chore: remove unnecessary binding of current message

[hugoduncan]

The binding is not used within the body expressions, and costs a thread
local binding, producing garbage. This changes ensures print-stream
produces no memory garbage.

---

Before submitting a PR make sure the following things have been done:

• The commits are consistent with our contribution guidelines
• You've added tests to cover your change(s)
• All tests are passing
• The new code is not generating reflection warnings
• You've updated the README (if adding/changing middleware)

Note: If you're just starting out to hack on cider-nrepl you might find
nREPL's documentation and the
"Design" section of the README extremely useful.*

Thanks!

[vemv]

The binding is not used within the body expressions

Perhaps while this is true, the binding is used by the underlying nrepl machinery?

Reading the code, there seems to be intentful work for binding msg to a precise thing out of a msg-seq. i.e. if something bothered to do that, probably it was for a reason?

If we couldn't determine the original intent, perhaps I'd default to not changing the code - a binding's runtime cost surely is marginal in the context of printing stuff to *out* (i.e. IO-heavy workload).

Each printed line might also imply transient garbage, so in terms of memory, the binding's cost also seems marginal.

[bbatsov]

If the output is still properly redirected it's fine to remove this.

[bbatsov]

I did notice that this code has been there since the first commit of the file. I'm guessing one reason to have it is to avoid the original msg that triggered some output being replaced by a subsequent evaluation.

[hugoduncan]

My understanding is:

The forking-printer just takes values and forwards them. There is no execution of user code, so it can not possibly use the message in any.

print-stream invokes the forking printer flush every 100ms, so the current code produces garbage every 100ms.

[hugoduncan]

I think my previous patch introduced a bug - the outout to original-output should happen outside the loop created by with-out-binding. #737

[bbatsov]

Okay, you've convinced me. :-) Thanks!