Python 2.0 Quick Reference

NB: features added in 2.0 since 1.5.2 are colored dark magenta.

Based on:
    Python Bestiary, Author: Ken Manheimer, [email protected]
    Python manuals, Authors: Guido van Rossum and Fred Drake
    What's new in Python 2.0, Authors: A.M. Kuchling and Moshe Zadka
    python-mode.el, Author: Tim Peters, [email protected]

    and the readers of comp.lang.python

Python's nest: http://www.python.org     Developement: http://python.sourceforge.net/    ActivePython : http://www.ActiveState.com/ASPN/Python/
newsgroup: comp.lang.python  Help desk: [email protected]
Resources: http://starship.python.net/ and http://www.vex.net/parnassus/
Full documentation: http://www.python.org/doc/
An excellent Python reference book: Python Essential Reference by David Beazley (New Riders)

Contents

• Invocation Options
• Environment Variables
• Lexical Entities : keywords, identifiers, strings, numbers, sequences, dictionaries, operators
• Basic Types And Their Operations
• Advanced Types
• Statements
• Built In Functions
• Built In Exceptions
• Standard methods & operators redefinition in user-created Classes
• Special informative state attributes for some types
• Important Modules : sys, os, posix, posixpath, shutil, time, string, re, math, getopt
• List of modules In base distribution
• Workspace Exploration And Idiom Hints
• Python Mode for Emacs
• The Python Debugger

Invocation Options

• Available IDEs in std distrib: IDLE (tkinter based, portable), Pythonwin (Windows).

Environment variables

Notable lexical entities

Keywords

and       del       for       is        raise    
assert    elif      from      lambda    return   
break     else      global    not       try      
class     except    if        or        while    
continue  exec      import    pass               
def       finally   in        print

• (list of keywords in std module: keyword)
• Illegitimate Tokens (only valid in strings): @ $ ?
• A statement must all be on a single line. To break a statement over multiple lines use "\", as with the C preprocessor.

Exception: can always break when inside any (), [], or {} pair, or in triple-quoted strings.
• More than one statement can appear on a line if they are separated with semicolons (";").
• Comments start with "#" and continue to end of line.

Identifiers

• Python identifiers keywords, attributes, etc. are case-sensitive.
• Special forms: _ident (not imported by 'from module import *'); __ident__ (system defined name);

           __ident (class-private name mangling)

Strings

"a string enclosed by double quotes"
'another string delimited by single quotes and with a " inside'
'''a string containing embedded newlines and quote (') marks, can be delimited with triple quotes.'''
""" may also use 3- double quotes as delimiters """
u'a unicode string'   U"Another unicode string"
r'a raw string where \ are kept (literalized): handy for regular expressions and windows paths!'
R"another raw string"    -- raw strings cannot end with a \
ur'a unicode raw string'   UR"another raw unicode"

Use \ at end of line to continue a string on next line.

adjacent strings are concatened, e.g. 'Monty' ' Python' is the same as 'Monty Python'.

u'hello' + ' world'  --> u'hello world'   (coerced to unicode)

String Literal Escapes

     \newline  Ignored (escape newline)
     \\ Backslash (\)        \e Escape (ESC)        \v Vertical Tab (VT)
     \' Single quote (')     \f Formfeed (FF)       \OOO char with octal value OOO 
     \" Double quote (")     \n Linefeed (LF) 
     \a Bell (BEL)           \r Carriage Return (CR) \xHH  char with hex value HH
     \b Backspace (BS)       \t Horizontal Tab (TAB)
     \uHHHH  unicode char with hex value HHHH, can only be used in unicode string
     \UHHHHHHHH  unicode char with hex value HHHHHHHH, can only be used in unicode string
     \AnyOtherChar is left as-is

• NUL byte (\000) is NOT an end-of-string marker; NULs may be embedded in strings.
• Strings (and tuples) are immutable: they cannot be modified.

Numbers

Decimal integer: 1234, 1234567890546378940L        (or l)

Octal integer: 0177, 0177777777777777777L (begin with a 0)

Hex integer: 0xFF, 0XFFFFffffFFFFFFFFFFL (begin with 0x or 0X)

Long integer (unlimited precision): 1234567890123456L (ends with L or l)

Float (double precision): 3.14e-10, .001, 10., 1E3

Complex: 1J, 2+3J, 4+5j (ends with J or j, + separates (float) real and imaginary parts)

Sequences

• String of length 0, 1, 2 (see above)
'', '1', "12", 'hello\n'
• Tuple of length 0, 1, 2, etc:
() (1,) (1,2)     # parentheses are optional if len > 0
• List of length 0, 1, 2, etc:
[] [1] [1,2]

Sequence slicing [starting-at-index : but-less-than-index]. Start defaults to '0'; End defaults to 'sequence-length'.

a = (0,1,2,3,4,5,6,7)
a[3] ==> 3
a[-1] ==> 7
a[2:4] ==> (2, 3)
a[1:] ==> (1, 2, 3, 4, 5, 6, 7)
a[:3] ==> (0, 1, 2)
a[:] ==> (0,1,2,3,4,5,6,7)  # makes a copy of the sequence.

