Next step with analyzer: split into libraries

[glennj]

The shellcheck analysis should be 1 of a number of steps in the analysis.

Next up:

• move the shellcheck analysis out of bin/run.sh into a lib/ directory,
• add an exercise-specific analysis
  • two-fer is a good starter, check if submission uses $(var:-you} in some form
• determine what language will be appropriate for handling the various data structures: continue with ruby?
• is this the time to add a representer?

[IsaacG]

two-fer is a good starter, check if submission uses $(var:-you} in some form

That presupposes that this exercise is about learning parameter expansions vs conditional and breaking up the logic into reasonable blocks. I find (( $# )) && who=$1 || who=you; echo "One for $who, one for me." is a pretty solid solution, too, and teaches students a fair bit, too

[ErikSchierboom]

is this the time to add a representer?

I'd argue that the representer might be more important than the analyzer, so starting with the representer seems like a good idea! That said, we still have some work to do on the website to fully integrate representers, but it might be nice to start with the representer anyways.

That presupposes that this exercise is about learning parameter expansions vs conditional and breaking up the logic into reasonable blocks. I find (( $# )) && who=$1 || who=you; echo "One for $who, one for me." is a pretty solid solution, too, and teaches students a fair bit, too

These are exactly the types of discussions to engage in once there is a basic analyzer 👍