Function Caching via Memoization

To explain memoization, see these different versions of Fibonacci:

Purely recursive. Highly inefficient. Recalculates every value for every calculation.

Iterative. Much faster.

Here is the memoization technique. A 'cache' dictionary of { (args) : (return_val), } is created. This allows us to quickly locate and reuse any value that has already been calculated.

Here is the same memoization design, implemented as a decorator.

Fabric is cool.

[Fabric](http://www.fabfile.org/) is cool. More python, less Bash. For example, I can automate my django db rebuild process, like so:

Using python pymssql to query SQLServer from linux

Getting pyssmql to work was pretty easy. (I had less luck using pyodbc.)

sudo apt-get install freetds

sudo pip install pymssql

Create entries for your sqlservers in /etc/freetds/freetds.conf like so:

[dev01]
host = YOUR_SQL_SERVER_HOST
port = 1433
tds version = 8.0


Here's a simple script using it:


import pymssql
import getpass

pwd = getpass.getpass()

conn = pymssql.connect(host='YOUR_SQL_SERVER_HOST',
                       user='YOUR_DOMAIN\\YOUR_USERNAME', password=pwd)
cur = conn.cursor()
cur.execute('use YOUR_DB')

query = "select top 10 id, name, lastmodifieddate from campaign"

cur.execute(query)

for row in cur:
    print "{0}\n{1}\n{2}\n".format(row[0], row[1], row[2])