Dictionaries (Mappings)

Dictionary of length 0, 1, 2, etc:
{} {1 : 'first'} {1 : 'first',  'next': 'second'}

Operators and their evaluation order

Basic Types and Their Operations

Comparisons (defined between *any* types)

Boolean values and operators

None

None is used as default return value on functions. Built-in single object with type NoneType.
Input that evaluates to None does not print when running Python interactively.

Numeric types

Floats, integers and long integers.

Floats are implemented with C doubles.
Integers are implemented with C longs.
Long integers have unlimited size (only limit is system resources)

Operators on all numeric types

Bit operators on integers and long integers

Complex Numbers

• represented as a pair of machine-level double precision floating point numbers.
• The real and imaginary value of a complex number z can be retrieved through

the attributes z.real and z.imag.

Numeric exceptions

TypeError

raised on application of arithmetic operation to non-number

OverflowError

 numeric bounds exceeded

ZeroDivisionError

 raised when zero second argument of div or modulo op

Operations on all sequence types (lists, tuples, strings)

Notes :

Operations on mutable (=modifiable) sequences (lists)

Notes :
    (1) raise a ValueError exception when x is not found in s (i.e. out of range).
     (2) The sort() method takes an optional argument specifying a comparison fct of 2 arguments (list items) which should
          return -1, 0, or 1 depending on whether the 1st argument is considered smaller than, equal to, or larger than the 2nd
          argument. Note that this slows the sorting process down considerably.
     (3) The sort() and reverse() methods modify the list in place for economy of space when sorting or reversing a large list.
           They don't return the sorted or reversed list to remind you of this side effect.
     (4) [New 1.5.2] The pop() method is experimental and not supported by other mutable sequence types than lists.
          The optional  argument i defaults to -1, so that by default the last item is removed and returned.
     (5) [New 1.5.2] Experimental ! Raises an exception when x is not a list object.
 
 

Operations on mappings (dictionaries)

Notes :
  TypeError is raised if key is not acceptable
  (1) KeyError is raised if key k is not in the map
  (2) Keys and values are listed in random order
  (3) d2 must be of the same type as d1
  (4) Never raises an exception if k is not in the map, instead it returns defaultVal.
      defaultVal is optional, when not provided and k is not in the map, None is returned.
  (5) Never raises an exception if k is not in the map, instead it returns defaultVal, and adds k to map with value defaultVal. defaultVal is optional. When not provided and k is not in the map, None is returned and added to map.

Operations on strings

Notes :
    (1) Padding is done using spaces.
    (2) If optional argument start is supplied, substring s[start:] is processed. If optional arguments start and end are supplied, substring s[start:end] is processed.
    (3) Optional argument errors may be given to set a different error handling scheme. The default for errors is 'strict', meaning that encoding errors raise a ValueError. Other possible values are 'ignore' and 'replace'.
    (4) If optional argument tabsize is not given, a tab size of 8 characters is assumed.
    (5) Returns false if string s does not contain at least one character.
    (6) Returns false if string s does not contain at least one cased character.
    (7) A titlecased string is a string in which uppercase characters may only follow uncased characters and lowercase characters only cased ones.
    (8) s is returned if width is less than len(s).
    (9) If the optional argument maxsplit is given, only the first maxsplit occurrences are replaced.
    (10) If sep is not specified or None, any whitespace string is a separator. If maxsplit is given, at most maxsplit splits are done.
    (11) Line breaks are not included in the resulting list unless keepends is given and true.
    (12) table must be a string of length 256. All characters occurring in the optional argument deletechars are removed prior to translation.

String formatting with the % operator

• formatString uses C printf format codes : %, c, s, i, d, u, o, x, X, e, E, f, g, G, r (details below).
• Width and precision may be a * to specify that an integer argument gives the actual width or precision.
• The flag characters -, +, blank, # and 0 are understood. (details below)
• %s will convert any type argument to string (uses str() function)
• args may be a single arg or a tuple of args

        '%s has %03d quote types.' % ('Python', 2)  # => 'Python has 002 quote types.'

• Right-hand-side can also be a mapping:

        a = '%(lang)s has %(c)03d quote types.' % {'c':2, 'lang':'Python}

File Objects

Operators on file objects

File Exceptions

Advanced Types

• Module objects
• Class objects
• Class instance objects
• Type objects (see module: types)
• File objects (see above)
• Slice objects
• XRange objects
• Callable types:
• User-defined (written in Python):
• User-defined Function objects
• User-defined Method objects
• Built-in (written in C):
• Built-in Function objects
• Built-in Method objects
• Internal Types:
• Code objects (byte-compile executable Python code: bytecode)
• Frame objects (execution frames)
• Traceback objects (stack trace of an exception)

Statements

pass            -- Null statement
del name[,name]* -- Unbind name(s) from object. Object will be indirectly
                    (and automatically) deleted only if no longer referenced.
print [>> fileobject,] [s1 [, s2 ]* [,]
                -- Writes to sys.stdout, or to fileobject if supplied.
                   Puts spaces between arguments. Puts newline at end
                   unless statement ends with comma.
                   Print is not required when running interactively,
                   simply typing an expression will print its value,
                   unless the value is None.
exec x [in globals [,locals]]
                -- Executes x in namespaces provided. Defaults
                   to current namespaces. x can be a string, file
                   object or a function object.
callable(value,... [id=value], [*args], [**kw])
                -- Call function callable with parameters. Parameters can
                   be passed by name or be omitted if function 
                   defines default values. E.g. if callable is defined as
                   "def callable(p1=1, p2=2)"
                   "callable()"       <=>  "callable(1, 2)"
                   "callable(10)"     <=>  "callable(10, 2)"
                   "callable(p2=99)"  <=>  "callable(1, 99)"
                   *args is a tuple of positional arguments.
                   **kw is a dictionary of keyword arguments.

