For future records, when creating OOXML files from scratch, make sure to use the ‘-D’ option to not store directory entries in the ZIP archive. This makes the Office apps choke (at least on OS X).
[Read more →]

This is part 1 of a 4 part series on EC2.

Part 1 – EC2 Background

What is EC2?

EC2, or ‘Elastic Compute Cloud’ is a primary component of AWS (‘Amazon Web Services’).  EC2 is Amazon’s market offering for cloud computing.  EC2 allows individuals and organizations the ability to run a full instance of their favorite version of Linux or Windows on Amazon’s gear – the same gear, ostensibly, that runs Amazon.com itself.  Users are given the opportunity to host their systems, applications, and files in Amazon’s data centers, on Amazon’s bandwidth and Internet connectivity, and run everything on Amazon’s server hardware (including redundant storage, power, etc.).
[Read more →]

I’m working on an extensive post about the market dynamics and pricing of Amazon’s AWS EC2 Spot Instance market.  Here’s a sneak peak:

This is a presentation I developed and gave at HRGeeks ][, on March 14. The bulk of this was original research.

Grab the PDF here.

iPhone forensics, without the iPhone

So, for the past few weeks, I’ve been hacking around with Python’s MySQLdb DBAPI v2 driver. After getting so used to mysqli in PHP, the move to MySQLdb was a mixed bag – some things are awesome, others not so much.

First the awesome – it’s so much cleaner to do safe, simple queries with python’s dbapi, than with the PHP mysqli stuff. Just call cursor.execute(query, arguments) and pass in your SQL and it’s arguments, and all the escaping and safety stuff is taken care of behind the scenes. No ugly hacks to check to see if something is a digit or a string and to properly escape/format the final query for you. In addition to that, connecting and setting the character set and forcing the driver to use unicode is all just 2 little options passed in the connect() method.

Now, the not so awesome – I really liked mysqli_fetch_assoc() for returning data from the database in a easy to use form – an associate array (in case it wasn’t obvious!). DBAPI drivers for python do not provide such a feature out of the box – your data comes back as a tuple, rather than a dict. While this makes some sense, the lack of a way to get a dict back seems to be rather un-pythonic. It violates the ‘Explicit is better than implicit’ and ‘Beautiful is better than ugly’ Zen of Python aphorisms. It requires implicit accessing of your data, as you must use tuple slices/subscripts instead of keywords to the dict, and that requires ugly numbers dropped into the otherwise easily readable code.

In an effort to save myself some trouble, here’s a quick way to emulate the fetch_assoc functionality, built in to a generator function so it can be used easily in ‘for’ statements and the like:

#!/usr/bin/env python
# Requires a version of python that supports generator functions (2.3+)
import  MySQLdb

connection = MySQLdb.connect(host="localhost", user="username", passwd="passwd", db="myInstance", use_unicode=True, charset="utf8")
cursor = connection.cursor()
def query(query, args):
        numRows = cursor.execute(query, args)
        tries = 0
        dbKeys = cursor.description
        row = {}
        while tries < numRows:
                tries = tries + 1
                dbRow = cursor.fetchone()
                for item, key in zip(dbRow, dbKeys):
                        row[key[0]] = item
                yield row

sql = """SELECT foo, bar FROM baz WHERE id = %(id)s"""
args = {'id':'adam'}

for rows in query(sql, args):
        print rows

That outputs:

	{'foo': u'fooValue', 'bar':'barValue'}
	{'foo': u'fooValue2', 'bar':'barValue2'}