24 04:09:56 +08:00
< / P >
< P class = "" >
< SPAN class = "stylepoint_section" > Pros: < / SPAN >
Allows definition of utility classes and functions that are only
used inside of a very limited scope. Very < a HREF = "http://en.wikipedia.org/wiki/Abstract_data_type" > ADT< / a > -y.
< / P >
< P class = "" >
< SPAN class = "stylepoint_section" > Cons: < / SPAN >
Instances of nested or local classes cannot be pickled.
< / P >
< P class = "" >
< SPAN class = "stylepoint_section" > Decision: < / SPAN >
They are fine.
< / P >
< / SPAN > < / SPAN >
< / SPAN >
< SPAN class = "" > < H3 > < A name = "List_Comprehensions" id = "List_Comprehensions" > List Comprehensions< / A > < / H3 >
< SPAN class = "showhide_button" onclick = "javascript:ShowHideByName('List_Comprehensions__body','List_Comprehensions__button')" name = "List_Comprehensions__button" id = "List_Comprehensions__button" > ▶< / SPAN >
< SPAN class = "" >
Okay to use for simple cases.
< / SPAN >
< SPAN class = "" > < BR > < SPAN class = "stylepoint_body" name = "List_Comprehensions__body" id = "List_Comprehensions__body" style = "display: none" > < SPAN class = "link_button" > < A href = "?showone=List_Comprehensions#List_Comprehensions" >
link
< / A > < / SPAN >
< P class = "" >
< SPAN class = "stylepoint_section" > Definition: < / SPAN >
List comprehensions and generator expressions provide a concise
and efficient way to create lists and iterators without
resorting to the use of < code > map()< / code > ,
< code > filter()< / code > , or < code > lambda< / code > .
< / P >
< P class = "" >
< SPAN class = "stylepoint_section" > Pros: < / SPAN >
Simple list comprehensions can be clearer and simpler than
other list creation techniques. Generator expressions can be
very efficient, since they avoid the creation of a list
entirely.
< / P >
< P class = "" >
< SPAN class = "stylepoint_section" > Cons: < / SPAN >
Complicated list comprehensions or generator expressions can be
hard to read.
< / P >
< P class = "" >
< SPAN class = "stylepoint_section" > Decision: < / SPAN >
Okay to use for simple cases. Each portion must fit on one line:
mapping expression, < code > for< / code > clause, filter expression.
Multiple < code > for< / code > clauses or filter expressions are not
permitted. Use loops instead when things get more complicated.
< / P >
< SPAN class = "" > < PRE class = "badcode" > No< span class = "external" > < / span > :
< span class = "external" > < / span > result = [(x, y) for x in range(10) for y in range(5) if x * y > 10]
< span class = "external" > < / span > return ((x, y, z)
< span class = "external" > < / span > for x in xrange(5)
< span class = "external" > < / span > for y in xrange(5)
< span class = "external" > < / span > if x != y
< span class = "external" > < / span > for z in xrange(5)
< span class = "external" > < / span > if y != z)< / PRE > < / SPAN >
< SPAN class = "" > < PRE > Ye< span class = "external" > < / span > s:
< span class = "external" > < / span > result = []
< span class = "external" > < / span > for x in range(10):
< span class = "external" > < / span > for y in range(5):
< span class = "external" > < / span > if x * y > 10:
< span class = "external" > < / span > result.append((x, y))
< span class = "external" > < / span > for x in xrange(5):
< span class = "external" > < / span > for y in xrange(5):
< span class = "external" > < / span > if x != y:
< span class = "external" > < / span > for z in xrange(5):
< span class = "external" > < / span > if y != z:
< span class = "external" > < / span > yield (x, y, z)
< span class = "external" > < / span > return ((x, complicated_transform(x))
< span class = "external" > < / span > for x in long_generator_function(parameter)
< span class = "external" > < / span > if x is not None)
< span class = "external" > < / span > squares = [x * x for x in range(10)]
< span class = "external" > < / span > eat(jelly_bean for jelly_bean in jelly_beans
< span class = "external" > < / span > if jelly_bean.color == 'black')< / PRE > < / SPAN >
< / SPAN > < / SPAN >
< / SPAN >
< SPAN class = "" > < H3 > < A name = "Default_Iterators_and_Operators" id = "Default_Iterators_and_Operators" > Default Iterators and Operators< / A > < / H3 >
< SPAN class = "showhide_button" onclick = "javascript:ShowHideByName('Default_Iterators_and_Operators__body','Default_Iterators_and_Operators__button')" name = "Default_Iterators_and_Operators__button" id = "Default_Iterators_and_Operators__button" > ▶< / SPAN >
< SPAN class = "" >
Use default iterators and operators for types that support them,
like lists, dictionaries, and files.
< / SPAN >
< SPAN class = "" > < BR > < SPAN class = "stylepoint_body" name = "Default_Iterators_and_Operators__body" id = "Default_Iterators_and_Operators__body" style = "display: none" > < SPAN class = "link_button" > < A href = "?showone=Default_Iterators_and_Operators#Default_Iterators_and_Operators" >
link
< / A > < / SPAN >
< P class = "" >
< SPAN class = "stylepoint_section" > Definition: < / SPAN >
Container types, like dictionaries and lists, define default
iterators and membership test operators ("in" and "not in").
< / P >
< P class = "" >
< SPAN class = "stylepoint_section" > Pros: < / SPAN >
The default iterators and operators are simple and efficient.
They express the operation directly, without extra method calls.
A function that uses default operators is generic. It can be
used with any type that supports the operation.
< / P >
< P class = "" >
< SPAN class = "stylepoint_section" > Cons: < / SPAN >
You can't tell the type of objects by reading the method names
(e.g. has_key() means a dictionary). This is also an advantage.
< / P >
< P class = "" >
< SPAN class = "stylepoint_section" > Decision: < / SPAN > Use default iterators and operators for types
that support them, like lists, dictionaries, and files. The
built-in types define iterator methods, too. Prefer these
methods to methods that return lists, except that you should not
mutate a container while iterating over it.
< SPAN class = "" > < PRE > Yes: < span class = "external" > < / span > for key in adict: ...
< span class = "external" > < / span > if key not in adict: ...
< span class = "external" > < / span > if obj in alist: ...
< span class = "external" > < / span > for line in afile: ...
< span class = "external" > < / span > for k, v in dict.iteritems(): ...< / PRE > < / SPAN >
< SPAN class = "" > < PRE class = "badcode" > No: < span class = "external" > < / span > for key in adict.keys(): ...
< span class = "external" > < / span > if not adict.has_key(key): ...
< span class = "external" > < / span > for line in afile.readlines(): ...< / PRE > < / SPAN >
< / P >
< / SPAN > < / SPAN >
< / SPAN >
< SPAN class = "" > < H3 > < A name = "Generators" id = "Generators" > Generators< / A > < / H3 >
< SPAN class = "showhide_button" onclick = "javascript:ShowHideByName('Generators__body','Generators__button')" name = "Generators__button" id = "Generators__button" > ▶< / SPAN >
< SPAN class = "" >
Use generators as needed.
< / SPAN >
< SPAN class = "" > < BR > < SPAN class = "stylepoint_body" name = "Generators__body" id = "Generators__body" style = "display: none" > < SPAN class = "link_button" > < A href = "?showone=Generators#Generators" >
link
< / A > < / SPAN >
< P class = "" >
< SPAN class = "stylepoint_section" > Definition: < / SPAN >
A generator function returns an iterator that yields a value each
time it executes a yield statement. After it yields a value, the
runtime state of the generator function is suspended until the
next value is needed.
< / P >
< P class = "" >
< SPAN class = "stylepoint_section" > Pros: < / SPAN >
Simpler code, because the state of local variables and control flow
are preserved for each call. A generator uses less memory than a
function that creates an entire list of values at once.
< / P >
< P class = "" >
< SPAN class = "stylepoint_section" > Cons: < / SPAN >
None.
< / P >
< P class = "" >
< SPAN class = "stylepoint_section" > Decision: < / SPAN >
Fine. Use "Yields:" rather than "Returns:" in the
doc string for generator functions.
< / P >
< / SPAN > < / SPAN >
< / SPAN >
< SPAN class = "" > < H3 > < A name = "Lambda_Functions" id = "Lambda_Functions" > Lambda Functions< / A > < / H3 >
< SPAN class = "showhide_button" onclick = "javascript:ShowHideByName('Lambda_Functions__body','Lambda_Functions__button')" name = "Lambda_Functions__button" id = "Lambda_Functions__button" > ▶< / SPAN >
< SPAN class = "" >
Okay for one-liners.
< / SPAN >
< SPAN class = "" > < BR > < SPAN class = "stylepoint_body" name = "Lambda_Functions__body" id = "Lambda_Functions__body" style = "display: none" > < SPAN class = "link_button" > < A href = "?showone=Lambda_Functions#Lambda_Functions" >
link
< / A > < / SPAN >
< P class = "" >
< SPAN class = "stylepoint_section" > Definition: < / SPAN >
Lambdas define anonymous functions in an expression, as
opposed to a statement. They are often used to define callbacks or
operators for higher-order functions like < code > map()< / code > and
< code > filter()< / code > .