Assignment operators

Control Flow

if condition: suite
[elif condition: suite]*
[else: suite]   -- usual if/else_if/else statement
while condition: suite
[else: suite]
                -- usual while statement. "else" suite is executed
                   after loop exits, unless the loop is exited with
                   "break"
for element in sequence: suite
[else: suite]
                -- iterates over sequence, assigning each element to element.
                   Use built-in range function to iterate a number of times.
                   "else" suite executed at end unless loop exited
                   with "break"
break           -- immediately exits "for" or "while" loop
continue        -- immediately does next iteration of "for" or "while" loop
return [result] -- Exits from function (or method) and returns result (use a tuple to
                   return more than one value). If no result given, then returns None.

Exception Statements

assert expr[, message]
                -- expr is evaluated. if false, raises exception AssertionError
                   with message. Inhibited if __debug__ is 0.

try: suite1
[except [exception [, value]: suite2]+
[else: suite3]
                -- statements in suite1 are executed. If an exception occurs, look
                   in "except" clauses for matching <exception>. If matches or bare
                   "except" execute suite of that clause. If no exception happens
                   suite in "else" clause is executed after suite1.
                   If exception has a value, it is put in value.
                   exception can also be tuple of exceptions, e.g.
                   "except (KeyError, NameError), val: print val"
try: suite1
finally: suite2
                -- statements in suite1 are executed. If no
                   exception, execute suite2 (even if suite1 is
                   exited with a "return", "break" or "continue"
                   statement). If exception did occur, executes 
                   suite2 and then immediately reraises exception.
raise exception [,value [, traceback]]
                -- raises exception with optional value
                   value. Arg traceback specifies a traceback object to
                   use when printing the exception's backtrace.
raise           -- a raise statement without arguments re-raises
                   the last exception raised in the current function

              my_exception = 'You did something wrong'
      try:
                   if bad:
              raise my_exception, bad
      except my_exception, value:
                    print 'Oops', value
 

            class text_exception(Exception): pass
            try:
                if bad:
                    raise text_exception()
                    # This is a shorthand for the form
                    # "raise <class>, <instance>"
             except Exception:
                 print 'Oops'
                 # This will be printed because
                 # text_exception is a subclass of Exception
When an error message is printed for an unhandled exception which is a
class, the class name is printed, then a colon and a space, and
finally the instance converted to a string using the built-in function
str().

All built-in exception classes derives from StandardError, itself
derived from Exception.

Name Space Statements

Packages (>1.5): a package is a name space which maps to a directory including
                module(s) and the special initialization module '__init__.py'
                (possibly empty). Packages/dirs can be nested. You address a
                module's symbol via '[package.[package...]module.symbol's.

import module1 [as name1] [, module2]*
                -- imports modules. Members of module must be 
                   referred to by qualifying with [package.]module name:
                   "import sys; print sys.argv:"
                   "import package1.subpackage.module; package1.subpackage.module.foo()"
                   module1 renamed as name1, if supplied.
from module import name1 [as othername1] [, name2]*
                -- imports names from module module in current namespace.
                   "from sys import argv; print argv"
                   "from package1 import module; module.foo()"
                   "from package1.module import foo; foo()"
                   name1 renamed as othername1, if supplied.
from module import *
                -- imports all names in module, except those starting with "_";
                   *to be used sparsely, beware of name clashes* :
                   "from sys import *; print argv"
                   "from package.module import *; print x'
                    NB: "from package import *" only imports the symbols defined
                    in the package's __init__.py file, not those in the
                    template modules!
global name1 [, name2]*
                -- names are from global scope (usually meaning from module)
                   rather than local (usually meaning only in function).
                -- E.g. in fct without "global" statements, assuming
                   "a" is name that hasn't been used in fct or module
                   so far:
                   -Try to read from "a" -> NameError
                   -Try to write to "a" -> creates "a" local to fcn
                   -If "a" not defined in fct, but is in module, then
                       -Try to read from "a", gets value from module
                       -Try to write to "a", creates "a" local to fct
                   But note "a[0]=3" starts with search for "a",
                   will use to global "a" if no local "a".

