4.More Control Flow Tools

Besides thewhile statement just introduced, Python uses the usualflow control statements known from other languages, with some twists.

4.1.if Statements

Perhaps the most well-known statement type is theif statement. Forexample:

>>>x=int(raw_input("Please enter an integer: "))Please enter an integer: 42>>>ifx<0:...x=0...print'Negative changed to zero'...elifx==0:...print'Zero'...elifx==1:...print'Single'...else:...print'More'...More

There can be zero or moreelif parts, and theelse part isoptional. The keyword ‘elif’ is short for ‘else if’, and is usefulto avoid excessive indentation. Anif …elif …elif … sequence is a substitute for theswitch orcase statements found in other languages.

4.2.for Statements

Thefor statement in Python differs a bit from what you may be usedto in C or Pascal. Rather than always iterating over an arithmetic progressionof numbers (like in Pascal), or giving the user the ability to define both theiteration step and halting condition (as C), Python’sfor statementiterates over the items of any sequence (a list or a string), in the order thatthey appear in the sequence. For example (no pun intended):

>>># Measure some strings:...words=['cat','window','defenestrate']>>>forwinwords:...printw,len(w)...cat 3window 6defenestrate 12

If you need to modify the sequence you are iterating over while inside the loop(for example to duplicate selected items), it is recommended that you firstmake a copy. Iterating over a sequence does not implicitly make a copy. Theslice notation makes this especially convenient:

>>>forwinwords[:]:# Loop over a slice copy of the entire list....iflen(w)>6:...words.insert(0,w)...>>>words['defenestrate', 'cat', 'window', 'defenestrate']

4.3.Therange() Function

If you do need to iterate over a sequence of numbers, the built-in functionrange() comes in handy. It generates lists containing arithmeticprogressions:

>>>range(10)[0, 1, 2, 3, 4, 5, 6, 7, 8, 9]

The given end point is never part of the generated list;range(10) generatesa list of 10 values, the legal indices for items of a sequence of length 10. Itis possible to let the range start at another number, or to specify a differentincrement (even negative; sometimes this is called the ‘step’):

>>>range(5,10)[5, 6, 7, 8, 9]>>>range(0,10,3)[0, 3, 6, 9]>>>range(-10,-100,-30)[-10, -40, -70]

To iterate over the indices of a sequence, you can combinerange() andlen() as follows:

>>>a=['Mary','had','a','little','lamb']>>>foriinrange(len(a)):...printi,a[i]...0 Mary1 had2 a3 little4 lamb

In most such cases, however, it is convenient to use theenumerate()function, seeLooping Techniques.

4.4.break andcontinue Statements, andelse Clauses on Loops

Thebreak statement, like in C, breaks out of the innermost enclosingfor orwhile loop.

Loop statements may have anelse clause; it is executed when the loopterminates through exhaustion of the list (withfor) or when thecondition becomes false (withwhile), but not when the loop isterminated by abreak statement. This is exemplified by thefollowing loop, which searches for prime numbers:

>>>forninrange(2,10):...forxinrange(2,n):...ifn%x==0:...printn,'equals',x,'*',n/x...break...else:...# loop fell through without finding a factor...printn,'is a prime number'...2 is a prime number3 is a prime number4 equals 2 * 25 is a prime number6 equals 2 * 37 is a prime number8 equals 2 * 49 equals 3 * 3

(Yes, this is the correct code. Look closely: theelse clause belongs tothefor loop,not theif statement.)

When used with a loop, theelse clause has more in common with theelse clause of atry statement than it does that ofif statements: atry statement’selse clause runswhen no exception occurs, and a loop’selse clause runs when nobreakoccurs. For more on thetry statement and exceptions, seeHandling Exceptions.

Thecontinue statement, also borrowed from C, continues with the nextiteration of the loop:

>>>fornuminrange(2,10):...ifnum%2==0:...print"Found an even number",num...continue...print"Found a number",numFound an even number 2Found a number 3Found an even number 4Found a number 5Found an even number 6Found a number 7Found an even number 8Found a number 9

4.5.pass Statements

Thepass statement does nothing. It can be used when a statement isrequired syntactically but the program requires no action. For example:

>>>whileTrue:...pass# Busy-wait for keyboard interrupt (Ctrl+C)...

This is commonly used for creating minimal classes:

>>>classMyEmptyClass:...pass...