< / P >
< P class = "" >
< SPAN class = "stylepoint_section" > Pros: < / SPAN >
Convenient.
< / P >
< P class = "" >
< SPAN class = "stylepoint_section" > Cons: < / SPAN > Harder to read and debug than local functions. The
lack of names means stack traces are more difficult to
understand. Expressiveness is limited because the function may
only contain an expression.
< / P >
< P class = "" >
< SPAN class = "stylepoint_section" > Decision: < / SPAN >
Okay to use them for one-liners. If the code inside the lambda
function is any longer than 60– 80 chars, it's probably better to
define it as a regular (nested) function.
< p >
For common operations like multiplication, use the functions from the
< code > operator< / code > module instead of lambda functions. For
example, prefer < code > operator.mul< / code > to < code > lambda
x, y: x * y< / code > .
< / p >
< / P >
< / SPAN > < / SPAN >
< / SPAN >
< SPAN class = "" > < H3 > < A name = "Default_Argument_Values" id = "Default_Argument_Values" > Default Argument Values< / A > < / H3 >
< SPAN class = "showhide_button" onclick = "javascript:ShowHideByName('Default_Argument_Values__body','Default_Argument_Values__button')" name = "Default_Argument_Values__button" id = "Default_Argument_Values__button" > ▶< / SPAN >
< SPAN class = "" >
Okay in most cases.
< / SPAN >
< SPAN class = "" > < BR > < SPAN class = "stylepoint_body" name = "Default_Argument_Values__body" id = "Default_Argument_Values__body" style = "display: none" > < SPAN class = "link_button" > < A href = "?showone=Default_Argument_Values#Default_Argument_Values" >
link
< / A > < / SPAN >
< P class = "" >
< SPAN class = "stylepoint_section" > Definition: < / SPAN >
You can specify values for variables at the end of a function's
parameter list, e.g., < code > def foo(a, b=0):< / code > . If
< code > foo< / code > is called with only one argument,
< code > b< / code > is set to 0. If it is called with two arguments,
< code > b< / code > has the value of the second argument.
< / P >
< P class = "" >
< SPAN class = "stylepoint_section" > Pros: < / SPAN >
Often you have a function that uses lots of default values,
but—rarely—you want to override the
defaults. Default argument values provide an easy way to do this,
without having to define lots of functions for the rare
exceptions. Also, Python does not support overloaded
methods/functions and default arguments are an easy way of
"faking" the overloading behavior.
< / P >
< P class = "" >
< SPAN class = "stylepoint_section" > Cons: < / SPAN >
Default arguments are evaluated once at module load
time. This may cause problems if the argument is a mutable
object such as a list or a dictionary. If the function modifies
the object (e.g., by appending an item to a list), the default
value is modified.
< / P >
< P class = "" >
< SPAN class = "stylepoint_section" > Decision: < / SPAN >
Okay to use with the following caveats:
< p >
Do not use mutable objects as default values in the function or method
definition.
< / p >
< SPAN class = "" > < PRE > Yes: < span class = "external" > < / span > def foo(a, b=None):
< span class = "external" > < / span > if b is None:
< span class = "external" > < / span > b = []< / PRE > < / SPAN >
< SPAN class = "" > < PRE class = "badcode" > No: < span class = "external" > < / span > def foo(a, b=[]):
< span class = "external" > < / span > ...< / PRE > < / SPAN >
< p >
2009-10-24 05:01:49 +08:00
Calling code must use named values for arguments with a default value.
This helps document the code somewhat and helps prevent and detect
2009-07-24 04:09:56 +08:00
interface breakage when more arguments are added.
< / p >
< SPAN class = "" > < PRE >
< span class = "external" > < / span > def foo(a, b=1):
< span class = "external" > < / span > ...< / PRE > < / SPAN >
< SPAN class = "" > < PRE > Yes: < span class = "external" > < / span > foo(1)
< span class = "external" > < / span > foo(1, b=2)< / PRE > < / SPAN >
< SPAN class = "" > < PRE class = "badcode" > No: < span class = "external" > < / span > foo(1, 2)< / PRE > < / SPAN >
< / P >
< / SPAN > < / SPAN >
< / SPAN >
< SPAN class = "" > < H3 > < A name = "Properties" id = "Properties" > Properties< / A > < / H3 >
< SPAN class = "showhide_button" onclick = "javascript:ShowHideByName('Properties__body','Properties__button')" name = "Properties__button" id = "Properties__button" > ▶< / SPAN >
< SPAN class = "" >
Use properties for accessing or setting data where you would
normally have used simple, lightweight accessor or setter methods.
< / SPAN >
< SPAN class = "" > < BR > < SPAN class = "stylepoint_body" name = "Properties__body" id = "Properties__body" style = "display: none" > < SPAN class = "link_button" > < A href = "?showone=Properties#Properties" >
link
< / A > < / SPAN >
< P class = "" >
< SPAN class = "stylepoint_section" > Definition: < / SPAN > A way to wrap method calls for getting and
setting an attribute as a standard attribute access when the
computation is lightweight.
< / P >
< P class = "" >
< SPAN class = "stylepoint_section" > Pros: < / SPAN > Readability is increased by eliminating explicit
get and set method calls for simple attribute access. Allows
calculations to be lazy. Considered the Pythonic way to
maintain the interface of a class. In terms of performance,
allowing properties bypasses needing trivial accessor methods
when a direct variable access is reasonable. This also allows
accessor methods to be added in the future without breaking the
interface.
< / P >
< P class = "" >
< SPAN class = "stylepoint_section" > Cons: < / SPAN > Properties are specified after the getter and
setter methods are declared, requiring one to notice they are
used for properties farther down in the code (except for readonly
properties created with the < code > @property< / code > decorator - see
below). Must inherit from
< code > object< / code > . Can hide side-effects much like operator
overloading. Can be confusing for subclasses.
< / P >
< P class = "" >
< SPAN class = "stylepoint_section" > Decision: < / SPAN > Use properties in new code to access or
set data where you would normally have used simple, lightweight
accessor or setter methods. Read-only properties should be created
with the < code > @property< / code >
< a HREF = "#Function_and_Method_Decorators" > decorator< / a > .
< p > < a id = "properties-template-dp" >
Inheritance with properties can be non-obvious if the property itself is
not overridden. Thus one must make sure that accessor methods are
called indirectly to ensure methods overridden in subclasses are called
by the property (using the Template Method DP).
< / a > < / p >
< SPAN class = "" > < PRE > Yes: < span class = "external" > < / span > import math
< span class = "external" > < / span > class Square(object):
< span class = "external" > < / span > """A square with two properties: a writable area and a read-only perimeter.
< span class = "external" > < / span > To use:
< span class = "external" > < / span > > > > sq = Square(3)
< span class = "external" > < / span > > > > sq.area
< span class = "external" > < / span > 9
< span class = "external" > < / span > > > > sq.perimeter
< span class = "external" > < / span > 12
< span class = "external" > < / span > > > > sq.area = 16
< span class = "external" > < / span > > > > sq.side
< span class = "external" > < / span > 4
< span class = "external" > < / span > > > > sq.perimeter
< span class = "external" > < / span > 16
< span class = "external" > < / span > """
< span class = "external" > < / span > def __init__(self, side):
< span class = "external" > < / span > self.side = side
< span class = "external" > < / span > def __get_area(self):
< span class = "external" > < / span > """Calculates the 'area' property."""
< span class = "external" > < / span > return self.side ** 2
< span class = "external" > < / span > def ___get_area(self):
< span class = "external" > < / span > """Indirect accessor for 'area' property."""
< span class = "external" > < / span > return self.__get_area()
< span class = "external" > < / span > def __set_area(self, area):
< span class = "external" > < / span > """Sets the 'area' property."""
< span class = "external" > < / span > self.side = math.sqrt(area)
< span class = "external" > < / span > def ___set_area(self, area):
< span class = "external" > < / span > """Indirect setter for 'area' property."""
< span class = "external" > < / span > self._SetArea(area)
< span class = "external" > < / span > area = property(___get_area, ___set_area,
< span class = "external" > < / span > doc="""Gets or sets the area of the square.""")
< span class = "external" > < / span > @property
< span class = "external" > < / span > def perimeter(self):
< span class = "external" > < / span > return self.side * 4
< span class = "external" > < / span >
< / PRE > < / SPAN >
< / P >
< / SPAN > < / SPAN >
< / SPAN >
< SPAN class = "" > < H3 > < A name = "True/False_evaluations" id = "True/False_evaluations" > True/False evaluations< / A > < / H3 >
< SPAN class = "showhide_button" onclick = "javascript:ShowHideByName('True/False_evaluations__body','True/False_evaluations__button')" name = "True/False_evaluations__button" id = "True/False_evaluations__button" > ▶< / SPAN >
< SPAN class = "" >
Use the "implicit" false if at all possible.
