Feb 7
continue data structures
Array : sequential memory locations; fixed size when created
#
# Here's what one can look like in python.
#
data = [0]*100 # allocate 100 slots
data[3] = 10 # put something in slot 3
x = data[52] # fetch something from slot 52
Linked List :
#
# Last class we implemented a linked list with a python class,
# using an instance for each node. We can also use python's dict,
# using keywords for the internal data and link pieces.
#
# +--------+ +--------+
# | node0 | <----| node1 |
# +--------+ +--------+
#
node0 = {'data': 3, 'up': None}
node1 = {'data': 10, 'up': node0}
Binary (Linked) Tree :
# root 40
# / \
# / \
# kid0 kid1 60
#
kid0 = {'data': 20, left: None, right: None}
kid1 = {'data': 60, left: None, right: None}
root = {'data': 40, left: kid0, right: kid1}
# balanced trees : nodes equally spread between left & right
Stack :
s = Stack()
s.push(data0)
s.push(data1)
dat1 = s.pop() # last in, first out
Queue :
q = Queue()
q.enqueue(data0)
q.enqueue(data1)
data0 = q.dequeue() # first in, first out
Dictionary :
d = Dictionary()
d.insert(data, key)
data = d.search(key)
d.delete(key)
# "priority queue" or "heap" (coming in chapter 4)
data = d.max() # Might have this, if so "priority" thing
data = d.min() # and/or this
# list like
data = d.after(key) # Or might have this (e.g. linked list)
data = d.before(key) # and/or this (ditto).
Hash Table :
# e.g. python's dictionary structure
index = hash(data) # try to get a unique array offset
storage[index] = data # ... unless there is already data there
example hash table & hash function (pg 89)
storage[] = [0]*10000
H('cat') = 2*char('c') + 4*char('a') + 8*char('t')
index('cat') = H('cat') mod 10000
storage[index('cat')] = 'cat'
- The array gives us a "jump to that spot" behavior.
- The hash function gives us a (hopefully unique) deterministic offset for each key.
Depending on time, continue to play around
with coding some of these structures in python,
using python dicts or classes.
javascript singly linked list
Here's the same idea in javascript
// A singly linked list in javascript,
// using the same ideas we discussed in python.
var a0 = { data: 0, up: null };
var a1 = { data: 1, up: a0 };
var a3 = { data: 3, up: a1 };
var print_list = function(node){
while (node != null){ // while not at end,
console.log(" " + node.data); // print
node = node.up; // follow the chain
}
};
print_list(a3)
