You can/should check these examples using the storage-visualization tool,


Lists are arrays that grow as needed. A list is written with this syntax: [val0, val1, ..., valn], e.g., [2, 5, "a", []], is a list of two ints, a string, and an empty list. (You can mix types within a Python list). Here are basic operations on lists:
x = [1,2,3]         # sets  x  to the list,  [1,2,3]
x.append(4)         # resets  x  to  [1,2,3,4]
y = x[1]            # sets y  to  2  (you index a list like an array)
z = [(y*y)+1, 4+2]  # sets  z  to  [5,6]
x = x + z           # sets  x  to  [1,2,3,4,5,6]

print x             # prints the list,  [1,2,3,4,5,6]
print len(x)        # prints  6,  the length of the list named by  x

a = x               # a is aliased to x  (a's cell holds x's handle)
a[0] = 9
print x             # prints the list,  [9,2,3,4,5,6]

b = x[:]            # sets  b  to a fresh copy of  x's elements (see "slicing", below)              

if isinstance(x, list):        # asks if  x's  value is a list
   for item in x :             # the for-loop iterates through  x's  elements 
      print item, item * item
                               # the loop prints   9 81
                               #                   2 4
                               #                    ...
                               #                   6 36
# this loop does the same work as the for-loop:
i = 0
while i != len(x) :
    item = x[i]
    print item, item * item
    i = i + 1

if y in x :    # asks if  y's  value is present within list  x
    print "yes"

# lists can be "sliced" (sublist-ed):
x = [2, 4, 6, 8, 10]
y = x[2:]       # "slice out" sublist  x[2], x[3], ..., to the end and assign to y
z = x[:3]       # "slice out" sublist up to, but not including, element 3
print x, y, z   #  prints  [2,4,6,8,10] [6,8,10] [2,4,6]
print x[1:4]    #  prints  [4,6,8],  namely, elements 1 up to, not including, 4

# how to emulate a stack with slicing, where the top is on the left:
st = []         # empty
st = [2] + st   # push 2
st = [4] + st   # push 4
print st[0]     # print top
st = st[1:]     # pop

# Using  append  and  pop  to emulate a stack, where the top is on the right:
st = []       # empty
st.append(4)  # st has value  [2,4]
print st[len(st)-1]   # prints top value, 4
print st[-1]   # also prints 4, since negative ints index from the right (!)
n = st.pop()   # pop  is a built-in method
print n, st    # prints  4 [2]

# Python has *list comprehensions*, which save a lot of coding!
vec = [2, 3, 4, 5, 6]

vec2 = [x*x  for x in vec]   # sets  vec2  to  [4,9,16,25,36]

vec3 = [ x*x  for x in vec  if x % 2 == 1 ]   # sets  vec3  to  [9,25]

# Here is cute trick that exposes that lists are really objects:
if x in x :
    print "is this possible?!"  # Actually, it is:   x = ["boo"]
                                #                    x.append(x)
                                # Now,  x in x  computes to True  (why?)


A dictionary (implemented in Python as a hash table) is similar to a record struct, but dictionaries can grow as needed and the key set is not fixed in advance.

A dictionary is written with this syntax: {key0: val0, key1: val1, ... keyn: valn}, e.g., {"p": True, "q": False, 99: {}} is a dictionary that has entries for the keys "p", "q", and 99. (You can mix keys and values of different type, but lists and dictionaries cannot themselves be keys.)

Here are basic operations on dictionaries:

d = {"p": True, "q": False, 99: {}}  # sets  d  to  {"p":True, "q":False, 99:{}}
x = d["q"]                           # sets  x  to  False
d["r"] = not x                       # adds new key:val pair to  d, which now is
                                     #  {"p":True, "q":False, 99:{}, "r":True}
d["q"] = 100                         # resets the value of the "q" key;  d  is
                                     #  {"p":True, "q":100, 99:{}, "r":True}

print d                         # prints the dict, but the keys might
                                # be in nonsorted order, since the dict
                                # is implemented as a hash table

if isinstance(d, dict):         # asks if  d's  value is a dictionary
    for key in d :              # like the for-loop for lists, enumerates each
        print key, d[key]       #  key  within  d  and uses it in the body

if "s" in d :                   # asks if  d  holds a key named  "s"
    print d["s"]

del d["q"]                      # deletes the entry for  "q"  from the dictionary.  Be careful.

# how to print a dictionary's contents in sorted order:
ids = d.keys()
for i in ids: 
    print i, ":", d[i]
You can build a list of dictionaries or a dictionary of lists, or you can mix together dictionaries, lists, ints, strings, etc., in a single data structure.

Python's ``global'' keyword

Python does not have variable declarations, but it does require that a function "declare" when it is changing the value of a variable that it does not own. (Such a variable is global to the function.) Here is a small example:
g = 4

def p(x) :
    print  x + g   // it is always ok to reference a global var

def q(y) :
    global g  // we are declaring that we want update rights to global var  g
    g = y     // we updated global var  g

def r(y) :
    g = 99    // BEWARE: the  g  assigned here is not declared "global" so it is a new local var
    print  y + g   

p(2)      // prints 6
print g  // prints 3
r(1)     //  prints 100
When a data structure is global, a function can change a component of the global structure without the global declaration! See how:
table = {'a': 4, 'b': 6}

def insert(key, value):
    table[key] = value    // the handle to  table  is not updated, but one of its components is....

def erase():
    global table;
    table = {}    // table's handle (value) is updated, so the  global  declaration is required

insert('b', 99)
print table   // prints  {'a': 4, 'b': 99}
print table        // prints  {}
If you write a script and you encounter an error where it seems that some variable "just isn't getting assigned when it is supposed to", then you are likely missing a global declaration!