< / SPAN >
< SPAN class = "" > < BR > < SPAN class = "stylepoint_body" name = "True/False_evaluations__body" id = "True/False_evaluations__body" style = "display: none" > < SPAN class = "link_button" > < A href = "?showone=True/False_evaluations#True/False_evaluations" >
link
< / A > < / SPAN >
< P class = "" >
< SPAN class = "stylepoint_section" > Definition: < / SPAN > Python evaluates certain values as < code > false< / code >
when in a boolean context. A quick "rule of thumb" is that all
"empty" values are considered < code > false< / code > so < code > 0, None, [], {},
""< / code > all evaluate as < code > false< / code > in a boolean context.
< / P >
< P class = "" >
< SPAN class = "stylepoint_section" > Pros: < / SPAN > Conditions using Python booleans are easier to read
and less error-prone. In most cases, they're also faster.
< / P >
< P class = "" >
< SPAN class = "stylepoint_section" > Cons: < / SPAN >
May look strange to C/C++ developers.
< / P >
< P class = "" >
< SPAN class = "stylepoint_section" > Decision: < / SPAN >
Use the "implicit" false if at all possible, e.g., < code > if
foo:< / code > rather than < code > if foo != []:< / code > . There are a
few caveats that you should keep in mind though:
< ul >
< li >
Never use < code > ==< / code > or < code > !=< / code > to compare
singletons like < code > None< / code > . Use < code > is< / code >
or < code > is not< / code > .< / li >
< li > Beware of writing < code > if x:< / code > when you really mean
< code > if x is not None:< / code > —e.g., when testing whether
a variable or argument that defaults to < code > None< / code > was
set to some other value. The other value might be a value
that's false in a boolean context!< / li >
< li >
Never compare a boolean variable to < code > False< / code > using
< code > ==< / code > . Use < code > if not x:< / code > instead. If
you need to distinguish < code > False< / code > from
< code > None< / code > then chain the expressions,
such as < code > if not x and x is not None:< / code > .
< / li >
< li >
For sequences (strings, lists, tuples), use the fact that
empty sequences are false, so < code > if not seq:< / code > or
< code > if seq:< / code > is preferable to < code > if
len(seq):< / code > or < code > if not
len(seq):< / code > .< / li >
< li >
When handling integers, implicit false may involve more risk than
benefit (i.e., accidentally handling < code > None< / code > as 0). You may
compare a value which is known to be an integer (and is not the
result of < code > len()< / code > ) against the integer 0.
< SPAN class = "" > < PRE > Yes: < span class = "external" > < / span > if not users:
< span class = "external" > < / span > print 'no users'
< span class = "external" > < / span > if foo == 0:
< span class = "external" > < / span > self.handle_zero()
< span class = "external" > < / span > if i % 10 == 0:
< span class = "external" > < / span > self.handle_multiple_of_ten()< / PRE > < / SPAN >
< SPAN class = "" > < PRE class = "badcode" > No: < span class = "external" > < / span > if len(users) == 0:
< span class = "external" > < / span > print 'no users'
< span class = "external" > < / span > if foo is not None and not foo:
< span class = "external" > < / span > self.handle_zero()
< span class = "external" > < / span > if not i % 10:
< span class = "external" > < / span > self.handle_multiple_of_ten()< / PRE > < / SPAN >
< / li >
< li >
Note that < code > '0'< / code > (i.e., < code > 0< / code > as string)
evaluates to true.< / li >
< / ul >
< / P >
< / SPAN > < / SPAN >
< / SPAN >
2009-10-24 05:01:49 +08:00
< SPAN class = "" > < H3 > < A name = "Deprecated_Language_Features" id = "Deprecated_Language_Features" > Deprecated Language Features< / A > < / H3 >
< SPAN class = "showhide_button" onclick = "javascript:ShowHideByName('Deprecated_Language_Features__body','Deprecated_Language_Features__button')" name = "Deprecated_Language_Features__button" id = "Deprecated_Language_Features__button" > ▶< / SPAN >
2009-07-24 04:09:56 +08:00
< SPAN class = "" >
2009-10-24 05:01:49 +08:00
Use string methods instead of the < code > string< / code > module
where possible. Use function call syntax instead
of < code > apply< / code > . Use list comprehensions
and < code > for< / code > loops instead of < code > filter< / code > ,
< code > map< / code > , and < code > reduce< / code > .
2009-07-24 04:09:56 +08:00
< / SPAN >
2009-10-24 05:01:49 +08:00
< SPAN class = "" > < BR > < SPAN class = "stylepoint_body" name = "Deprecated_Language_Features__body" id = "Deprecated_Language_Features__body" style = "display: none" > < SPAN class = "link_button" > < A href = "?showone=Deprecated_Language_Features#Deprecated_Language_Features" >
2009-07-24 04:09:56 +08:00
link
< / A > < / SPAN >
< P class = "" >
2009-10-24 05:01:49 +08:00
< SPAN class = "stylepoint_section" > Definition: < / SPAN >
Current versions of Python provide alternative constructs
that people find generally preferable.
2009-07-24 04:09:56 +08:00
< / P >
< P class = "" >
2009-10-24 05:01:49 +08:00
< SPAN class = "stylepoint_section" > Decision: < / SPAN >
We do not use any Python version which does not support
these features, so there is no reason not to use the new
styles.
< SPAN class = "" > < PRE class = "badcode" > No: < span class = "external" > < / span > words = string.split(foo, ':')
< span class = "external" > < / span > map(lambda x: x[1], filter(lambda x: x[2] == 5, my_list))
< span class = "external" > < / span > apply(fn, args, kwargs)< / PRE > < / SPAN >
< SPAN class = "" > < PRE > Yes: < span class = "external" > < / span > words = foo.split(':')
< span class = "external" > < / span > [x[1] for x in my_list if x[2] == 5]
< span class = "external" > < / span > fn(*args, **kwargs)< / PRE > < / SPAN >
2009-07-24 04:09:56 +08:00
< / P >
< / SPAN > < / SPAN >
< / SPAN >
< SPAN class = "" > < H3 > < A name = "Lexical_Scoping" id = "Lexical_Scoping" > Lexical Scoping< / A > < / H3 >
< SPAN class = "showhide_button" onclick = "javascript:ShowHideByName('Lexical_Scoping__body','Lexical_Scoping__button')" name = "Lexical_Scoping__button" id = "Lexical_Scoping__button" > ▶< / SPAN >
< SPAN class = "" >
Okay to use.
< / SPAN >
< SPAN class = "" > < BR > < SPAN class = "stylepoint_body" name = "Lexical_Scoping__body" id = "Lexical_Scoping__body" style = "display: none" > < SPAN class = "link_button" > < A href = "?showone=Lexical_Scoping#Lexical_Scoping" >
link
< / A > < / SPAN >
< P class = "" >
< SPAN class = "stylepoint_section" > Definition: < / SPAN >
A nested Python function can refer to variables defined in
enclosing functions, but can not assign to them. Variable
bindings are resolved using lexical scoping, that is, based on
the static program text. Any assignment to a name in a block
will cause Python to treat all references to that name as a
local variable, even if the use precedes the assignment. If a
global declaration occurs, the name is treated as a global
variable.
< p >
An example of the use of this feature is:
< / p >
< SPAN class = "" > < PRE >
< span class = "external" > < / span > def get_adder(summand1):
< span class = "external" > < / span > """Returns a function that adds numbers to a given number."""
< span class = "external" > < / span > def adder(summand2):
< span class = "external" > < / span > return summand1 + summand2
< span class = "external" > < / span > return adder
< span class = "external" > < / span >
< / PRE > < / SPAN >
< / P >
< P class = "" >
< SPAN class = "stylepoint_section" > Pros: < / SPAN >
Often results in clearer, more elegant code. Especially comforting
to experienced Lisp and Scheme (and Haskell and ML and …)
programmers.
< / P >
< P class = "" >
< SPAN class = "stylepoint_section" > Cons: < / SPAN >
Can lead to confusing bugs. Such as this example based on
< a HREF = "http://www.python.org/dev/peps/pep-0227/" > PEP-0227< / a > :
< SPAN class = "" > < PRE class = "badcode" >
< span class = "external" > < / span > i = 4
< span class = "external" > < / span > def foo(x):
< span class = "external" > < / span > def bar():
< span class = "external" > < / span > print i,
< span class = "external" > < / span > # ...
< span class = "external" > < / span > # A bunch of code here
< span class = "external" > < / span > # ...
< span class = "external" > < / span > for i in x: # Ah, i *is* local to Foo, so this is what Bar sees
< span class = "external" > < / span > print i,
< span class = "external" > < / span > bar()< / PRE > < / SPAN >
< p >
So < code > foo([1, 2, 3])< / code > will print < code > 1 2 3 3< / code > , not
< code > 1 2 3 4< / code > .
< / p >
< / P >
< P class = "" >
< SPAN class = "stylepoint_section" > Decision: < / SPAN >
Okay to use.