Function Definition

def func_id ([param_list]): suite
                -- Creates a function object & binds it to name func_id.

param_list ::= [id [, id]*]
id ::= value | id = value | *id | **id

[Args are passed by value.Thus only args representing a mutable object
can be modified (are inout parameters). Use a tuple to return more than
one value]

Example:
        def test (p1, p2 = 1+1, *rest, **keywords):
            -- Parameters with "=" have default value (v is
               evaluated when function defined).
               If list has "*id" then id is assigned a tuple of
               all remaining args passed to function (like C vararg)
               If list has "**id" then id is assigned a dictionary of
               all extra arguments passed as keywords.

Class Definition

class <class_id> [(<super_class1> [,<super_class2>]*)]: <suite>
        -- Creates a class object and assigns it name <class_id>
           <suite> may contain local "defs" of class methods and
           assignments to class attributes.
Example:
       class my_class (class1, class_list[3]): ...
                  Creates a class object inheriting from both "class1" and whatever  
                  class object "class_list[3]" evaluates to. Assigns new
                  class object to name "my_class".
        - First arg to class methods is always instance object, called 'self'
          by convention.
        - Special method __init__() is called when instance is created.
        - Special method __del__() called when no more reference to object.
        - Create instance by "calling" class object, possibly with arg
          (thus instance=apply(aClassObject, args...) creates an instance!)
        - In current implementation, can't subclass off built-in
          classes. But can "wrap" them, see UserDict & UserList modules,
          and see __getattr__() below.
Example:
        class c (c_parent): 
           def __init__(self, name): self.name = name
           def print_name(self): print "I'm", self.name
           def call_parent(self): c_parent.print_name(self)
           instance = c('tom')
           print instance.name 
           'tom'
           instance.print_name()
           "I'm tom"
        Call parent's super class by accessing parent's method
        directly and passing "self" explicitly (see "call_parent"
        in example above).
        Many other special methods available for implementing
        arithmetic operators, sequence, mapping indexing, etc.

Documentation Strings

Example:
        class C:
            "A description of C"
            def __init__(self):
                "A description of the constructor"
                # etc.
Then c.__doc__ == "A description of C".
Then c.__init__.__doc__ == "A description of the constructor".

Others

lambda [param_list]: returnedExpr
                -- Creates an anonymous function. returnedExpr must be
                   an expression, not a statement (e.g., not "if xx:...", 
                   "print xxx", etc.) and thus can't contain newlines.
                   Used mostly for filter(), map(), reduce() functions, and GUI callbacks..

result = [expression for item1 in sequence1  [if condition1]
                        [for item2 in sequence2 ... for itemN in sequenceN]
                      ]

result = []
for item1 in sequence1:
    for item2 in sequence2:
    ...
        for itemN in sequenceN:
             if (condition1) and furthur conditions:
                  result.append(expression)

Built-In Functions

Built-In Exceptions

Exception>

         Root class for all exceptions

    SystemExit

         On 'sys.exit()'

    StandardError

                 Base class for all built-in exceptions; derived from Exception root class.

        ArithmeticError

                 Base class for OverflowError, ZeroDivisionError, FloatingPointError

            FloatingPointError

                       When a floating point operation fails.

            OverflowError

                        On excessively large arithmetic operation

            ZeroDivisionError

              On division or modulo operation with 0 as 2nd arg

        AssertionError

                When an assert statement fails.

        AttributeError

                On attribute reference or assignment failure

        EnvironmentError    [new in 1.5.2]

                On error outside Python; error arg tuple is (errno, errMsg...)

            IOError    [changed in 1.5.2]

           I/O-related operation failure

            OSError    [new in 1.5.2]

           used by the os module's os.error exception.

        EOFError

                Immediate end-of-file hit by input() or raw_input()

        ImportError

     On failure of `import' to find module or name

        KeyboardInterrupt

     On user entry of the interrupt key (often `Control-C')

        LookupError

                base class for IndexError, KeyError

            IndexError

             On out-of-range sequence subscript

            KeyError

             On reference to a non-existent mapping (dict) key

        MemoryError

     On recoverable memory exhaustion

        NameError

     On failure to find a local or global (unqualified) name

        RuntimeError

     Obsolete catch-all; define a suitable error instead

      NotImplementedError   [new in 1.5.2]

            On method not implemented

        SyntaxError

     On parser encountering a syntax error

   IndentationError

       On parser encountering an indentation syntax error

   TabError

       On parser encountering an indentation syntax error

        SystemError

     On non-fatal interpreter error - bug - report it

        TypeError

     On passing inappropriate type to built-in op or func

        ValueError

     On arg error not covered by TypeError or more precise

Standard methods & operators redefinition in classes

Standard methods & operators map to special '__methods__' and thus may be
 redefined (mostly in in user-defined classes), e.g.:
    class x: 
         def __init__(self, v): self.value = v
         def __add__(self, r): return self.value + r
    a = x(3) # sort of like calling x.__init__(a, 3)
    a + 4    # is equivalent to a.__add__(4)

Special methods for any class

        __init__(s, args) instance initialization (on construction) 
        __del__(s)        called on object demise (refcount becomes 0)
        __repr__(s)       repr() and `...` conversions
        __str__(s)        str() and 'print' statement
        __cmp__(s, o)     Compares s to o and returns <0, 0, or >0. 
                          Implements >, <, == etc...
        __hash__(s)       Compute a 32 bit hash code; hash() and dictionary ops
        __nonzero__(s)    Returns 0 or 1 for truth value testing
        __getattr__(s, name)  called when attr lookup doesn't find <name>
        __setattr__(s, name, val) called when setting an attr
                                  (inside, don't use "self.name = value"
                                   use "self.__dict__[name] = val")
        __delattr__(s, name)  called to delete attr <name>
        __call__(self, *args) called when an instance is called as function.