Another placepass can be used is as a place-holder for a function orconditional body when you are working on new code, allowing you to keep thinkingat a more abstract level. Thepass is silently ignored:

>>>definitlog(*args):...pass# Remember to implement this!...

4.6.Defining Functions

We can create a function that writes the Fibonacci series to an arbitraryboundary:

>>>deffib(n):# write Fibonacci series up to n..."""Print a Fibonacci series up to n."""...a,b=0,1...whilea<n:...printa,...a,b=b,a+b...>>># Now call the function we just defined:...fib(2000)0 1 1 2 3 5 8 13 21 34 55 89 144 233 377 610 987 1597

The keyworddef introduces a functiondefinition. It must befollowed by the function name and the parenthesized list of formal parameters.The statements that form the body of the function start at the next line, andmust be indented.

The first statement of the function body can optionally be a string literal;this string literal is the function’s documentation string, ordocstring.(More about docstrings can be found in the sectionDocumentation Strings.)There are tools which use docstrings to automatically produce online or printeddocumentation, or to let the user interactively browse through code; it’s goodpractice to include docstrings in code that you write, so make a habit of it.

Theexecution of a function introduces a new symbol table used for the localvariables of the function. More precisely, all variable assignments in afunction store the value in the local symbol table; whereas variable referencesfirst look in the local symbol table, then in the local symbol tables ofenclosing functions, then in the global symbol table, and finally in the tableof built-in names. Thus, global variables cannot be directly assigned a valuewithin a function (unless named in aglobal statement), although theymay be referenced.

The actual parameters (arguments) to a function call are introduced in the localsymbol table of the called function when it is called; thus, arguments arepassed usingcall by value (where thevalue is always an objectreference,not the value of the object).1 When a function calls another function, a newlocal symbol table is created for that call.

A function definition introduces the function name in the current symbol table.The value of the function name has a type that is recognized by the interpreteras a user-defined function. This value can be assigned to another name whichcan then also be used as a function. This serves as a general renamingmechanism:

>>>fib<function fib at 10042ed0>>>>f=fib>>>f(100)0 1 1 2 3 5 8 13 21 34 55 89

Coming from other languages, you might object thatfib is not a function buta procedure since it doesn’t return a value. In fact, even functions without areturn statement do return a value, albeit a rather boring one. Thisvalue is calledNone (it’s a built-in name). Writing the valueNone isnormally suppressed by the interpreter if it would be the only value written.You can see it if you really want to usingprint:

>>>fib(0)>>>printfib(0)None

It is simple to write a function that returns a list of the numbers of theFibonacci series, instead of printing it:

>>>deffib2(n):# return Fibonacci series up to n..."""Return a list containing the Fibonacci series up to n."""...result=[]...a,b=0,1...whilea<n:...result.append(a)# see below...a,b=b,a+b...returnresult...>>>f100=fib2(100)# call it>>>f100# write the result[0, 1, 1, 2, 3, 5, 8, 13, 21, 34, 55, 89]

This example, as usual, demonstrates some new Python features:

  • Thereturn statement returns with a value from a function.return without an expression argument returnsNone. Falling offthe end of a function also returnsNone.

  • The statementresult.append(a) calls amethod of the list objectresult. A method is a function that ‘belongs’ to an object and is namedobj.methodname, whereobj is some object (this may be an expression),andmethodname is the name of a method that is defined by the object’s type.Different types define different methods. Methods of different types may havethe same name without causing ambiguity. (It is possible to define your ownobject types and methods, usingclasses, seeClasses)The methodappend() shown in the example is defined for list objects; itadds a new element at the end of the list. In this example it is equivalent toresult=result+[a], but more efficient.

4.7.More on Defining Functions

It is also possible to define functions with a variable number of arguments.There are three forms, which can be combined.

4.7.1.Default Argument Values

The most useful form is to specify a default value for one or more arguments.This creates a function that can be called with fewer arguments than it isdefined to allow. For example:

defask_ok(prompt,retries=4,complaint='Yes or no, please!'):whileTrue:ok=raw_input(prompt)ifokin('y','ye','yes'):returnTrueifokin('n','no','nop','nope'):returnFalseretries=retries-1ifretries<0:raiseIOError('refusenik user')printcomplaint

This function can be called in several ways:

  • giving only the mandatory argument:ask_ok('Doyoureallywanttoquit?')

  • giving one of the optional arguments:ask_ok('OKtooverwritethefile?',2)

  • or even giving all arguments:ask_ok('OKtooverwritethefile?',2,'Comeon,onlyyesorno!')