< / P >
< / SPAN > < / SPAN >
< / SPAN >
< SPAN class = "" > < H3 > < A name = "Function_and_Method_Decorators" id = "Function_and_Method_Decorators" > Function and Method Decorators< / A > < / H3 >
< SPAN class = "showhide_button" onclick = "javascript:ShowHideByName('Function_and_Method_Decorators__body','Function_and_Method_Decorators__button')" name = "Function_and_Method_Decorators__button" id = "Function_and_Method_Decorators__button" > ▶< / SPAN >
< SPAN class = "" >
Use decorators judiciously when there is a clear advantage.
< / SPAN >
< SPAN class = "" > < BR > < SPAN class = "stylepoint_body" name = "Function_and_Method_Decorators__body" id = "Function_and_Method_Decorators__body" style = "display: none" > < SPAN class = "link_button" > < A href = "?showone=Function_and_Method_Decorators#Function_and_Method_Decorators" >
link
< / A > < / SPAN >
< P class = "" >
< SPAN class = "stylepoint_section" > Definition: < / SPAN >
< a HREF = "http://www.python.org/doc/2.4.3/whatsnew/node6.html" > Decorators
for Functions and Methods< / a >
(a.k.a "the < code > @< / code > notation").
The most common decorators are < code > @classmethod< / code > and
< code > @staticmethod< / code > , for converting ordinary methods to class or
static methods. However, the decorator syntax allows for
user-defined decorators as well. Specifically, for some function
< code > my_decorator< / code > , this:
< SPAN class = "" > < PRE >
< span class = "external" > < / span > class C(object):
< span class = "external" > < / span > @my_decorator
< span class = "external" > < / span > def method(self):
< span class = "external" > < / span > # method body ...
< span class = "external" > < / span >
< / PRE > < / SPAN >
is equivalent to:
< SPAN class = "" > < PRE >
< span class = "external" > < / span > class C(object):
< span class = "external" > < / span > def method(self):
< span class = "external" > < / span > # method body ...
< span class = "external" > < / span > method = my_decorator(method)
< span class = "external" > < / span >
< / PRE > < / SPAN >
< / P >
< P class = "" >
< SPAN class = "stylepoint_section" > Pros: < / SPAN > Elegantly specifies some transformation on a method; the
transformation might eliminate some repetitive code, enforce invariants,
etc.
< / P >
< P class = "" >
< SPAN class = "stylepoint_section" > Cons: < / SPAN > Decorators can perform arbitrary operations on a
function's arguments or return values, resulting in surprising
implicit behavior.
Additionally, decorators execute at import time. Failures in decorator
code are pretty much impossible to recover from.
< / P >
< P class = "" >
< SPAN class = "stylepoint_section" > Decision: < / SPAN > Use decorators judiciously when there is a clear
advantage. Decorators should follow the same import and naming
guidelines as functions. Decorator pydoc should clearly state that the
function is a decorator. Write unit tests for decorators.
< p >
Avoid external dependencies in the decorator itself (e.g. don't rely on
files, sockets, database connections, etc.), since they might not be
available when the decorator runs (at import time, perhaps from
< code > pychecker< / code > or other tools). A decorator that is
called with valid parameters should (as much as possible) be guaranteed
to succeed in all cases.
< / p >
< p >
Decorators are a special case of "top level code" - see
< a HREF = "#Main" > main< / a > for more discussion.
< / p >
< / P >
< / SPAN > < / SPAN >
< / SPAN >
< SPAN class = "" > < H3 > < A name = "Threading" id = "Threading" > Threading< / A > < / H3 >
< SPAN class = "showhide_button" onclick = "javascript:ShowHideByName('Threading__body','Threading__button')" name = "Threading__button" id = "Threading__button" > ▶< / SPAN >
< SPAN class = "" >
Do not rely on the atomicity of built-in types.
< / SPAN >
< SPAN class = "" > < BR > < SPAN class = "stylepoint_body" name = "Threading__body" id = "Threading__body" style = "display: none" > < SPAN class = "link_button" > < A href = "?showone=Threading#Threading" >
link
< / A > < / SPAN >
< p >
While Python's built-in data types such as dictionaries appear
to have atomic operations, there are corner cases where they
aren't atomic (e.g. if < code > __hash__< / code > or
< code > __eq__< / code > are implemented as Python methods) and their
atomicity should not be relied upon. Neither should you rely on
atomic variable assignment (since this in turn depends on
dictionaries).
< / p >
< p >
Use the Queue module's < code > Queue< / code > data type as the preferred
way to
communicate data between threads. Otherwise, use the threading
module and its locking primitives. Learn about the proper use
of condition variables so you can use
< code > threading.Condition< / code > instead of using lower-level
locks.
< / p >
< / SPAN > < / SPAN >
< / SPAN >
< SPAN class = "" > < H3 > < A name = "Power_Features" id = "Power_Features" > Power Features< / A > < / H3 >
< SPAN class = "showhide_button" onclick = "javascript:ShowHideByName('Power_Features__body','Power_Features__button')" name = "Power_Features__button" id = "Power_Features__button" > ▶< / SPAN >
< SPAN class = "" >
Avoid these features.
< / SPAN >
< SPAN class = "" > < BR > < SPAN class = "stylepoint_body" name = "Power_Features__body" id = "Power_Features__body" style = "display: none" > < SPAN class = "link_button" > < A href = "?showone=Power_Features#Power_Features" >
link
< / A > < / SPAN >
< P class = "" >
< SPAN class = "stylepoint_section" > Definition: < / SPAN > Python is an extremely flexible language and
gives you many fancy features such as metaclasses, access to bytecode,
on-the-fly compilation, dynamic inheritance, object reparenting,
import hacks, reflection, modification of system internals,
etc.
< / P >
< P class = "" >
< SPAN class = "stylepoint_section" > Pros: < / SPAN > These are powerful language features. They can
make your code more compact.
< / P >
< P class = "" >
< SPAN class = "stylepoint_section" > Cons: < / SPAN > It's very tempting to use these "cool" features
when they're not absolutely necessary. It's harder to read,
understand, and debug code that's using unusual features
underneath. It doesn't seem that way at first (to the original
author), but when revisiting the code, it tends to be more
difficult than code that is longer but is straightforward.
< / P >
< P class = "" >
< SPAN class = "stylepoint_section" > Decision: < / SPAN >
Avoid these features in
your code.
< / P >
< / SPAN > < / SPAN >
< / SPAN >
< / SPAN >
< SPAN class = "" > < H2 name = "Python_Style_Rules" id = "Python_Style_Rules" > Python Style Rules< / H2 >
< SPAN class = "" > < H3 > < A name = "Semicolons" id = "Semicolons" > Semicolons< / A > < / H3 >
< SPAN class = "showhide_button" onclick = "javascript:ShowHideByName('Semicolons__body','Semicolons__button')" name = "Semicolons__button" id = "Semicolons__button" > ▶< / SPAN >
< SPAN class = "" >
Do not terminate your lines with semi-colons and do not use
semi-colons to put two commands on the same line.
< / SPAN >
< SPAN class = "" > < BR > < SPAN class = "stylepoint_body" name = "Semicolons__body" id = "Semicolons__body" style = "display: none" > < SPAN class = "link_button" > < A href = "?showone=Semicolons#Semicolons" >
link
< / A > < / SPAN >
< / SPAN > < / SPAN >
< / SPAN >
< SPAN class = "" > < H3 > < A name = "Line_length" id = "Line_length" > Line length< / A > < / H3 >
< SPAN class = "showhide_button" onclick = "javascript:ShowHideByName('Line_length__body','Line_length__button')" name = "Line_length__button" id = "Line_length__button" > ▶< / SPAN >
< SPAN class = "" >
Maximum line length is < em > 80 characters< / em > .
< / SPAN >
< SPAN class = "" > < BR > < SPAN class = "stylepoint_body" name = "Line_length__body" id = "Line_length__body" style = "display: none" > < SPAN class = "link_button" > < A href = "?showone=Line_length#Line_length" >
link
< / A > < / SPAN >
< p >
Exception: lines importing modules may end up longer than 80
characters only if using Python 2.4 or
earlier.
< / p >
< p >
Make use of Python's
< a HREF = "http://www.python.org/doc/ref/implicit-joining.html" > implicit
line joining inside parentheses, brackets and braces< / a > .
If necessary, you can add an extra pair of parentheses around an
expression.
< / p >
< SPAN class = "" > < PRE > Yes: foo_bar(self, width, height, color='black', design=None, x='foo',
emphasis=None, highlight=0)
if (width == 0 and height == 0 and
color == 'red' and emphasis == 'strong'):< / PRE > < / SPAN >
< p >
When a literal string won't fit on a single line, use parentheses for
implicit line joining.
< / p >
< SPAN class = "" > < PRE >
< span class = "external" > < / span > x = ('This will build a very long long '
< span class = "external" > < / span > 'long long long long long long string')< / PRE > < / SPAN >
< p >
Make note of the indentation of the elements in the line
continuation examples above; see the
< a HREF = "#indentation" > indentation< / a >
section for explanation.
