I was almost through my first mag when I experienced a failure-to-fire. The round had fed properly and the slide appeared to be fully closed, but all I got when I squeezed the trigger on my Glock 23 was… nothing. No click. Just… nothing. I racked the slide again to eject the unfired round and I was able to finish the stage, but it cost me at least 2 seconds, perhaps more.

After retrieving my ejected round, I noticed it looked a bit odd. Comparing it to another round, I discovered the rear of the case hadn’t been trimmed properly. It was a few millimeters too long, as you can clearly see in the photo I’ve posted here. My guess is that it prevented the slide from fully engaging. This is the first time I’ve seen anything like this, and talking with a few other shooters at the match, this was the first time they had seen anything like this either. Cool!

I’ve sent the questionable round to Federal’s service department to see what they have to say. Hopefully they’ll make things right. The delay ended up costing me first place in the Novice class. What did I learn from this? First, that even quality factory ammo shouldn’t be relied upon to function flawlessly every time. Malfunctions will occur. Practice for them. Second, that every malfunction is different. I’ve practiced my failure-to-feed drill pretty consistently, but running into a situation where the slide closed and the trigger didn’t even click was something I just wasn’t expecting. Instead of treating it like any other failure, I let my surprise delay me from taking action.

Final note: I’ve been extremely happy with Federal ammo thus far and would be surprised if they don’t make this right for me. This post wasn’t intended to criticize their company in any way, it was just a fascinating occurrence that I thought might interest others as well. Has anyone else seen a factory round that was too long like this?

before_update do
  if started_at.to_s(:abbrev_date) != started_at_was.to_s(:abbrev_date)
    errors.add :started_at, "cannot be set to a different day"
  end
end

The started_at date was coming back in the Eastern time zone as expected. The started_at_was attribute, which was supposed to reflect the value prior to the update, was coming back as a UTC date/time. I would expect it to be returned in the Eastern time zone too. Apparently, I’m not alone in this assumption because a ticket was opened for this issue last month.

An official fix hasn’t been made yet, but I got around the problem by calling in_time_zone on my dirty attribute:

before_update do
  if started_at.to_s(:abbrev_date) != started_at_was.in_time_zone.to_s(:abbrev_date)
    errors.add :started_at, "cannot be set to a different day"
  end
end

That is all.

One nice benefit to upgrading was that I’m now able to leverage the Syntax Highlighter plugin. It does nifty stuff like this:

class HelloWorld
  def say_hello
    puts "Hello world!"
  end
end
HelloWorld.new.say_hello

If you run a WordPress blog, I would definitely recommend checking it into some sort of source control. Also, being able to run the blog on my development system is very beneficial. I was able to verify that the upgrade hadn’t borked my layout before making everything live. I use Apache to serve it up locally.

What platform do you use for blogging? What do you like about it?

This problem also occurs quite frequently for me when logging into a remote server. For example, a default Slicehost instance usually suffers from this behavior.

If you truly want to understand what causes the problem, check out this article.

To move to a new API version, I copy all of the existing integration tests and rename their prefix to the new version. Instead of renaming the files by hand, there is a nifty UNIX command that handles it for me. For example, to rename all the “v2_*.rb” files to “v3_*.rb” I would type:

for file in *; do mv "$file" "v3_${file#v2_}"; done

no such file to load -- active_support/cache/memory_store

The fix is simple. Assuming you’re using a vendored copy of Radiant or Rails, you probably have an entry for “cache” in your .gitignore file or your svn:ignore property. Remove it, commit the changes, and deploy again. Bingo!

Don’t ask me how I know this.

“The heavens declare the glory of God, and the sky above proclaims his handiwork. Day to day pours out speech, and night to night reveals knowledge.” Psalm 19:1-2

“For what can be known about God is plain to them, because God has shown it to them. For his invisible attributes, namely, his eternal power and divine nature, have been clearly perceived, ever since the creation of the world, in the things that have been made. So they are without excuse.” Romans 1:19-20