This example also introduces thein keyword. This tests whether ornot a sequence contains a certain value.

The default values are evaluated at the point of function definition in thedefining scope, so that

i=5deff(arg=i):printargi=6f()

will print5.

Important warning: The default value is evaluated only once. This makes adifference when the default is a mutable object such as a list, dictionary, orinstances of most classes. For example, the following function accumulates thearguments passed to it on subsequent calls:

deff(a,L=[]):L.append(a)returnLprintf(1)printf(2)printf(3)

This will print

[1][1,2][1,2,3]

If you don’t want the default to be shared between subsequent calls, you canwrite the function like this instead:

deff(a,L=None):ifLisNone:L=[]L.append(a)returnL

4.7.2.Keyword Arguments

Functions can also be called usingkeyword argumentsof the formkwarg=value. For instance, the following function:

defparrot(voltage,state='a stiff',action='voom',type='Norwegian Blue'):print"-- This parrot wouldn't",action,print"if you put",voltage,"volts through it."print"-- Lovely plumage, the",typeprint"-- It's",state,"!"

accepts one required argument (voltage) and three optional arguments(state,action, andtype). This function can be called in anyof the following ways:

parrot(1000)# 1 positional argumentparrot(voltage=1000)# 1 keyword argumentparrot(voltage=1000000,action='VOOOOOM')# 2 keyword argumentsparrot(action='VOOOOOM',voltage=1000000)# 2 keyword argumentsparrot('a million','bereft of life','jump')# 3 positional argumentsparrot('a thousand',state='pushing up the daisies')# 1 positional, 1 keyword

but all the following calls would be invalid:

parrot()# required argument missingparrot(voltage=5.0,'dead')# non-keyword argument after a keyword argumentparrot(110,voltage=220)# duplicate value for the same argumentparrot(actor='John Cleese')# unknown keyword argument

In a function call, keyword arguments must follow positional arguments.All the keyword arguments passed must match one of the argumentsaccepted by the function (e.g.actor is not a valid argument for theparrot function), and their order is not important. This also includesnon-optional arguments (e.g.parrot(voltage=1000) is valid too).No argument may receive a value more than once.Here’s an example that fails due to this restriction:

>>>deffunction(a):...pass...>>>function(0,a=0)Traceback (most recent call last):  File"<stdin>", line1, in<module>TypeError:function() got multiple values for keyword argument 'a'

When a final formal parameter of the form**name is present, it receives adictionary (seeMapping Types — dict) containing all keyword arguments except forthose corresponding to a formal parameter. This may be combined with a formalparameter of the form*name (described in the next subsection) whichreceives a tuple containing the positional arguments beyond the formal parameterlist. (*name must occur before**name.) For example, if we define afunction like this:

defcheeseshop(kind,*arguments,**keywords):print"-- Do you have any",kind,"?"print"-- I'm sorry, we're all out of",kindforarginarguments:printargprint"-"*40keys=sorted(keywords.keys())forkwinkeys:printkw,":",keywords[kw]

It could be called like this:

cheeseshop("Limburger","It's very runny, sir.","It's really very, VERY runny, sir.",shopkeeper='Michael Palin',client="John Cleese",sketch="Cheese Shop Sketch")

and of course it would print:

-- Do you have any Limburger ?-- I'm sorry, we're all out of LimburgerIt's very runny, sir.It's really very, VERY runny, sir.----------------------------------------client : John Cleeseshopkeeper : Michael Palinsketch : Cheese Shop Sketch

Note that the list of keyword argument names is created by sorting the resultof the keywords dictionary’skeys() method before printing its contents;if this is not done, the order in which the arguments are printed is undefined.

4.7.3.Arbitrary Argument Lists

Finally, the least frequently used option is to specify that a function can becalled with an arbitrary number of arguments. These arguments will be wrappedup in a tuple (seeTuples and Sequences). Before the variable number of arguments,zero or more normal arguments may occur.

defwrite_multiple_items(file,separator,*args):file.write(separator.join(args))

4.7.4.Unpacking Argument Lists

The reverse situation occurs when the arguments are already in a list or tuplebut need to be unpacked for a function call requiring separate positionalarguments. For instance, the built-inrange() function expects separatestart andstop arguments. If they are not available separately, write thefunction call with the*-operator to unpack the arguments out of a listor tuple:

>>>range(3,6)# normal call with separate arguments[3, 4, 5]>>>args=[3,6]>>>range(*args)# call with arguments unpacked from a list[3, 4, 5]

In the same fashion, dictionaries can deliver keyword arguments with the**-operator:

>>>defparrot(voltage,state='a stiff',action='voom'):...print"-- This parrot wouldn't",action,...print"if you put",voltage,"volts through it.",...print"E's",state,"!"...>>>d={"voltage":"four million","state":"bleedin' demised","action":"VOOM"}>>>parrot(**d)-- This parrot wouldn't VOOM if you put four million volts through it. E's bleedin' demised !