< / p >
< / SPAN > < / SPAN >
< / SPAN >
< SPAN class = "" > < H3 > < A name = "Parentheses" id = "Parentheses" > Parentheses< / A > < / H3 >
< SPAN class = "showhide_button" onclick = "javascript:ShowHideByName('Parentheses__body','Parentheses__button')" name = "Parentheses__button" id = "Parentheses__button" > ▶< / SPAN >
< SPAN class = "" >
Use parentheses sparingly.
< / SPAN >
< SPAN class = "" > < BR > < SPAN class = "stylepoint_body" name = "Parentheses__body" id = "Parentheses__body" style = "display: none" > < SPAN class = "link_button" > < A href = "?showone=Parentheses#Parentheses" >
link
< / A > < / SPAN >
< p >
Do not use them in return statements or conditional statements unless
using parentheses for implied line continuation. (See above.)
It is however fine to use parentheses around tuples.
< / p >
< SPAN class = "" > < PRE > Yes: < span class = "external" > < / span > if foo:
< span class = "external" > < / span > bar()
< span class = "external" > < / span > while x:
< span class = "external" > < / span > x = bar()
< span class = "external" > < / span > if x and y:
< span class = "external" > < / span > bar()
< span class = "external" > < / span > if not x:
< span class = "external" > < / span > bar()
< span class = "external" > < / span > return foo
< span class = "external" > < / span > for (x, y) in dict.items(): ...< / PRE > < / SPAN >
< SPAN class = "" > < PRE class = "badcode" > No: < span class = "external" > < / span > if (x):
< span class = "external" > < / span > bar()
< span class = "external" > < / span > if not(x):
< span class = "external" > < / span > bar()
< span class = "external" > < / span > return (foo)< / PRE > < / SPAN >
< / SPAN > < / SPAN >
< / SPAN >
< SPAN class = "" > < H3 > < A name = "Indentation" id = "Indentation" > Indentation< / A > < / H3 >
< SPAN class = "showhide_button" onclick = "javascript:ShowHideByName('Indentation__body','Indentation__button')" name = "Indentation__button" id = "Indentation__button" > ▶< / SPAN >
< SPAN class = "" >
Indent your code blocks with < em > 4 spaces< / em > .
< / SPAN >
< SPAN class = "" > < BR > < SPAN class = "stylepoint_body" name = "Indentation__body" id = "Indentation__body" style = "display: none" > < SPAN class = "link_button" > < A href = "?showone=Indentation#Indentation" >
link
< / A > < / SPAN >
< p >
Never use tabs or mix tabs and spaces.
In cases of implied line continuation, you should align wrapped elements
either vertically, as per the examples in the
< a HREF = "#Line_length" > line length< / a > section; or using a hanging
indent of 4 spaces, in which case there should be no argument on
the first line.
< / p >
< SPAN class = "" > < PRE > Yes: # Aligned with opening delimiter
foo = long_function_name(var_one, var_two,
var_three, var_four)
# 4-space hanging indent; nothing on first line
foo = long_function_name(
var_one, var_two, var_three,
var_four)< / PRE > < / SPAN >
< SPAN class = "" > < PRE class = "badcode" > No: < span class = "external" > < / span > # Stuff on first line forbidden
< span class = "external" > < / span > foo = long_function_name(var_one, var_two,
< span class = "external" > < / span > var_three, var_four)
< span class = "external" > < / span > # 2-space hanging indent forbidden
< span class = "external" > < / span > foo = long_function_name(
< span class = "external" > < / span > var_one, var_two, var_three,
< span class = "external" > < / span > var_four)< / PRE > < / SPAN >
< / SPAN > < / SPAN >
< / SPAN >
< SPAN class = "" > < H3 > < A name = "Blank_Lines" id = "Blank_Lines" > Blank Lines< / A > < / H3 >
< SPAN class = "showhide_button" onclick = "javascript:ShowHideByName('Blank_Lines__body','Blank_Lines__button')" name = "Blank_Lines__button" id = "Blank_Lines__button" > ▶< / SPAN >
< SPAN class = "" >
Two blank lines between top-level definitions, one blank line
between method definitions.
< / SPAN >
< SPAN class = "" > < BR > < SPAN class = "stylepoint_body" name = "Blank_Lines__body" id = "Blank_Lines__body" style = "display: none" > < SPAN class = "link_button" > < A href = "?showone=Blank_Lines#Blank_Lines" >
link
< / A > < / SPAN >
< p >
Two blank lines between top-level definitions, be they function
or class definitions. One blank line between method definitions
and between the < code > class< / code > line and the first method.
Use single blank lines as you judge appropriate within functions or
methods.
< / p >
< / SPAN > < / SPAN >
< / SPAN >
< SPAN class = "" > < H3 > < A name = "Whitespace" id = "Whitespace" > Whitespace< / A > < / H3 >
< SPAN class = "showhide_button" onclick = "javascript:ShowHideByName('Whitespace__body','Whitespace__button')" name = "Whitespace__button" id = "Whitespace__button" > ▶< / SPAN >
< SPAN class = "" >
Follow standard typographic rules for the use of spaces around
punctuation.
< / SPAN >
< SPAN class = "" > < BR > < SPAN class = "stylepoint_body" name = "Whitespace__body" id = "Whitespace__body" style = "display: none" > < SPAN class = "link_button" > < A href = "?showone=Whitespace#Whitespace" >
link
< / A > < / SPAN >
< p >
No whitespace inside parentheses, brackets or braces.
< / p >
< SPAN class = "" > < PRE > Yes: < span class = "external" > < / span > spam(ham[1], {eggs: 2}, [])< / PRE > < / SPAN >
< SPAN class = "" > < PRE class = "badcode" > No: < span class = "external" > < / span > spam( ham[ 1 ], { eggs: 2 }, [ ] )< / PRE > < / SPAN >
< p >
No whitespace before a comma, semicolon, or colon. Do use
whitespace after a comma, semicolon, or colon except at the end
of the line.
< / p >
< SPAN class = "" > < PRE > Yes: < span class = "external" > < / span > if x == 4:
< span class = "external" > < / span > print x, y
< span class = "external" > < / span > x, y = y, x< / PRE > < / SPAN >
< SPAN class = "" > < PRE class = "badcode" > No: < span class = "external" > < / span > if x == 4 :
< span class = "external" > < / span > print x , y
< span class = "external" > < / span > x , y = y , x< / PRE > < / SPAN >
< p >
No whitespace before the open paren/bracket that starts an argument list,
indexing or slicing.
< / p >
< SPAN class = "" > < PRE > Yes: < span class = "external" > < / span > spam(1)< / PRE > < / SPAN >
< SPAN class = "" > < PRE class = "badcode" > No: < span class = "external" > < / span > spam (1)< / PRE > < / SPAN >
< SPAN class = "" > < PRE > Yes: < span class = "external" > < / span > dict['key'] = list[index]< / PRE > < / SPAN >
< SPAN class = "" > < PRE class = "badcode" > No: < span class = "external" > < / span > dict ['key'] = list [index]< / PRE > < / SPAN >
< p >
Surround binary operators with a single space on either side for
assignment (< code > =< / code > ), comparisons (< code > ==, < , > , !=,
< > , < =, > =, in, not in, is, is not< / code > ), and Booleans
(< code > and, or, not< / code > ). Use your better judgment for the
insertion of spaces around arithmetic operators but always be
consistent about whitespace on either side of a binary operator.
< / p >
< SPAN class = "" > < PRE > Yes: < span class = "external" > < / span > x == 1< / PRE > < / SPAN >
< SPAN class = "" > < PRE class = "badcode" > No: < span class = "external" > < / span > x< 1< / PRE > < / SPAN >
< p >
Don't use spaces around the '=' sign when used to indicate a
keyword argument or a default parameter value.
< / p >
< SPAN class = "" > < PRE > Yes: < span class = "external" > < / span > def complex(real, imag=0.0): return magic(r=real, i=imag)< / PRE > < / SPAN >
< SPAN class = "" > < PRE class = "badcode" > No: < span class = "external" > < / span > def complex(real, imag = 0.0): return magic(r = real, i = imag)< / PRE > < / SPAN >
< p >
Don't use spaces to vertically align tokens on consecutive lines, since it
becomes a maintenance burden (applies to < code > :< / code > , < code > #< / code > ,
< code > =< / code > , etc.):
< / p >
< SPAN class = "" > < PRE > Yes:
foo = 1000 # comment
long_name = 2 # comment that should not be aligned
dictionary = {
"foo": 1,
"long_name": 2,
}< / PRE > < / SPAN >
< SPAN class = "" > < PRE class = "badcode" > No:
foo = 1000 # comment
long_name = 2 # comment that should not be aligned
dictionary = {
"foo" : 1,
"long_name": 2,
}< / PRE > < / SPAN >
< / SPAN > < / SPAN >
< / SPAN >
< SPAN class = "" > < H3 > < A name = "Python_Interpreter" id = "Python_Interpreter" > Python Interpreter< / A > < / H3 >
< SPAN class = "showhide_button" onclick = "javascript:ShowHideByName('Python_Interpreter__body','Python_Interpreter__button')" name = "Python_Interpreter__button" id = "Python_Interpreter__button" > ▶< / SPAN >
< SPAN class = "" >
Modules should begin with
< code > #!/usr/bin/env python< version> < / code >
< / SPAN >
< SPAN class = "" > < BR > < SPAN class = "stylepoint_body" name = "Python_Interpreter__body" id = "Python_Interpreter__body" style = "display: none" > < SPAN class = "link_button" > < A href = "?showone=Python_Interpreter#Python_Interpreter" >
link
< / A > < / SPAN >
< p >
Modules should begin with a "shebang" line specifying the Python
interpreter used to execute the program:
< / p >
< SPAN class = "" > < PRE >
< span class = "external" > < / span > #!/usr/bin/env python2.4< / PRE > < / SPAN >
< p >
Always use the most specific version you can use, e.g.,
< code > /usr/bin/python2.4< / code > , not
< code > /usr/bin/python2< / code > . This makes it easier to find
dependencies when
upgrading to a different Python version
and also avoids confusion and breakage during use. E.g., Does
< code > /usr/bin/python2< / code > mean Python 2.0.1 or Python
2.3.0?
