Ruby Vulnerability Fixed In Ubuntu Hardy Heron

It appears that Ubuntu guys pushed the patch to fix the security vulnerability we’re all going crazy over. I just applied the patch this morning and below is the result of my test for Ruby array.

irb(main):001:0> a = []
=> []
irb(main):002:0> a[0x7fffffff] = "A"
IndexError: index 2147483647 too big
from (irb):2:in `[]='
from (irb):2
irb(main):003:0>

Rather than crashing, it just throws IndexError exception. After a short testing, it appears that nothing’s breaking. Very nicely done.

Off To RailsConf

I’m leaving very early tomorrow to Portland for RailsConf. It’s really odd not being a speaker at a conference and I intend to enjoy it, rather than working.

I plan on posting things I learn or anything interesting on this blog, so stay tuned.

Merb 0.9.3 Released

Ezra just released Merb 0.9.3. He has the details listed here.

I really like Merb. I believe it accommodates Flex/AIR and other front-end apps better since it’s lot leaner than the Rails. Oh, another thing, frameworks are not religions! They’re all toolsets. Just use the one that’s best for your needs. I strongly urge everyone to check out Merb.

no such file to load — libxml_so

“no such file to load — libxml_so” is the lovely error message I got when I upgraded/installed 0.5.3 version of libxml-ruby using gem. Now that I only use Leopard, I’m not sure if this applies to Tiger. My guess is that this is Leopard issue. The solution to this problem is relatively quick. I just removed 0.5.3 version and went back to older version.

sudo gem uninstall libxml-ruby -v 0.5.3

You’ll also have this problem on Ubuntu if you installed Ruby by compiling it yourself. It’s pain in the ass, but you can go through the code and install the components yourself. Let me know if anyone wants the steps.

By the way, libxml is about 100 times faster than REXML that comes with Ruby. If you don’t believe me, try it for yourself. I didn’t believe it either, but I was SHOCKED when I saw how much faster it was.

Linux + Firefox = #1 at RubyHead

While testing out Google Analytics and new Adobe AIR application, I found out that Firefox on Linux is now #1 OS and browser combo for those visiting this site. I’m extremely happy that Linux is gaining momentum. This means that people are breaking free from proprietary software and leaving the technology prison built by Microsoft.

It’s only been few weeks since I became Microsoft-free, and I’m somewhat shocked that I don’t notice anything different. In fact, I don’t even boot my VM anymore (not personally - I have to test webapps on Explorer at work). I became extremely comfortable with NeoOffice now that I got that stupid program to work, and haven’t found a need to open Microsoft Office.

Yeah! Going to RailsConf 2008!

My lovely wife is letting me take a mini-vacation by myself by letting me go to this year’s RailsConf. This is my first conference where I’m not speaking and don’t have to attend million meetings. It’s also not sponsored by anyone other than myself. Yes, I lost my mind.

If you want to see me there, just look for a dorky looking Korean dude with RubyHead.com shirt. I’ll set up a page where you can win this shirt.

I’m so happy that I work on Ruby and Rails. Don’t get me wrong, I was extremely passionate about what I did before as it allows me to do whatever I want now, but it’s just a pure joy at this point in my life. At some point, I will go back to being a CTO or CEO, but I need a break.

uninitialized constant Gem::GemRunner (NameError) - Ubuntu

I mentioned that I now use Ubuntu 7.10 on my Dell. After installing Ruby and RubyGems using apt-get, I got this nasty error AFTER doing ’sudo gem update –system’.

It appears that similar problems have occurred with Mac OS X as well. For one thing, you can try adding following to the gem file:

require ‘rubygems/gem_runner’

However, it was irritating me that along with this problem, I had to install irb separately and ruby sitting in /usr/bin directory.

My solution? Just install using the source. It always works for me without any problem and any subsequent stuff I add works without any issues. What I also like is that some of my scripts explicitly points to the location of ruby. Although I can correct this using env, why bother?

« Previous PageNext Page »