Add it blocks

[locochris]

Don't know if anyone will think this sugar is sweet enough, but ...

Instead of writing:

it "assert equal $((1+1)) 2"
  assert equals $((1+1)) 2

write:

it <<-'EOF'
  assert equal $((1+1)) 2
EOF

[rylnd]

My feeling is there are few cases in which this would be useful, since it requires a single-line, self-explanatory assertion.

Besides the possible addition of an end_it command, I think we've probably hit a wall with how far we can push the current syntax.

I'd like to eventually replace the end_it and end_describe with something like the { } that's used in bats. It would require a preprocessor, but I think that could solve a lot of the issues we currently have open.

[twe4ked]

@rylnd I had a quick look at how the bats DSL works and that looks perfect.

This would potentially help solve #15, #30 and #31.

[locochris]

-1 for the bats preprocessor pattern, or at least be very careful it doesn't cause more problems than it solves.

I had problems with that exact thing in bats which led me to shpec in the first place.
I think it was when I was trying to put shpecs in a loop, and/or trying to output anything other than a shpec result.
(Although that might have been because of its use of the TAP.)

(My shpecs currently have verbose output that are summarized by a shpec result and can be in loops.)