< / p >
< / SPAN > < / SPAN >
< / SPAN >
< SPAN class = "" > < H3 > < A name = "Comments" id = "Comments" > Comments< / A > < / H3 >
< SPAN class = "showhide_button" onclick = "javascript:ShowHideByName('Comments__body','Comments__button')" name = "Comments__button" id = "Comments__button" > ▶< / SPAN >
< SPAN class = "" >
Be sure to use the right style for module, function, method and in-line
comments.
< / SPAN >
< SPAN class = "" > < BR > < SPAN class = "stylepoint_body" name = "Comments__body" id = "Comments__body" style = "display: none" > < SPAN class = "link_button" > < A href = "?showone=Comments#Comments" >
link
< / A > < / SPAN >
< P class = "" >
< SPAN class = "stylepoint_subsection" > Doc Strings< / SPAN >
< p >
Python has a unique commenting style using doc strings. A doc
string is a string that is the first statement in a package,
module, class or function. These strings can be extracted
automatically through the < code > __doc__< / code > member of the
object and are used by < code > pydoc< / code > . (Try running
< code > pydoc< / code > on your module to see how it looks.) Our
convention for doc strings is to use the three double-quote
format for strings. A doc string should be organized as a
summary line (one physical line) terminated by a period,
question mark, or exclamation point, followed by a blank line,
followed by the rest of the doc string starting at the same
cursor position as the first quote of the first line. There are
more formatting guidelines for doc strings below.
< / p >
< / P >
< P class = "" >
< SPAN class = "stylepoint_subsection" > Modules< / SPAN >
< p >
Every file should contain the following items, in order:
< ul >
< li > a copyright statement (for example,
< code > Copyright 2008 Google Inc.< / code > )< / li >
< li > a license boilerplate. Choose the appropriate boilerplate
for the license used by the project (for example, Apache 2.0, BSD,
LGPL, GPL)< / li >
< li > an author line to identify the original author of the file< / li >
< / ul >
< / p >
< / P >
< P class = "" >
< SPAN class = "stylepoint_subsection" > Functions and Methods< / SPAN >
< p >
Any function or method which is not both obvious and very short
needs a doc string. Additionally, any externally accessible
function or method regardless of length or simplicity needs a
doc string. The doc string should include what the function does
and have detailed descriptions of the input and output. It
should not, generally, describe how it does it unless it's some
complicated algorithm. For tricky code block/inline comments
within the code are more appropriate. The doc string should give
enough information to write a call to the function without
looking at a single line of the function's code. Args should be
individually documented, an explanation following after a colon,
and should use a uniform hanging indent of 2 or 4 spaces. The
doc string should specify the expected types where specific types
are required. A "Raises:" section should list all exceptions
that can be raised by the function. The doc string for generator
functions should use "Yields:" rather than "Returns:".
< / p >
< SPAN class = "" > < PRE >
< span class = "external" > < / span > def fetch_bigtable_rows(big_table, keys, other_silly_variable=None):
< span class = "external" > < / span > """Fetches rows from a Bigtable.
< span class = "external" > < / span > Retrieves rows pertaining to the given keys from the Table instance
< span class = "external" > < / span > represented by big_table. Silly things may happen if
< span class = "external" > < / span > other_silly_variable is not None.
< span class = "external" > < / span > Args:
< span class = "external" > < / span > big_table: An open Bigtable Table instance.
< span class = "external" > < / span > keys: A sequence of strings representing the key of each table row
< span class = "external" > < / span > to fetch.
< span class = "external" > < / span > other_silly_variable: Another optional variable, that has a much
< span class = "external" > < / span > longer name than the other args, and which does nothing.
< span class = "external" > < / span > Returns:
< span class = "external" > < / span > A dict mapping keys to the corresponding table row data
< span class = "external" > < / span > fetched. Each row is represented as a tuple of strings. For
< span class = "external" > < / span > example:
< span class = "external" > < / span > {'Serak': ('Rigel VII', 'Preparer'),
< span class = "external" > < / span > 'Zim': ('Irk', 'Invader'),
< span class = "external" > < / span > 'Lrrr': ('Omicron Persei 8', 'Emperor')}
< span class = "external" > < / span > If a key from the keys argument is missing from the dictionary,
< span class = "external" > < / span > then that row was not found in the table.
< span class = "external" > < / span > Raises:
< span class = "external" > < / span > IOError: An error occurred accessing the bigtable.Table object.
< span class = "external" > < / span > """
< span class = "external" > < / span > pass
< span class = "external" > < / span >
< / PRE > < / SPAN >
< / P >
< P class = "" >
< SPAN class = "stylepoint_subsection" > Classes< / SPAN >
< p >
Classes should have a doc string below the class definition describing
the class. If your class has public attributes, they should be documented
here in an Attributes section and follow the same formatting as a
function's Args section.
< / p >
< SPAN class = "" > < PRE >
< span class = "external" > < / span > class SampleClass(object):
< span class = "external" > < / span > """Summary of class here.
< span class = "external" > < / span > Longer class information....
< span class = "external" > < / span > Longer class information....
< span class = "external" > < / span > Attributes:
< span class = "external" > < / span > likes_spam: A boolean indicating if we like SPAM or not.
< span class = "external" > < / span > eggs: An integer count of the eggs we have laid.
< span class = "external" > < / span > """
< span class = "external" > < / span > def __init__(self, likes_spam=False):
< span class = "external" > < / span > """Inits SampleClass with blah."""
< span class = "external" > < / span > self.likes_spam = likes_spam
< span class = "external" > < / span > self.eggs = 0
< span class = "external" > < / span > def public_method(self):
< span class = "external" > < / span > """Performs operation blah."""
< span class = "external" > < / span >
< / PRE > < / SPAN >
< / P >
< P class = "" >
< SPAN class = "stylepoint_subsection" > Block and Inline Comments< / SPAN >
< p >
The final place to have comments is in tricky parts of the
code. If you're going to have to explain it at the next
< a HREF = "http://en.wikipedia.org/wiki/Code_review" > code review< / a > ,
you should comment it now. Complicated operations get a few lines of
comments before the operations
commence. Non-obvious ones get comments at the end of the line.
< / p >
< SPAN class = "" > < PRE >
< span class = "external" > < / span > # We use a weighted dictionary search to find out where i is in
< span class = "external" > < / span > # the array. We extrapolate position based on the largest num
< span class = "external" > < / span > # in the array and the array size and then do binary search to
< span class = "external" > < / span > # get the exact number.
< span class = "external" > < / span > if i & (i-1) == 0: # true iff i is a power of 2
< span class = "external" > < / span >
< / PRE > < / SPAN >
< p >
To improve legibility, these comments should be at least 2 spaces away
from the code.
< / p >
< p >
On the other hand, never describe the code. Assume the person
reading the code knows Python (though not what you're trying to
do) better than you do.
< / p >
< SPAN class = "" > < PRE class = "badcode" >
< span class = "external" > < / span > # BAD COMMENT: Now go through the b array and make sure whenever i occurs
< span class = "external" > < / span > # the next element is i+1
< span class = "external" > < / span >
< / PRE > < / SPAN >
< / P >
< / SPAN > < / SPAN >
< / SPAN >
< SPAN class = "" > < H3 > < A name = "Classes" id = "Classes" > Classes< / A > < / H3 >
< SPAN class = "showhide_button" onclick = "javascript:ShowHideByName('Classes__body','Classes__button')" name = "Classes__button" id = "Classes__button" > ▶< / SPAN >
< SPAN class = "" >
If a class inherits from no other base classes, explicitly inherit
from < code > object< / code > . This also applies to nested classes.