Operators

See list in the operator module. Operator function names are provided with 2 variants, with or without
ading & trailing '__' (eg. __add__ or add).

Numeric operations special methods
(s: self, o: other)

        s+o       =  __add__(s,o)         s-o        =  __sub__(s,o)
        s*o       =  __mul__(s,o)         s/o        =  __div__(s,o)
        s%o       =  __mod__(s,o)         divmod(s,o) = __divmod__(s,o)
        s**o      =  __pow__(s,o)
        s&o       =  __and__(s,o)         
        s^o       =  __xor__(s,o)         s|o        =  __or__(s,o)
        s<<o      =  __lshift__(s,o)      s>>o       =  __rshift__(s,o)
        nonzero(s) = __nonzero__(s) (used in boolean testing)
        -s        =  __neg__(s)           +s         =  __pos__(s)  
        abs(s)    =  __abs__(s)           ~s         =  __invert__(s)  (bitwise)
        s+=o      =  __iadd__(s,o)        s-=o       =  __isub__(s,o)
        s*=o      =  __imul__(s,o)        s/=o       =  __idiv__(s,o)
        s%=o      =  __imod__(s,o)
        s**=o     =  __ipow__(s,o)
        s&=o      =  __iand__(s,o)         
        s^=o      =  __ixor__(s,o)        s|=o       =  __ior__(s,o)
        s<<=o     =  __ilshift__(s,o)     s>>=o      =  __irshift__(s,o)

        Conversions
        int(s)    =  __int__(s)           long(s)    =  __long__(s)
        float(s)  =  __float__(s)         complex(s)    =  __complex__(s)
        oct(s)    =  __oct__(s)           hex(s)     =  __hex__(s)
        coerce(s,o) = __coerce__(s,o)
        Right-hand-side equivalents for all binary operators exist;
        are called when class instance is on r-h-s of operator:
        a + 3  calls __add__(a, 3)
        3 + a  calls __radd__(a, 3)

All seqs and maps, general operations plus:
(s: self, i: index or key)

        len(s)    = __len__(s)        length of object, >= 0.  Length 0 == false
        s[i]      = __getitem__(s,i)  Element at index/key i, origin 0

Sequences, general methods, plus:

  s[i]=v           = __setitem__(s,i,v)
  del s[i]         = __delitem__(s,i)
  s[i:j]           = __getslice__(s,i,j)
  s[i:j]=seq       = __setslice__(s,i,j,seq)
  del s[i:j]       = __delslice__(s,i,j)   == s[i:j] = []
  seq * n          = __repeat__(seq, n)
  s1 + s2          = __concat__(s1, s2)
  i in s           = __contains__(s, i)

Mappings, general methods, plus

  hash(s)          = __hash__(s) - hash value for dictionary references
  s[k]=v           = __setitem__(s,k,v)
  del s[k]         = __delitem__(s,k)

Special informative state attributes for some types:

    Lists & Dictionaries:
        __methods__ (list, R/O): list of method names of the object
 
    Modules:
        __doc__ (string/None, R/O): doc string (<=> __dict__['__doc__'])
        __name__(string, R/O): module name (also in __dict__['__name__'])
        __dict__ (dict, R/O): module's name space
        __file__(string/undefined, R/O): pathname of .pyc, .pyo or .pyd (undef for
                 modules statically linked to the interpreter)
        __path__(string/undefined, R/O): fully qualified package name when applies.
 
    Classes:    [in bold: writable since 1.5.2]
        __doc__ (string/None, R/W): doc string (<=> __dict__['__doc__'])
        __name__(string, R/W): class name (also in __dict__['__name__'])
        __bases__ (tuple, R/W): parent classes
        __dict__ (dict, R/W): attributes (class name space)
 
    Instances:
        __class__ (class, R/W): instance's class
        __dict__ (dict, R/W): attributes
    User-defined functions: [bold: writable since 1.5.2]
        __doc__ (string/None, R/W): doc string
        __name__(string, R/O): function name
        func_doc (R/W): same as __doc__
        func_name (R/O): same as __name__
        func_defaults (tuple/None, R/W): default args values if any
        func_code (code, R/W): code object representing the compiled function body
        func_globals (dict, R/O): ref to dictionary of func global variables
 
