Comments on: Coding: Single Level of Abstraction Principle

By: Border Crossings » Blog Archive » The Elements of Style

Border Crossings » Blog Archive » The Elements of Style — Thu, 16 Jun 2011 10:38:40 +0000

[...] (above) is closely related to the idea of cohesion in routines (Code Complete, p168), and to the “Single Level of Abstraction Principle” (SLAP). The first says that a routine should do one thing and one thing only. The second states that the [...]

By: Shaun Abram » Blog Archive » OSCON Day1: The Productive Programmer, part 2

Fri, 23 Jul 2010 03:12:57 +0000

[...] Single Layer of Abstraction Principle Keep all lines of code in a method at the same level of abstraction Jumping abstraction layers makes code hard to understand (See a blog posting on this here) [...]

By: • Will a single level of abstraction result in a higher complexity? | test.ical.ly

Wed, 26 May 2010 03:11:34 +0000

[...] bore you with explanations instead I would like those of you who haven’t heard about it to Mark Needhams blog post which also links to two much recommended books covering this and other [...]

By: Aligning the abstraction level with constant booleans « Schneide Blog

Aligning the abstraction level with constant booleans « Schneide Blog — Mon, 25 Jan 2010 21:18:31 +0000

[...] that is the principle to operate on only one level of abstraction within a method or function. See here for a detailed explanation. I have been applying this technique more or less unconsciously already [...]

By: Mark Needham

Mark Needham — Mon, 15 Jun 2009 12:00:45 +0000

@Jonas yeh just because all the other’s are var that one’s var. I tend to have all local variables as var these days. Rarely seem to miss explicitly typing on the left hand side of assignments.

By: Jonas Elfström

Jonas Elfström — Mon, 15 Jun 2009 08:46:23 +0000

I’m a fan of “var” as much as the next guy but what’s the reasoning behind “var hasExpectedPrefix” instead of “bool hasExpectedPrefix”? Since it only saves you from typing one single character I guess it’s some kind of code convention of yours?

By: Mark Needham

Mark Needham — Sun, 14 Jun 2009 22:09:38 +0000

@Dan yeh you can have it simplified down to exactly what you said and yeh it can’t be true – the example is slightly wrong actually.

It should be:

var followedBy7Digits = Regex.IsMatch(policyNumber.Substring(6,7), “^[0-9]{7}$”);

By: Dan

Dan — Sun, 14 Jun 2009 16:20:56 +0000

Perhaps this is because I’m from a Java background and don’t understand C# regex, but …

Assuming it’s like a POSIX regex, doesn’t the ^ and $ in followedBy7Digits mean “starts with 7 characters of [0-9] and then ends”. If so, how can hasExpectedPrefix && followedBy7Digits ever be true?

With a bit of refactoring I think you could just have one line:

return Regex.IsMatch(policyNumber, “^POLIC[0-9]{7}$”);

By: Mark Needham

Mark Needham — Sun, 14 Jun 2009 00:18:52 +0000

Hey Zach,

I actually came across this code in an acceptance test/end to end test so I guess then you have the debate of whether or not you should test your test code or not .

I think you’re absolutely right that it would have been much easier if it had been TDD’s in this case though.

My pair & I’s goal when we came across this code was to try and pull stuff out so that we could understand what was going on a bit more clearly and although it has rightly been pointed out that we don’t need all of those assertions it did help us at the time.

Mark

By: Zach Shaw

Zach Shaw — Sun, 14 Jun 2009 00:01:50 +0000

So… I’ve been wondering this since the first post on this topic, why wasn’t ValidPolicyNumber public / protected / what ever and just TDDed? Then you would have gotten around a lot of the tricky implementation details… like is policy.number == 12 needed.

-Zach