< / SPAN >
< SPAN class = "" > < BR > < SPAN class = "stylepoint_body" name = "Classes__body" id = "Classes__body" style = "display: none" > < SPAN class = "link_button" > < A href = "?showone=Classes#Classes" >
link
< / A > < / SPAN >
< SPAN class = "" > < PRE class = "badcode" > No: < span class = "external" > < / span > class SampleClass:
< span class = "external" > < / span > pass
< span class = "external" > < / span > class OuterClass:
< span class = "external" > < / span > class InnerClass:
< span class = "external" > < / span > pass
< span class = "external" > < / span >
< / PRE > < / SPAN >
< SPAN class = "" > < PRE > Yes: < span class = "external" > < / span > class SampleClass(object):
< span class = "external" > < / span > pass
< span class = "external" > < / span > class OuterClass(object):
< span class = "external" > < / span > class InnerClass(object):
< span class = "external" > < / span > pass
< span class = "external" > < / span > class ChildClass(ParentClass):
< span class = "external" > < / span > """Explicitly inherits from another class already."""
< span class = "external" > < / span >
< / PRE > < / SPAN >
< p > Inheriting from < code > object< / code > is needed to make properties work
properly, and it will protect your code from one particular potential
incompatibility with Python 3000. It also defines
special methods that implement the default semantics of objects including
< code > __new__< / code > , < code > __init__< / code > , < code > __delattr__< / code > ,
< code > __getattribute__< / code > , < code > __setattr__< / code > ,
< code > __hash__< / code > , < code > __repr__< / code > , and < code > __str__< / code > .
< / p >
< / SPAN > < / SPAN >
< / SPAN >
< SPAN class = "" > < H3 > < A name = "Strings" id = "Strings" > Strings< / A > < / H3 >
< SPAN class = "showhide_button" onclick = "javascript:ShowHideByName('Strings__body','Strings__button')" name = "Strings__button" id = "Strings__button" > ▶< / SPAN >
< SPAN class = "" >
Use the < code > %< / code > operator for formatting strings,
even when the parameters are all strings. Use your best judgement
to decide between < code > +< / code > and < code > %< / code > though.
< / SPAN >
< SPAN class = "" > < BR > < SPAN class = "stylepoint_body" name = "Strings__body" id = "Strings__body" style = "display: none" > < SPAN class = "link_button" > < A href = "?showone=Strings#Strings" >
link
< / A > < / SPAN >
< SPAN class = "" > < PRE class = "badcode" > No: < span class = "external" > < / span > x = '%s%s' % (a, b) # use + in this case
< span class = "external" > < / span > x = imperative + ', ' + expletive + '!'
< span class = "external" > < / span > x = 'name: ' + name + '; score: ' + str(n)< / PRE > < / SPAN >
< SPAN class = "" > < PRE > Yes: < span class = "external" > < / span > x = a + b
< span class = "external" > < / span > x = '%s, %s!' % (imperative, expletive)
< span class = "external" > < / span > x = 'name: %s; score: %d' % (name, n)< / PRE > < / SPAN >
< p >
Avoid using the < code > +< / code > and < code > +=< / code > operators to
accumulate a string within a loop. Since strings are immutable, this
creates unnecessary temporary objects and results in quadratic rather
than linear running time. Instead, add each substring to a list and
< code > ''.join< / code > the list after the loop terminates (or, write each
substring to a < code > cStringIO.StringIO< / code > buffer).
< / p >
< SPAN class = "" > < PRE class = "badcode" > No: < span class = "external" > < / span > employee_table = '< table> '
< span class = "external" > < / span > for last_name, first_name in employee_list:
< span class = "external" > < / span > employee_table += '< tr> < td> %s, %s< /td> < /tr> ' % (last_name, first_name)
< span class = "external" > < / span > employee_table += '< /table> '< / PRE > < / SPAN >
< SPAN class = "" > < PRE > Yes: < span class = "external" > < / span > items = ['< table> ']
< span class = "external" > < / span > for last_name, first_name in employee_list:
< span class = "external" > < / span > items.append('< tr> < td> %s, %s< /td> < /tr> ' % (last_name, first_name))
< span class = "external" > < / span > items.append('< /table> ')
< span class = "external" > < / span > employee_table = ''.join(items)< / PRE > < / SPAN >
< p >
Use < code > """< / code > for multi-line strings rather than
< code > '''< / code > . Note, however, that it is often cleaner to
use implicit line joining since multi-line strings do
not flow with the indentation of the rest of the program:
< / p >
< SPAN class = "" > < PRE class = "badcode" > No< span class = "external" > < / span > :
< span class = "external" > < / span > print """This is pretty ugly.
Don'< span class = "external" > < / span > t do this.
"""< span class = "external" > < / span >
< / PRE > < / SPAN >
< SPAN class = "" > < PRE > Ye< span class = "external" > < / span > s:
< span class = "external" > < / span > print ("This is much nicer.\n"
< span class = "external" > < / span > "Do it this way.\n")< / PRE > < / SPAN >
< / SPAN > < / SPAN >
< / SPAN >
< SPAN class = "" > < H3 > < A name = "TODO_Comments" id = "TODO_Comments" > TODO Comments< / A > < / H3 >
< SPAN class = "showhide_button" onclick = "javascript:ShowHideByName('TODO_Comments__body','TODO_Comments__button')" name = "TODO_Comments__button" id = "TODO_Comments__button" > ▶< / SPAN >
< SPAN class = "" >
Use < code > TODO< / code > comments for code that is temporary, a
short-term solution, or good-enough but not perfect.
< / SPAN >
< SPAN class = "" > < BR > < SPAN class = "stylepoint_body" name = "TODO_Comments__body" id = "TODO_Comments__body" style = "display: none" > < SPAN class = "link_button" > < A href = "?showone=TODO_Comments#TODO_Comments" >
link
< / A > < / SPAN >
< p >
< code > TODO< / code > s should include the string < code > TODO< / code > in
all caps, followed by your
name, e-mail address, or other
identifier
in parentheses. A colon is optional. A comment explaining what there
is to do is required. The main purpose is to have
a consistent < code > TODO< / code > format searchable by the person
adding the comment (who can provide more details upon request). A
< code > TODO< / code > is not a commitment to provide the fix yourself.
< / p >
< SPAN class = "" > < PRE > # TODO(kl@gmail.com): Drop the use of "has_key".
# TODO(Zeke) change this to use relations.< / PRE > < / SPAN >
< p >
If your < code > TODO< / code > is of the form "At a future date do
something" make sure that you either include a very specific
date ("Fix by November 2009") or a very specific event
("Remove this code when all clients can handle XML responses.").
< / p >
< / SPAN > < / SPAN >
< / SPAN >
< SPAN class = "" > < H3 > < A name = "Imports_formatting" id = "Imports_formatting" > Imports formatting< / A > < / H3 >
< SPAN class = "showhide_button" onclick = "javascript:ShowHideByName('Imports_formatting__body','Imports_formatting__button')" name = "Imports_formatting__button" id = "Imports_formatting__button" > ▶< / SPAN >
< SPAN class = "" >
Imports should be on separate lines.
< / SPAN >
< SPAN class = "" > < BR > < SPAN class = "stylepoint_body" name = "Imports_formatting__body" id = "Imports_formatting__body" style = "display: none" > < SPAN class = "link_button" > < A href = "?showone=Imports_formatting#Imports_formatting" >
link
< / A > < / SPAN >
< p >
E.g.:
< / p >
< SPAN class = "" > < PRE > Yes: < span class = "external" > < / span > import os
< span class = "external" > < / span > import sys< / PRE > < / SPAN >
< SPAN class = "" > < PRE class = "badcode" > No: < span class = "external" > < / span > import os, sys< / PRE > < / SPAN >
< p >
Imports are always put at the top of the file, just after any
module comments and doc strings and before module globals and
constants. Imports should be grouped with the order being most generic
to least generic:
< / p >
< ul >
< li > standard library imports< / li >
< li > third-party imports< / li >
< li > application-specific imports< / li >
< / ul >
< p >
Within each grouping, imports should be sorted lexicographically,
ignoring case, according to each module's full package path.
< / p >
< SPAN class = "" > < PRE >
< span class = "external" > < / span > import foo
< span class = "external" > < / span > from foo import bar
< span class = "external" > < / span > from foo.bar import baz
< span class = "external" > < / span > from foo.bar import Quux
< span class = "external" > < / span > from Foob import ar< / PRE > < / SPAN >
< / SPAN > < / SPAN >
< / SPAN >
< SPAN class = "" > < H3 > < A name = "Statements" id = "Statements" > Statements< / A > < / H3 >
< SPAN class = "showhide_button" onclick = "javascript:ShowHideByName('Statements__body','Statements__button')" name = "Statements__button" id = "Statements__button" > ▶< / SPAN >
< SPAN class = "" >
Generally only one statement per line.
< / SPAN >
< SPAN class = "" > < BR > < SPAN class = "stylepoint_body" name = "Statements__body" id = "Statements__body" style = "display: none" > < SPAN class = "link_button" > < A href = "?showone=Statements#Statements" >
link
< / A > < / SPAN >
< p >
However, you may put the
result of a test on the same line as the test only if the entire
statement fits on one line. In particular, you can never do so
with < code > try< / code > /< code > except< / code > since the
< code > try< / code > and < code > except< / code > can't both fit on the
same line, and you can only do so with an < code > if< / code > if
there is no < code > else< / code > .
