In which Erin gets great news.

Q: How does a very large, very dead, very fly-ridden mouse end up on the 2nd floor deck?

A: I have no idea, but that shit was gross.

Q: How does one properly dispose of said corpse?

A: It’s a trade secret, but the items you’ll need include: a stick, an empty shoe box, and lots of duct tape.

Note to wife: DO NOT open the pink shoe box with duct tape on it!

In which Erin learns that not all is as it seems. Or at least, all is not as it was promised. I refrain from cussing, but it’s not that difficult to fill in the fucking blanks.

In my previous post “Why the Caps Need to Sign Huet“, I argued that Huet performed better than the Caps’ other goaltenders in the 2007-2008 NHL season, at least with respect to their goals-against averages, and save percent.

One day after Huet signed with Chicago, and the Caps signed Jose Theodore, I wondered how he stacked up against those 3 netminders from last year. The previous post’s calculations for all 3 guys included only games played as a Washington Capital. Since Theodore has yet to make his debut at Verizon Center, his GAA and SV% were calculated using all his games from this past year (regular season and playoffs) for the Avalanche.

Here’s the dirt. Reference that other post for the GAA and SV% algorithms.

I suppose a case could be made that by counting only Huet’s 20.3 games as a Capital artificially inflates his stats. But I’d point out that Johnson played only 17.2 games last year.

From a purely statistical standpoint, Theodore is better than both Kolzig and Johnson, but not as good as Huet by a long shot.

Their career stats tell a different story. That is–for all intents and purposes–they’re pretty much interchangeable. Here’s hoping that Theodore, standing in net behind the same guys as those other 3 Caps goalies, can match Huet’s success.

In which we learn something truly amazing–or truly fucked-up–about Erin’s placenta. Also, new images of The Beanlet, including his impersonations of 2 famous music stars… and his massive balls.

In which we discuss the not-so-common phenomenon called placenta previa.

Post #1,000 on this blog. Fitting that it’s Python nerd shit, huh?

I needed a way to search for MAC addresses, which are unique identifiers for networking hardware. For example, if your computer has a built-in Ethernet port, as well as wireless capability, then it has 2 MAC addresses. These are always 6 groups of 2 hexadecimal characters (0 through 9, and A through F). E.g., a valid MAC address would be: 01:98:DF:9E:10:37. Theoretically, every MAC address on every computer in the world will be unique, as the naming scheme provides over 281 trillion possible combinations (281,474,976,710,656).

Canonically these groups of 2 hex digits are separated by a colon, but many people record them with hyphens instead. So I needed to search for this particular pattern of characters amid a potentially-vast amount of text. Enter regular expressions (which I totally suck at using).

The regex I came up with is:

([a-fA-F0-9]{2}[:|\-]?){6}

Going through it, piece by piece:
[a-fA-F0-9] = find any character A-F, upper and lower case, as well as any number
[a-fA-F0-9]{2} = find that twice in a row
[a-fA-F0-9]{2}[:|\-] = followed by either a “:” or a “-” character (the backslash escapes the hyphen, since the hyphen itself is a valid metacharacter for that type of expression; this tells the regex to look for the hyphen character, and ignore its role as an operator in this piece of the expression)
[a-fA-F0-9]{2}[:|\-]? = make that final “:” or “-” character optional; since the last pair of characters won’t be followed by anything, and we want them to be included, too; that’s a chunk of 2 or 3 characters, so far
([a-fA-F0-9]{2}[:|\-]?){6} = find this type of chunk 6 times in a row

Let’s give it a shot.

First, a list of strings… e.g., a row from a comma-delimited file (returned via the csv module):

import re
L = ['aseredf', '55:A8:99:66:77:11', 'wefgcre', '98-75-64-52-48-21']
X = '([a-fA-F0-9]{2}[:|\-]?){6}' # this is the regex
for s in L:
    a = re.compile(X).search(s)
    if a:
        print s[a.start(): a.end()]

Run it:

[gary@roscoe ~]$ python test.py
55:A8:99:66:77:11
98-75-64-52-48-21

Next, a string:

import re
L = 'aseredf 55:A8:99:66:77:11 wefgcre 98-75-64-52-48-21'
X = '([a-fA-F0-9]{2}[:|\-]?){6}' # same regex as above
c = re.compile(X).finditer(L)
if c:
    for y in c:
        print L[y.start(): y.end()]

Run it:

[gary@roscoe ~]$ python test.py
55:A8:99:66:77:11
98-75-64-52-48-21

Fuckin’ bickety-bam, the whole stage comes crashing down.

As promised, I updated The Beanlet Blog.

The latest post includes an image from the 20-week sonogram.

I promised Erin that I would not use profanity on the new baby website. A completely separate blog seemed to be in order.

http://beanlet.wordpress.com/

Note the permanent link over in the right sidebar under “See Also”.

I’ve cross-posted 2 (edited) posts from here, and will be adding to that blog soon (hopefully tonight). Also, I’m hoping to get Erin in on the blogging over there… it must get tiresome reading only my view of the baby situation.

Me: Erin found some baby clothes that said something on the front in PHP.

Me: Like: <?php “baby();”> or something like that.

Me: So that made me think: “Python onesie”.

class Baby(Human):
   def __init__(self):
       return self

Gregg: Isn’t that a singleton?

Me: I don’t know what a “singleton” is… at least how it relates to programming.

Gregg: A class that only exists to return a single object.

Me: Then, yes, it would be.

Gregg: Technically, I guess it should test if other instances exist and either kill them or refuse to create.

Gregg: But it’s a fucking baby thing, there’s not enough space on the child’s front for real design patterns.

Gregg: Looking over those last few lines, I now know why I never get invited to parties.

Me: You wouldn’t want to imply that the baby kills other instances of Baby which subclass Human.

Me: Not yet, anyway.

Me: You could always subclass differently to imply it.

class Baby(Cthulu):
    RECURSION