Three Dollar Quill

Okay, I've got a bit more time. Let's see about improving MOTR.

After doing some planning for rewrites, I decided to take a look at coverage misses, because I'm currently blocking these rewrites on getting the coverage up.

One really prominent set of coverage misses focuses around the EnvVar class, which it looks like I'm currently not using at all. EnvVar is one of three wrapper classes that go around a Flex. These wrapper classes take the value in the Flex, and inject it into the generated command in various places. EnvVar is for injecting a Flex value into an environment variable. I don't know offhand of a program that works that way, but I think I shouldn't discount the possibility.

As far as testing this behavior, I could try to verify the actions of the functions directly, but I should be looking into verifying the design, and I'd like to try and verify the intended higher-level behavior, which is that you can't "double up" on a Flex's value by parameterizing fewer variables than the Flex contains.

Let's see what this looks like...

The fundamental range of behavior is between FlexIn and FlexOut, because LessFlexOut should mostly act like a FlexIn.

FlexIn: here is a specific range of values that something else generated.

FlexOut: here is a means of generating values.

The basic check here is that there is:

• Every command gets exactly one value for the EnvVar, and no EnvVar of a FlexOut produces the same value twice.

Thinking about this, I'd kind of rather have some kind of framework for testing this rather than trying to put stuff together ad-hoc. I'm going to have to spend some time thinking about that.

Anyway, it's late.

Good night.