    User-defined Methods:
        __doc__ (string/None, R/O): doc string
        __name__(string, R/O): method name (same as im_func.__name__)
        im_class (class, R/O): class defining the method (may be a base class)
        im_self (instance/None, R/O): target instance object (None if unbound)
        im_func (function, R/O): function object
    Built-in Functions & methods:
        __doc__ (string/None, R/O): doc string
        __name__ (string, R/O): function name
        __self__ : [methods only] target object
        __members__ = list of attr names: ['__doc__','__name__','__self__'])
    Codes:
        co_name (string, R/O): function name
        co_argcount (int, R/0): number of positional args
        co_nlocals (int, R/O): number of local vars (including args)
        co_varnames (tuple, R/O): names of local vars (starting with args)
        co_code (string, R/O): sequence of bytecode instructions
        co_consts (tuple, R/O): litterals used by the bytecode, 1st one is
                                fct doc (or None)
        co_names (tuple, R/O): names used by the bytecode
        co_filename (string, R/O): filename from which the code was compiled
        co_firstlineno (int, R/O): first line number of the function
        co_lnotab (string, R/O): string encoding bytecode offsets to line numbers.
        co_stacksize (int, R/O): required stack size (including local vars)
        co_firstlineno (int, R/O): first line number of the function
        co_flags (int, R/O): flags for the interpreter
                           bit 2 set if fct uses "*arg" syntax
                           bit 3 set if fct uses '**keywords' syntax
    Frames:
        f_back (frame/None, R/O): previous stack frame (toward the caller)
        f_code (code, R/O): code object being executed in this frame
        f_locals (dict, R/O): local vars
        f_globals (dict, R/O): global vars
        f_builtins (dict, R/O): built-in (intrinsic) names
        f_restricted (int, R/O): flag indicating whether fct is executed in
                                 restricted mode
        f_lineno (int, R/O): current line number
        f_lasti (int, R/O): precise instruction (index into bytecode)
        f_trace (function/None, R/W): debug hook called at start of each source line
        f_exc_type (Type/None, R/W): Most recent exception type
        f_exc_value (any, R/W): Most recent exception value
        f_exc_traceback (traceback/None, R/W): Most recent exception traceback
    Tracebacks:
        tb_next (frame/None, R/O): next level in stack trace (toward the frame where
                                  the exception occurred)
        tb_frame (frame, R/O): execution frame of the current level
        tb_lineno (int, R/O): line number where the exception occured
        tb_lasti (int, R/O): precise instruction (index into bytecode)
 
    Slices:
        start (any/None, R/O): lowerbound
        stop (any/None, R/O): upperbound
        step (any/None, R/O): step value
 
    Complex numbers:
        real (float, R/O): real part
        imag (float, R/O): imaginary part
    XRanges:
        tolist (Built-in method, R/O): ?

Important Modules

Handles Unicode strings. Implemented in new module sre, re now a mere front-end for compatibility.
Patterns are specified as strings. Tip: Use raw strings (e.g. r'\w*') to litteralize backslashes.
 

Regular expression syntax

Form	Description
.	matches any character (including newline if DOTALL flag specified)
^	matches start of the string (of every line in MULTILINE mode)
$	matches end of the string (of every line in MULTILINE mode)
*	0 or more of preceding regular expression (as many as possible)
+	1 or more of preceding regular expression (as many as possible)
?	0 or 1 occurence of preceding regular expression
*?, +?, ??	Same as *, + and ? but matches as few characters as possible
{m,n}	matches from m to n repetitions of preceding RE
{m,n}?	idem, attempting to match as few repetitions as possible
[ ]	defines character set: e.g. '[a-zA-Z]' to match all letters(see also \w \S)
[^ ]	defines complemented character set: matches if char is NOT in set
\	escapes special chars '*?+&$|()' and introduces special sequences(see below). Due to Python string rules, write as '\\' orr'\' in the pattern string.
\\	matches a litteral '\'; due to Python string rules, write as '\\\\'in pattern string, or better using raw string: r'\\'.
|	specifies alternative: 'foo|bar' matches 'foo' or 'bar'
(...)	matches any RE inside (), and delimits a group.
(?:...)	idem but doesn't delimit a group.
(?=...)	matches if ... matches next, but doesn't consume any of the string e.g. 'Isaac (?=Asimov)' matches 'Isaac' only if followed by 'Asimov'.
(?!...)	matches if ... doesn't match next. Negative of (?=...)
(?P<name>...)	matches any RE inside (), and delimits a named group. (e.g. r'(?P<id>[a-zA-Z_]\w*)' defines a group named id)
(?P=name)	matches whatever text was matched by the earlier group named name.
(?#...)	A comment; ignored.
(?letter)	letter is one of 'i','L', 'm', 's', 'x'. Set the corresponding flags (re.I, re.L, re.M, re.S, re.X) for the entire RE.

Special sequences

Sequence	Description
number	matches content of the group of the same number; groups are numbered starting from 1
\A	matches only at the start of the string
\b	empty str at beg or end of word: '\bis\b' matches 'is', but not 'his'
\B	empty str NOT at beginning or end of word
\d	any decimal digit (<=> [0-9])
\D	any non-decimal digit char (<=> [^O-9])
\s	any whitespace char (<=> [ \t\n\r\f\v])
\S	any non-whitespace char (<=> [^ \t\n\r\f\v])
\w	any alphaNumeric char (depends on LOCALE flag)
\W	any non-alphaNumeric char (depends on LOCALE flag)
\Z	matches only at the end of the string

Variables

Variable	Meaning
error	Exception when pattern string isn't a valid regexp.

