Author: blueZhift

I’ve been setting up a test installation for a ColdFusion application I’m supporting and dreading having to set up ColdFusion on Ubuntu. I had been planning to install ColdFusion 7 because the earlier 6.1 no longer worked after I upgraded Apache to version 2.2.3. I found instructions for installing ColdFusion 7 on Ubuntu, but in the comments on that blog someone had said that ColdFusion 8 installed on Ubuntu 7 without the need for any special steps.

So I downloaded the ColdFusion 8 Developer’s Edition for Linux and ran the installer. No trouble at all, and so far it appears to be working. Ubuntu is not an officially supported Linux for ColdFusion, but I suppose it has become popular enough that maybe someone decided to at least allow it to install on Ubuntu. In any case, thanks Adobe!

I just read that Windows Vista Service Pack 1 will be out Q1 2008 and weighs in at a hefty 1GB! OMFG, how in the world has it come to this? How do you grow an OS to such size? There’s no good reason that this should be the case and IMHO, Microsoft needs a serious palace revolution to get things back on track. In the past, faster computers have been able to outrun OS bloat so that now, the initially piggy Windows XP is a pretty good performer on today’s hardware. But I wonder if those days are coming to an end. Chips may get faster, but PCs have a lot more competition for their functionality now than in the past which could slow down PC sales to consumers. And, as usual, businesses are slow to adopt new OSs and probably won’t be picking up Vista until they buy new hardware, Vista’s specs have pretty much assured this. A 1GB service pack definitely is not going to help matters.

It’s been a few weeks since my last foray into the virtual world Second Life. I’ll probably logon this weekend having heard that the Second Life Community Conference is taking place right here in Chicago. Even if I wanted to go physically, I couldn’t because the event has been sold out for a while now! And part of me wonders if crossing the streams of one’s real life and second life, to borrow a Ghostbusters analogy, is a good thing. Insert the usual geek/nerd jokes here.

I only learned of the SLCC yesterday while reading an article on CNet concerning the conference and what they labeled as a “backlash” against Second Life due to some long standing technical issues and the failure of SL to live up to the wet dreams of a number of corporations hoping to make some real money in the virtual world.

After reading this, I remembered a lot of skepticism about whether or not anyone would be able to make money on the web and how companies like Amazon.com were destined to failure. Obviously, a lot, if not most, of the early web ventures failed, some spectacularly. But in the end, a number of companies were successful and now Amazon, Google, Yahoo, and others are household names and for many, it’s hard to imagine doing business without the web. Second Life may not go on to the success that Amazon ultimately had, but it’s really too early to tell. I think that virtual spaces like SL probably are the next big thing, but it will take some time to develop the new business paradigms that are valid in that space. If Linden Lab can hang on long enough and don’t make too many mistakes, they, like Amazon, may be around to reap the rewards of pioneering a new medium.

While very popular, MySQL is a database with a number of sometimes annoying quirks. On Windows, MySQL stores table names in lower case and does case insensitive comparisons by default. On Linux, it stores table names in their original cases and comparisons are case sensitive. This has obvious implications if one is trying to develop a database on Windows and then restore it to a Linux based production server, and is just the problem that I ran into. I know that I can reset those defaults to make this easier, but that really wasn’t an option on the production server, which hosts other databases that would have been effected by a change like this. So I decided that the best bet, being pressed for time, was to simply convert the lowercase table names in the mysqldump generated SQL dump, to the uppercase versions needed for the app running on the server. Having over 150 tables, in a 12MB dump file made changing these by hand extremely undesirable.

Regular expressions and Python to the rescue! Searching the web for a script that would find all text enclosed in back ticks and uppercase it, proved fruitless. And which my favorite editor for this, notepad++, can find and replace using regular expressions, it did not have any function I could invoke to perform the uppercasing automatically. So I present the following code I wrote to do the job. It’s very simple, only the very last part of how to grab the matched text that I wanted to uppercase, took any time. Please feel free to use this any way you like! I think it can serve as a good base for any similar search and replace operations. I used A.M. Kuchling’s Regular Expression HOWTO to learn how to use regex in python.

# uppercase_table_names.py
#
# utility script to find and uppercase all
# table names and references in mysql sql
# script. this actually uppercases anything
# in back ticks which is fine for my purposes
#
# feel free to use this script anyway you like
# -- blueZhift 2007-08-19
# http://bluezhift.proliphus.com

# see http://www.amk.ca/python/howto/regex/
# for more about using regular expressions in python
import re

def make_uppercase (match):
    input_string = str(match.group())
    return input_string.upper()

print "Beginning Program\n"

output_file=open("outputfile.sql", "w")
output_file.write("-- Uppercased Output\n")

# complile regular expression to look for
search_pattern = re.compile(r"`(.[^`]*)`", re.VERBOSE)

input_file = open("inputfile.sql")
for line in input_file.readlines():
    # find and replace
    edited_line = search_pattern.sub(make_uppercase,line)
    output_file.write(edited_line)
    print edited_line


input_file.close
output_file.close
print "Program Completed\n"

Click here to download source code plus input file example. (uppercase_table_names.zip)