< / p >
< SPAN class = "" > < PRE > Ye< span class = "external" > < / span > s:
< span class = "external" > < / span > if foo: bar(foo)< / PRE > < / SPAN >
< SPAN class = "" > < PRE class = "badcode" > No< span class = "external" > < / span > :
< span class = "external" > < / span > if foo: bar(foo)
< span class = "external" > < / span > else: baz(foo)
< span class = "external" > < / span > try: bar(foo)
< span class = "external" > < / span > except ValueError: baz(foo)
< span class = "external" > < / span > try:
< span class = "external" > < / span > bar(foo)
< span class = "external" > < / span > except ValueError: baz(foo)
< span class = "external" > < / span >
< / PRE > < / SPAN >
< / SPAN > < / SPAN >
< / SPAN >
< SPAN class = "" > < H3 > < A name = "Access_Control" id = "Access_Control" > Access Control< / A > < / H3 >
< SPAN class = "showhide_button" onclick = "javascript:ShowHideByName('Access_Control__body','Access_Control__button')" name = "Access_Control__button" id = "Access_Control__button" > ▶< / SPAN >
< SPAN class = "" >
If an accessor function would be trivial you should use public variables
instead of accessor functions to avoid the extra cost of function
calls in Python. When more functionality is added you can use
< code > property< / code > to keep the syntax consistent.
< / SPAN >
< SPAN class = "" > < BR > < SPAN class = "stylepoint_body" name = "Access_Control__body" id = "Access_Control__body" style = "display: none" > < SPAN class = "link_button" > < A href = "?showone=Access_Control#Access_Control" >
link
< / A > < / SPAN >
< p >
On the other hand, if access is more complex, or the cost of accessing
the variable is significant, you should use function calls (following the
< a HREF = "#naming" > Naming< / a > guidelines) such as < code > get_foo()< / code >
and < code > set_foo()< / code > . If the past behavior allowed access through a
property, do not bind the new accessor functions to the property. Any
code still attempting to access the variable by the old method should
break visibly so they are made aware of the change in complexity.
< / p >
< / SPAN > < / SPAN >
< / SPAN >
< SPAN class = "" > < H3 > < A name = "Naming" id = "Naming" > Naming< / A > < / H3 >
< SPAN class = "showhide_button" onclick = "javascript:ShowHideByName('Naming__body','Naming__button')" name = "Naming__button" id = "Naming__button" > ▶< / SPAN >
< SPAN class = "" >
< code > module_name, package_name, ClassName, method_name, ExceptionName,
function_name, GLOBAL_VAR_NAME, instance_var_name,
function_parameter_name, local_var_name.< / code >
< / SPAN >
< SPAN class = "" > < BR > < SPAN class = "stylepoint_body" name = "Naming__body" id = "Naming__body" style = "display: none" > < SPAN class = "link_button" > < A href = "?showone=Naming#Naming" >
link
< / A > < / SPAN >
< P class = "" >
< SPAN class = "stylepoint_subsection" > Names to Avoid< / SPAN >
< ul >
< li > single character names except for counters or iterators< / li >
< li > dashes (< code > -< / code > ) in any package/module name< / li >
< li >
< code > __double_leading_and_trailing_underscore__< / code > names
(reserved by Python)< / li >
< / ul >
< / P >
< P class = "" >
< SPAN class = "stylepoint_subsection" > Naming Convention< / SPAN >
< ul >
< li >
"Internal" means internal to a module or protected
or private within a class.< / li >
< li >
Prepending a single underscore (< code > _< / code > ) has some
support for protecting module variables and functions (not included
with < code > import * from< / code > ). Prepending a double underscore
(< code > __< / code > ) to an instance variable or method
effectively serves to make the variable or method private to its class
(using name mangling).< / li >
< li >
Place related classes and top-level functions together in a
module. Unlike Java,
there is no need to limit yourself to one class per module.< / li >
< li >
Use CapWords for class names, but lower_with_under.py for module names.
Although there are many existing modules named CapWords.py, this is now
discouraged because it's confusing when the module happens to be
named after a class. ("wait -- did I write
< code > import StringIO< / code > or < code > from StringIO import
StringIO< / code > ?")< / li >
< / ul >
< / P >
< P class = "" >
< SPAN class = "stylepoint_subsection" > Guidelines derived from Guido's Recommendations< / SPAN >
< table rules = "all" border = "1" cellspacing = "2" cellpadding = "2" >
< tr >
< th > Type< / th >
< th > Public< / th >
< th > Internal< / th >
< / tr >
< tr >
< td > Packages< / td >
< td > < code > lower_with_under< / code > < / td >
< td > < / td >
< / tr >
< tr >
< td > Modules< / td >
< td > < code > lower_with_under< / code > < / td >
< td > < code > _lower_with_under< / code > < / td >
< / tr >
< tr >
< td > Classes< / td >
< td > < code > CapWords< / code > < / td >
< td > < code > _CapWords< / code > < / td >
< / tr >
< tr >
< td > Exceptions< / td >
< td > < code > CapWords< / code > < / td >
< td > < / td >
< / tr >
< tr >
< td > Functions< / td >
< td > < code > lower_with_under()< / code > < / td >
< td > < code > _lower_with_under()< / code > < / td >
< / tr >
< tr >
< td > Global/Class Constants< / td >
< td > < code > CAPS_WITH_UNDER< / code > < / td >
< td > < code > _CAPS_WITH_UNDER< / code > < / td >
< / tr >
< tr >
< td > Global/Class Variables< / td >
< td > < code > lower_with_under< / code > < / td >
< td > < code > _lower_with_under< / code > < / td >
< / tr >
< tr >
< td > Instance Variables< / td >
< td > < code > lower_with_under< / code > < / td >
< td > < code > _lower_with_under (protected) or __lower_with_under (private)< / code > < / td >
< / tr >
< tr >
< td > Method Names< / td >
< td > < code > lower_with_under()< / code > < / td >
< td > < code > _lower_with_under() (protected) or __lower_with_under() (private)< / code > < / td >
< / tr >
< tr >
< td > Function/Method Parameters< / td >
< td > < code > lower_with_under< / code > < / td >
< td > < / td >
< / tr >
< tr >
< td > Local Variables< / td >
< td > < code > lower_with_under< / code > < / td >
< td > < / td >
< / tr >
< / table >
< / P >
< / SPAN > < / SPAN >
< / SPAN >
< SPAN class = "" > < H3 > < A name = "Main" id = "Main" > Main< / A > < / H3 >
< SPAN class = "showhide_button" onclick = "javascript:ShowHideByName('Main__body','Main__button')" name = "Main__button" id = "Main__button" > ▶< / SPAN >
< SPAN class = "" >
Even a file meant to be used as a script should be importable and a
mere import should not have the side effect of executing the script's
main functionality. The main functionality should be in a main()
function.
< / SPAN >
< SPAN class = "" > < BR > < SPAN class = "stylepoint_body" name = "Main__body" id = "Main__body" style = "display: none" > < SPAN class = "link_button" > < A href = "?showone=Main#Main" >
link
< / A > < / SPAN >
< p >
In Python,
< code > pychecker< / code > , < code > pydoc< / code > , and unit tests
require modules to be importable. Your code should always check
< code > if __name__ == '__main__'< / code > before executing your
main program so that the main program is not executed when the
module is imported.
< / p >
< SPAN class = "" > < PRE >
< span class = "external" > < / span > def main():
< span class = "external" > < / span > ...
< span class = "external" > < / span > if __name__ == '__main__':
< span class = "external" > < / span > main()
< span class = "external" > < / span >
< / PRE > < / SPAN >
< p >
All code at the top level will be executed when the module is
imported. Be careful not to call functions, create objects, or
perform other operations that should not be executed when the
file is being < code > pycheck< / code > ed or < code > pydoc< / code > ed.
< / p >
< / SPAN > < / SPAN >
< / SPAN >
< / SPAN >
< H2 > Parting Words< / H2 >
< p >
< em > BE CONSISTENT< / em > .
< / p >
< p >
If you're editing code, take a few minutes to look at the code
around you and determine its style. If they use spaces around
all their arithmetic operators, you should too. If their
comments have little boxes of hash marks around them, make your
comments have little boxes of hash marks around them too.
< / p >
< p >
The point of having style guidelines is to have a common vocabulary
of coding so people can concentrate on what you're saying rather
than on how you're saying it. We present global style rules here so
people know the vocabulary, but local style is also important. If
code you add to a file looks drastically different from the existing
code around it, it throws readers out of their rhythm when they go to
read it. Avoid this.
< / p >
< p align = "right" >
2009-10-24 05:01:49 +08:00
Revision 2.14
2009-07-24 04:09:56 +08:00
< / p >
< address >
Amit Patel< br >
Antoine Picard< br >
Eugene Jhong< br >
Jeremy Hylton< br >
Matt Smart< br >
Mike Shields< br >
< / address >
< / BODY >
< / HTML >