Travis CI: No output has been received in the last 10 minutes, this potentially indicates a stalled build or something wrong with the build itself.

Travis fails with
test_bitcoin: key.cpp:198: void ECC_Start(): Assertion secp256k1_context == __null' failed.
unknown location(0): fatal error in "script_PushData": signal: SIGABRT (application abort requested)`
Problem location: https://github.com/bitcoin/bitcoin/blob/master/src/key.cpp#L198

Looks like a libsecp256k issue (@sipa)?
Same travis problem in #6509.

This PR does the exact opposite of what the "block style example" says in https://github.com/bitcoin/bitcoin/blob/master/doc/developer-notes.md#coding

Too risky for what is gained, IMO. We don't usually merge these kinds of random code refactors because - even when they do improve readability - in the past some very nasty bugs have been introduced by them, accidentally.

That these changes are in consensus/validation code compounds the above-mentioned risks.

If your goal is to make the code more understandable, adding of comments is preferable.

Hi!
@laanwj I agree this is a small change in a very critical code. But if test pass and build pass. How does adding a new variable to store the return value should produce any bugs? (This kind of change is too simple to fail).
Thank you so much for reviewing

@pouta The tests hardly ever cover all possible cases. I agree that there is a class of changes that is too simple to fail, but just as well there are some seeming innocent changes that break bug-for-bug compatibility which is very important in the consensus code.

@jonasschnelli opened an issue for that, #6515. Looks like secp256k1 context is the victim, not the cause of the problem.