Functions

Function	Result
compile(pattern[, flags=0])	Compile a RE pattern string into a regular expression object. Flags (combinable by |): I or IGNORECASE or (?i) case insensitive matching L or LOCALE or (?L) make \w, \W, \b, \B dependent on thecurrent locale M or MULTILINE or (?m) matches every new line and not onlystart/end of the whole string S or DOTALL or (?s) '.' matches ALL chars, including newline X or VERBOSE or (?x) Ignores whitespace outside character sets
escape(string)	return (a copy of) string with all non-alphanumerics backslashed.
match(pattern, string[, flags])	if 0 or more chars at beginning of <string> match the RE pattern string,return a corresponding MatchObject instance, or None if no match.
search(pattern, string[, flags])	scan thru <string> for a location matching <pattern>, return acorresponding MatchObject instance, or None if no match.
split(pattern, string[, maxsplit=0])	split <string> by occurrences of <pattern>. If capturing () are used inpattern, then occurrences of patterns or subpatterns are also returned.
findall(pattern, string)	return a list of non-overlapping matches in <pattern>, either a list ofgroups or a list of tuples if the pattern has more than 1 group.
sub(pattern, repl, string[, count=0])	return string obtained by replacing the (<count> first) lefmost non-overlapping occurrences of <pattern> (a string or a RE object) in <string>by <repl>; <repl> can be a string or a fct called with a single MatchObj arg, which must return the replacement string.
subn(pattern, repl, string[, count=0])	same as sub(), but returns a tuple (newString, numberOfSubsMade)

Regular Expression Objects

Match Objects

Variables:

pi
e

acos(x)
asin(x)
atan(x)
atan2(x, y)
ceil(x)
cos(x)
cosh(x)
exp(x)
fabs(x)
floor(x)
fmod(x, y)
frexp(x)        -- Unlike C: (float, int) = frexp(float)
ldexp(x, y)
log(x)
log10(x)
modf(x)         -- Unlike C: (float, float) = modf(float)
pow(x, y)
sin(x)
sinh(x)
sqrt(x)
tan(x)
tanh(x)

Functions:

getopt(list, optstr)    -- Similar to C. <optstr> is option
                           letters to look for. Put ':' after letter
                           if option takes arg. E.g.
    # invocation was "python test.py -c hi -a arg1 arg2"
       opts, args =  getopt.getopt(sys.argv[1:], 'ab:c:')   
    # opts would be
       [('-c', 'hi'), ('-a', '')]
    # args would be
       ['arg1', 'arg2']

List of modules and packages in base distribution

(built-ins and content of python Lib directory)
(Python NT distribution, may be slightly different in other distributions)

(following list not revised)

* Built-ins *

            sys                 Interpreter state vars and functions
            __built-in__        Access to all built-in python identifiers
            __main__            Scope of the interpreters main program, script or stdin
            array               Obj efficiently representing arrays of basic values
            math                Math functions of C standard
            time                Time-related functions
            regex               Regular expression matching operations
            marshal             Read and write some python values in binary format
            struct              Convert between python values and C structs

* Standard *

            getopt              Parse cmd line args in sys.argv.  A la UNIX 'getopt'.
            os                  A more portable interface to OS dependent functionality
            re                  Functions useful for working with regular expressions
            string              Useful string and characters functions and exceptions
            whrandom            Wichmann-Hill pseudo-random number generator
            thread              Low-level primitives for working with process threads
            threading           idem, new recommanded interface.

* Unix/Posix *

            dbm                 Interface to Unix ndbm database library
            grp                 Interface to Unix group database
            posix               OS functionality standardized by C and POSIX standards
            posixpath           POSIX pathname functions
            pwd                 Access to the Unix password database
            select              Access to Unix select multiplex file synchronization
            socket              Access to BSD socket interface

* Tk User-interface Toolkit *

            tkinter             Main interface to Tk

* Multimedia *

            audioop             Useful operations on sound fragments
            imageop             Useful operations on images
            jpeg                Access to jpeg image compressor and decompressor
            rgbimg              Access SGI imglib image files

* Cryptographic Extensions *

            md5                 Interface to RSA's MD5 message digest algorithm
            mpz                 Interface to int part of GNU multiple precision library
            rotor               Implementation of a rotor-based encryption algorithm

* Stdwin * Standard Window System

            stdwin              Standard Window System interface
            stdwinevents        Stdwin event, command, and selection constants
            rect                Rectangle manipulation operations

* SGI IRIX * (4 & 5)

            al          SGI audio facilities
            AL          al constants
            fl          Interface to FORMS library
            FL          fl constants
            flp Functions for form designer
            fm          Access to font manager library
            gl          Access to graphics library
            GL          Constants for gl
            DEVICE      More constants for gl
            imgfile     Imglib image file interface