4.7.5.Lambda Expressions

Small anonymous functions can be created with thelambda keyword.This function returns the sum of its two arguments:lambdaa,b:a+b.Lambda functions can be used wherever function objects are required. They aresyntactically restricted to a single expression. Semantically, they are justsyntactic sugar for a normal function definition. Like nested functiondefinitions, lambda functions can reference variables from the containingscope:

>>>defmake_incrementor(n):...returnlambdax:x+n...>>>f=make_incrementor(42)>>>f(0)42>>>f(1)43

The above example uses a lambda expression to return a function. Another useis to pass a small function as an argument:

>>>pairs=[(1,'one'),(2,'two'),(3,'three'),(4,'four')]>>>pairs.sort(key=lambdapair:pair[1])>>>pairs[(4, 'four'), (1, 'one'), (3, 'three'), (2, 'two')]

4.7.6.Documentation Strings

There are emerging conventions about the content and formatting of documentationstrings.

The first line should always be a short, concise summary of the object’spurpose. For brevity, it should not explicitly state the object’s name or type,since these are available by other means (except if the name happens to be averb describing a function’s operation). This line should begin with a capitalletter and end with a period.

If there are more lines in the documentation string, the second line should beblank, visually separating the summary from the rest of the description. Thefollowing lines should be one or more paragraphs describing the object’s callingconventions, its side effects, etc.

The Python parser does not strip indentation from multi-line string literals inPython, so tools that process documentation have to strip indentation ifdesired. This is done using the following convention. The first non-blank lineafter the first line of the string determines the amount of indentation forthe entire documentation string. (We can’t use the first line since it isgenerally adjacent to the string’s opening quotes so its indentation is notapparent in the string literal.) Whitespace “equivalent” to this indentation isthen stripped from the start of all lines of the string. Lines that areindented less should not occur, but if they occur all their leading whitespaceshould be stripped. Equivalence of whitespace should be tested after expansionof tabs (to 8 spaces, normally).

Here is an example of a multi-line docstring:

>>>defmy_function():..."""Do nothing, but document it.......    No, really, it doesn't do anything....    """...pass...>>>printmy_function.__doc__Do nothing, but document it.    No, really, it doesn't do anything.

4.8.Intermezzo: Coding Style

Now that you are about to write longer, more complex pieces of Python, it is agood time to talk aboutcoding style. Most languages can be written (or moreconcise,formatted) in different styles; some are more readable than others.Making it easy for others to read your code is always a good idea, and adoptinga nice coding style helps tremendously for that.

For Python,PEP 8 has emerged as the style guide that most projects adhere to;it promotes a very readable and eye-pleasing coding style. Every Pythondeveloper should read it at some point; here are the most important pointsextracted for you:

  • Use 4-space indentation, and no tabs.

    4 spaces are a good compromise between small indentation (allows greaternesting depth) and large indentation (easier to read). Tabs introduceconfusion, and are best left out.

  • Wrap lines so that they don’t exceed 79 characters.

    This helps users with small displays and makes it possible to have severalcode files side-by-side on larger displays.

  • Use blank lines to separate functions and classes, and larger blocks ofcode inside functions.

  • When possible, put comments on a line of their own.

  • Use docstrings.

  • Use spaces around operators and after commas, but not directly insidebracketing constructs:a=f(1,2)+g(3,4).

  • Name your classes and functions consistently; the convention is to useCamelCase for classes andlower_case_with_underscores for functionsand methods. Always useself as the name for the first method argument(seeA First Look at Classes for more on classes and methods).

  • Don’t use fancy encodings if your code is meant to be used in internationalenvironments. Plain ASCII works best in any case.

Footnotes

1

Actually,call by object reference would be a better description,since if a mutable object is passed, the caller will see any changes thecallee makes to it (items inserted into a list).