* Suns *

            sunaudiodev Access to sun audio interface

Workspace exploration and idiom hints

        dir(<module>)                                list functions, variables in <module>
        dir();                                       get object keys, defaults to local name space
        X.__methods__;                               list of methods supported by X (if any)
        X.__members__                                List of X's data attributes
        if __name__ == '__main__': main()            invoke main if running as script
        map(None, lst1, lst2, ...)                   merge lists
        b = a[:]                                     create copy of seq structure
        _                                            in interactive mode, is last value printed

Python Mode for Emacs

(Not revised, possibly not up to date)

Type C-c ? when in python-mode for extensive help.
INDENTATION
Primarily for entering new code:
        TAB      indent line appropriately
        LFD      insert newline, then indent
        DEL      reduce indentation, or delete single character
Primarily for reindenting existing code:
        C-c :    guess py-indent-offset from file content; change locally
        C-u C-c :        ditto, but change globally
        C-c TAB  reindent region to match its context
        C-c <    shift region left by py-indent-offset
        C-c >    shift region right by py-indent-offset
MARKING & MANIPULATING REGIONS OF CODE
C-c C-b         mark block of lines
M-C-h           mark smallest enclosing def
C-u M-C-h       mark smallest enclosing class
C-c #           comment out region of code
C-u C-c #       uncomment region of code
MOVING POINT
C-c C-p         move to statement preceding point
C-c C-n         move to statement following point
C-c C-u         move up to start of current block
M-C-a           move to start of def
C-u M-C-a       move to start of class
M-C-e           move to end of def
C-u M-C-e       move to end of class
EXECUTING PYTHON CODE
C-c C-c sends the entire buffer to the Python interpreter
C-c |   sends the current region
C-c !   starts a Python interpreter window; this will be used by
        subsequent C-c C-c or C-c | commands
VARIABLES
py-indent-offset        indentation increment
py-block-comment-prefix comment string used by py-comment-region
py-python-command       shell command to invoke Python interpreter
py-scroll-process-buffer        t means always scroll Python process buffer
py-temp-directory       directory used for temp files (if needed)
py-beep-if-tab-change   ring the bell if tab-width is changed

The Python Debugger

(Not revised, possibly not up to date, see 1.5.2 Library Ref section 9.1; in 1.5.2, you may also use debugger integrated in IDLE)

Accessing

import pdb      (it's a module written in Python)
        -- defines functions :
           run(statement[,globals[, locals]])
                        -- execute statement string under debugger control, with optional
                           global & local environment.
           runeval(expression[,globals[, locals]])
                        -- same as run, but evaluate expression and return value.
           runcall(function[, argument, ...])
                        -- run function object with given arg(s)
           pm()         -- run postmortem on last exception (like debugging a core file)
           post_mortem(t)
                        -- run postmortem on traceback object <t>
        
        -- defines class Pdb :
           use Pdb to create reusable debugger objects. Object
           preserves state (i.e. break points) between calls.
 
        runs until a breakpoint hit, exception, or end of program
        If exception, variable '__exception__' holds (exception,value).

Commands

h, help
        brief reminder of commands
b, break [<arg>]
        if <arg> numeric, break at line <arg> in current file
        if <arg> is function object, break on entry to fcn <arg>
        if no arg, list breakpoints
cl, clear [<arg>]
        if <arg> numeric, clear breakpoint at <arg> in current file
        if no arg, clear all breakpoints after confirmation
w, where
        print current call stack
u, up
        move up one stack frame (to top-level caller)
d, down
        move down one stack frame 
s, step
        advance one line in the program, stepping into calls
n, next
        advance one line, stepping over calls
r, return
        continue execution until current function returns
        (return value is saved in variable "__return__", which
        can be printed or manipulated from debugger)
c, continue
        continue until next breakpoint
a, args
        print args to current function
rv, retval
        prints return value from last function that returned
p, print <arg>
        prints value of <arg> in current stack frame
l, list [<first> [, <last>]]
               List source code for the current file.
               Without arguments, list 11 lines around the current line
               or continue the previous listing.
               With one argument, list 11 lines starting at that line.
               With two arguments, list the given range;
               if the second argument is less than the first, it is a count.
whatis <arg>
        prints type of <arg>
! 
        executes rest of line as a Python statement in the current stack frame
q quit
        immediately stop execution and leave debugger
<return>
        executes last command again
Any input debugger doesn't recognize as a command is assumed to be a
Python statement to execute in the current stack frame, the same way
the exclamation mark ("!") command does.

Example

(1394) python
Python 1.0.3 (Sep 26 1994)
Copyright 1991-1994 Stichting Mathematisch Centrum, Amsterdam
>>> import rm
>>> rm.run()
Traceback (innermost last):
         File "<stdin>", line 1
         File "./rm.py", line 7
           x = div(3)
         File "./rm.py", line 2
           return a / r
ZeroDivisionError: integer division or modulo
>>> import pdb
>>> pdb.pm()
> ./rm.py(2)div: return a / r
(Pdb) list
         1     def div(a):
         2  ->     return a / r
         3  
         4     def run():
         5         global r
         6         r = 0
         7         x = div(3)
         8         print x
[EOF]
(Pdb) print r
0
(Pdb) q
>>> pdb.runcall(rm.run)
etc.

Quirks

Always single-steps through top-most stack frame. That is, "c" acts like "n".